Practical Nitriding and Ferritic Nitrocarburizing [illustrated edition] 0871707918, 9780871707918

Citation preview

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

Practical

NITRIDING and Ferritic Nitrocarburizing

David Pye

ASM International Materials Park, Ohio 44073-0002 www.asminternational.org

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G) Copyright © 2003 by ASM International® All rights reserved

www.asminternational.org

No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the written permission of the copyright owner. First printing, December 2003

Great care is taken in the compilation and production of this book, but it should be made clear that NO WARRANTIES, EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE GIVEN IN CONNECTION WITH THIS PUBLICATION. Although this information is believed to be accurate by ASM, ASM cannot guarantee that favorable results will be obtained from the use of this publication alone. This publication is intended for use by persons having technical skill, at their sole discretion and risk. Since the conditions of product or material use are outside of ASM’s control, ASM assumes no liability or obligation in connection with any use of this information. No claim of any kind, whether as to products or information in this publication, and whether or not based on negligence, shall be greater in amount than the purchase price of this product or publication in respect of which damages are claimed. THE REMEDY HEREBY PROVIDED SHALL BE THE EXCLUSIVE AND SOLE REMEDY OF BUYER, AND IN NO EVENT SHALL EITHER PARTY BE LIABLE FOR SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES WHETHER OR NOT CAUSED BY OR RESULTING FROM THE NEGLIGENCE OF SUCH PARTY. As with any material, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specific testing under actual conditions is recommended. Nothing contained in this book shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters patent, copyright, or trademark, and nothing contained in this book shall be construed as a defense against any alleged infringement of letters patent, copyright, or trademark, or as a defense against liability for such infringement. Comments, criticisms, and suggestions are invited and should be forwarded to ASM International. Prepared under the direction of the ASM International Technical Book Committee (2002–2003), Charles A. Parker, Chair ASM International staff who worked on this project include Charles Moosbrugger, Acquisitions Editor; Bonnie Sanders, Manager of Production; Nancy Hrivnak, Jill Kinson, and Carol Polakowski, Production Editors; and Scott Henry, Assistant Director of Reference Publications. Library of Congress Cataloging-in-Publication Data Pye, David, 1939– Practical nitriding and ferritic nitrocarburizing / David Pye p. cm. Includes bibliographical references and index. 1. Nitriding. 2. Case hardening. 3. Steel—Heat treatment. I. Title. TN752.C3P4 2003 671.3'6—dc21 2003056298 ISBN: 0-87170-791-8 SAN: 204-7586 ASM International® Materials Park, OH 44073-0002 www.asminternational.org Printed in the United States of America

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

Contents Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix PART 1

Nitriding

CHAPTER 1 An Introduction to Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Metallurgical Considerations and Process Requirements . . . . . . . . . . . 1 The Pioneering Work of Machlet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Parallel Work in Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Developments in the United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Other Early Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Current Status of Nitriding Technology . . . . . . . . . . . . . . . . . . . . . . . . 11 CHAPTER 2 Why Nitride? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Key Process Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 CHAPTER 3 How Does the Nitriding Process Work? . . . . . . . . . . . . . . . . . . . . . 23 The Liberation of Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Dissociation of the Gas at the Selected Nitriding Temperature . . . . . 25 Why Ammonia Is Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Preheat Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 CHAPTER 4 Microstructures of Nitrided Iron and Steel . . . . . . . . . . . . . . . . . . 31 Influence of Carbon on the Compound Zone . . . . . . . . . . . . . . . . . . . 32 Controlling Compound Zone Thickness . . . . . . . . . . . . . . . . . . . . . . . 32 What Happens Below the Compound Zone? . . . . . . . . . . . . . . . . . . . . 35 iii

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

Can Plain Carbon Steel Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Calculating the Compound Zone Thickness . . . . . . . . . . . . . . . . . . . . 36 Other Factors Affecting Surface Case Formation . . . . . . . . . . . . . . . . 36 CHAPTER 5 Furnace Equipment and Control Systems . . . . . . . . . . . . . . . . . . . 39 Essential Furnace Design Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Types of Nitriding Furnaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Determining Appropriate Furnace Design . . . . . . . . . . . . . . . . . . . . . . 43 Retort Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Retort Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Sealing the Retort to Prevent Ammonia Leaks . . . . . . . . . . . . . . . . . . 44 Safety Precautions When Using Ammonia . . . . . . . . . . . . . . . . . . . . . 46 Furnace Heating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Process Control and Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . 49 CHAPTER 6 Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Salts Used and Process Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Types of Salt Bath Nitriding Processes . . . . . . . . . . . . . . . . . . . . . . . . 54 Salt Bath Nitriding Equipment and Procedure . . . . . . . . . . . . . . . . . . . 55 Using a New Salt Bath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Bath Replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Bath Testing and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Bath Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Operating the Salt Bath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Safety Precautions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Design Parameters for Furnace Equipment . . . . . . . . . . . . . . . . . . . . . 63 CHAPTER 7 Control of the Compound Zone or White Layer . . . . . . . . . . . . . . . 65 A Test to Determine the Presence of the White Layer . . . . . . . . . . . . . 66 Reduction of the Compound Zone by the Two-Stage Process . . . . . . 66 Other Methods for Controlling Compound Zone Formation . . . . . . . 67 Case Depth of Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 CHAPTER 8 Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 History of Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 How the Ion Nitriding Process Works . . . . . . . . . . . . . . . . . . . . . . . . . 72 Glow Discharge Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Process Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Other Uses for Plasma Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 iv

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

What Happens in the Ion Nitriding Process . . . . . . . . . . . . . . . . . . . . 77 Gas Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Reactions at the Steel Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 Surface Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 “Corner Effect” and Nitride Networking . . . . . . . . . . . . . . . . . . . . . . . 80 Degradation of Surface Finish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Control of the Compound Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Process Gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Process Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Plasma Generation Philosophies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Oxynitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 CHAPTER 9 Ion Nitriding Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Cold-Wall Continuous dc Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . 89 Hot-Wall Pulsed dc Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . 94 Work Cooling after Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . 101 Other Considerations for Ion Nitriding Equipment and Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Summary: Advantages of Plasma Nitriding . . . . . . . . . . . . . . . . . . . . 107 CHAPTER 10 Nitriding in Fluidized Beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Heating Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Nitriding in the Fluidized-Bed Furnace . . . . . . . . . . . . . . . . . . . . . . . 114 Oxynitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 Operating the Fluid Bed for Nitriding . . . . . . . . . . . . . . . . . . . . . . . . 117 Measurement of the Gas Dissociation . . . . . . . . . . . . . . . . . . . . . . . . 117 CHAPTER 11 Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Size Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Shape Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Control of Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Distortion in Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Stock Removal Prior to Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Postmachining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 CHAPTER 12 Steels For Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Steel Selection Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Requirements for a Nitriding Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 v

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

Can Stainless Steels Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Plasma Nitride Case Depths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 CHAPTER 13 Control of the Process Gas in Plasma Conditions . . . . . . . . . . . . 139 Analysis by Photo Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 Analysis by Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 Difficulties Associated with Gas Analysis . . . . . . . . . . . . . . . . . . . . . 141 Kinetic Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Appendix: The Role of Sputtering in Plasma Nitriding . . . . . . . . . . . . . 142 Experimental Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 CHAPTER 14 Processing with Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Hot-Work Tool Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 High-Speed Steel Cutters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Pure Irons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 Low-Alloy Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Maraging Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Higher Alloyed Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 CHAPTER 15 Stop-Off Procedures for Selective Nitriding . . . . . . . . . . . . . . . . 163 Methods for Selective Gas Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . 163 Methods for Selective Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . 164 Methods for Selective Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . 164 CHAPTER 16 Examination of the Nitrided Case . . . . . . . . . . . . . . . . . . . . . . . . 167 Hardness Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Etching of the Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Safety Precautions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Optical Light Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 CHAPTER 17 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Gas Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 vi

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

PART 2

www.asminternational.org

Ferritic Nitrocarburizing

CHAPTER 18 What Is Meant by Ferritic Nitrocarburizing? . . . . . . . . . . . . . . . . 193 Process Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Early History of Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . 195 Why Ferritic Nitrocarburize? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 CHAPTER 19 Salt Bath Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . 201 Low-Cyanide Salt Bath Ferritic Nitrocarburizing . . . . . . . . . . . . . . . 202 Salt Bath Nitrocarburizing plus Post Treatment . . . . . . . . . . . . . . . . 207 Kolene Nu-Tride Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 Other Methods for Salt Bath Nitrocarburizing . . . . . . . . . . . . . . . . . . 217 CHAPTER 20 Gaseous Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . 219 Development of the Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Process Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Gaseous Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Properties of Gaseous Ferritic Nitrocarburized Components . . . . . . 221 Industrial Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 Safety Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Appendix: Gaseous Nitrocarburizing—A Suitable Alternative for the Heat Treatment of Automotive Crankshafts . . . . . . . . . . . . 223 Process Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 Typical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 CHAPTER 21 Equipment for Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . 231 Salt Bath Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Atmosphere Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Plasma-Assisted Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . 233 Ferritic Oxynitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 CHAPTER 22 Preparation for Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . 241 Gas Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Enhanced Plasma Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 vii

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

CHAPTER 23 Evaluating the Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Case Depth Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Case Hardness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 What If the Formed Case Has Low Hardness Values? . . . . . . . . . . . 246 Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

viii

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

Preface Usually one seeks out a career, and, although I chose my career, I never realized how that choice was to impact my life. Although I chose heat treatment as a career, I did not make a conscious choice for nitriding; the subject of nitriding chose me. In 1960, I was a final-year apprentice at DeHavilland Propellors, Lostock, United Kingdom. My project with a colleague was to evaluate the nitriding process for the DeHavilland Aircraft Group. It was at that time that the subject of nitriding chose me. No matter where I have been, in the United Kingdom, South Africa, and now the United States, the subject of nitriding has followed me. Yet each time that I have researched the subject, I have found very few resource materials available. Unlike carburizing, for example, the subject of nitriding has had very few reference books or “cook books” written on the subject. A few books include a chapter or two on nitriding, and some conference papers are available in proceedings volumes. Up to now, however, there has not been a practical “how to” or “why to” book available on the subject. In 1991, Rodney Allwood of the ASM Education Department urged me to present a one-day class on nitriding and somehow got me to agree. Despite feeling that I did not know enough to pull it off, to my surprise I was able to put together the notes for that course. This was the foundation for a book on nitriding. Mrs. Veronica Flint of the ASM Reference Publications Department challenged me to write such a book and, with her tremendous patience and persistence, forced me to find the time to put pen to paper. Many books are dedicated to husbands, wives, children, or even dogs and cats (we have 8!). I can only dedicate this book to the young heat treater or metallurgist who is coming into the industry and to my colleagues who have, without exception, given me tremendous encouragement. I want to remember my colleagues in South Africa, where to a large extent I learned my trade. Without that 20 years experience in South Africa, I could not do what I am doing today. I also want to remember my ix

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

colleagues at DeHavilland Propellors with whom I worked from 1956 to 1963, especially Wally Simms, our foreman (now deceased), who had the foresight and tenacity to fight to bring apprentices into the world of heat treatment and the DeHavilland Propellers (now British Aerospace) apprentice program; Bill Oddey, our heat treatment superintendent; Joe Aspinall, our metallurgist; and Alex Thexton, our chemist (who sold me his briefcase, which I still have), who is now somewhere in Australia. We will always be grateful to Paul Huber at Seco Warwick, who brought my family and me to the United States, and I thank my other colleagues there. I would also like to dedicate this book to all those who have gone before me whose work in the field of nitriding has brought the process to maturity. I stand humbly before all of them and thank them for this opportunity. For help and growth in the field of pulsed plasma nitriding, I give thanks to Dr. Siegfeid Stramke in Germany. I especially would like to recognize Dr. Reinar Grun, who has put up with me, debated with me vigorously, and helped me learn the subject. He is a man whom I hold in great respect in the field of plasma physics. I would like to acknowledge both George Totten and Maurice Howes for being mentors to me and for their very positive encouragement. Their support has been a privilege and continues to be a very rewarding experience. I would like also to acknowledge Grace Pye. She always believed that I was capable of “going on my own.” It took my wife Lynn to push me to the successful completion of this and other goals. She said to me, “You have a dream. Don’t keep dreaming; live the dream.” I could not have accomplished what I have without her patience, dedication, and support. This book could not have come together without her. She typed the original manuscript from my almost illegible handwriting and “chicken scratching.” Thank you for your support of me and for listening to my heat treatment and furnace stories. You really are a gift to me. I also would like to recognize Valerie Sales, my mother-in-law (deceased) who interpreted my hand drawn sketches and turned them into illustrations, and Robin Maloney, who also helped with the illustrations. Thank you for being patient with me. I would like to thank all the reviewers who gave freely of their time and efforts to reviewing the manuscript. I would also like to acknowledge the editorial contribution of Joseph R. Davis, Davis & Associates, who helped polish the chapters and prepare them for the production process.

Purpose Nitriding and ferritic nitrocarburizing in some ways can be seen as Cinderella processes in comparison to the other surface modification techniques, as it was around but often ignored. Though the begining of nitriding can be traced to Adolph Machlet who applied for a patent in

x

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

1908 followed by Dr. Adolph Fry’s patent in 1922, it was considered in many areas as “a new process” when Drs. Wehnheldt and Berghaus developed the ion nitride process in 1932. Many fine metallurgists have contributed to the growth in our knowledge of the subject of nitriding. I have attempted to bring together all of their work by capturing the accumulated knowledge about the process in a book that will contribute to the continued growth of the nitriding process. The bell has not struck midnight on the subject of nitriding, however. Its technology, the refinement of its control, and most of all, our understanding of the process and methods will continue to grow. I expect the technology will be adapted for application to nonferrous metals, such as aluminum and the refractory metals. This book is intended to assist all of the practitioners of the technology in the day-to-day process operation of nitriding and ferritic nitrocarburizing. The contents are based upon my lifetime of experience and the knowledge gained from my peers. I hope that you, the reader, will gain some useful knowledge about the subject of nitriding and its derivative process techniques. The chapters in this book address many important questions related to the nitriding process: •

• • •

• • • •

Of the many nitriding methods, which one is for you? There are many different and valid reasons for choosing each relevant nitriding process technique, be it the reduction in thickness of the compound zone, the elimination of the compound zone, deep case formation, shallow case formation, high wear resistance, or corrosion resistance. Of the many different nitridable steels that can be chosen to manufacture the component in question, which one should be chosen? What hardness should the steel have prior to nitriding? How much surface stock should be removed prior to nitriding, and what problems are caused if the appropriate amount of stock is not removed? Which furnace should be used? How should the process be controlled? How should the steel be prepared? How should the steel be handled after nitriding?

Up to now, many of the answers have been determined mostly by personal experience (“what works for you”) and possibly by the age old method of trial and error. This book is a practical approach to the subject and not an academic or scientific work, although the subject is of a scientific nature. As with any other surface treatment process, such as carburizing or carbonitriding, the process of nitriding draws on many disciplines such as physics, chemistry, mechanical engineering, and electrical engineering. The “art” of nitriding

xi

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

also requires individual ingenuity and dogged determination, tempered with patience to accomplish the process and produce a metallurgically sound part. The process cannot be guessed. It can offer many metallurgical benefits, but it needs to be managed and controlled in order to produce the desired and acceptable surface metallurgy. While it is a very simple process, there is little widespread understanding of it. This book is my attempt to help remedy this situation. Over the years of heat treating and the years in the furnace industry, many examples of material selection for varied applications, process techniques, and failure evaluations have been both seen and experienced by myself and shared by others. My years as a consultant have exposed me to many industrial problems and process techniques. All of these experiences contributed significantly to this publication. There are many, many steels available to the engineer who designs the part and chooses the steel. The problem that the engineer is faced with is “How do I source my information on steel and metallurgical processing techniques?” The simple answer (which may sound glib) is “With great difficulty.” It is hoped that this publication will provide the reader, who might be a heat treater, metallurgist, or design engineer, with a clearer insight into the techniques, material selection, equipment, control, testing, evaluation, and trouble shooting. The analogy of an iceberg can be used to consider the factors that influence process costs related to nitriding. Nine-tenths of an iceberg is below the surface of the water. The top one-tenth could be likened to representing both the material cost and the process cost. However, inappropriate material selection and a lack of understanding of process techniques (and their results) can greatly inflate these costs. The submerged nine-tenths of this iceberg is the labor cost, machining costs, equipment costs, and time that is lost if the part does not function properly and must be scrapped because of improper heat treatment procedures and material selection. The combination of steel selection and choice of heat treatment can therefore “make or break the product.” This book is offered to help you “make the product” and not “break the product.” David Pye Pye Metallurgical Consulting Inc. Meadville, PA

xii

ASM International is the society for materials

engineers and scientists, a worldwide network dedicated to advancing industry, technology, and applications of metals and materials. ASM International, Materials Park, Ohio, USA www.asminternational.org This publication is copyright © ASM International®. All rights reserved. Publication title

Product code

Practical Nitriding and Ferritic Nitrocarburizing

#06950G

To order products from ASM International: Online Visit www.asminternational.org/bookstore Telephone 1-800-336-5152 (US) or 1-440-338-5151 (Outside US) Fax 1-440-338-4634 Mail

Customer Service, ASM International 9639 Kinsman Rd, Materials Park, Ohio 44073-0002, USA

Email [email protected]

American Technical Publishers Ltd. 27-29 Knowl Piece, Wilbury Way, Hitchin Hertfordshire SG4 0SX, In Europe United Kingdom Telephone: 01462 437933 (account holders), 01462 431525 (credit card)

www.ameritech.co.uk Neutrino Inc. In Japan Takahashi Bldg., 44-3 Fuda 1-chome, Chofu-Shi, Tokyo 182 Japan Telephone: 81 (0) 424 84 5550 Terms of Use. This publication is being made available in PDF format as a benefit to members and customers of ASM International. You may download and print a copy of this publication for your personal use only. Other use and distribution is prohibited without the express written permission of ASM International. No warranties, express or implied, including, without limitation, warranties of merchantability or fitness for a particular purpose, are given in connection with this publication. Although this information is believed to be accurate by ASM, ASM cannot guarantee that favorable results will be obtained from the use of this publication alone. This publication is intended for use by persons having technical skill, at their sole discretion and risk. Since the conditions of product or material use are outside of ASM's control, ASM assumes no liability or obligation in connection with any use of this information. As with any material, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specific testing under actual conditions is recommended. Nothing contained in this publication shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters patent, copyright, or trademark, and nothing contained in this publication shall be construed as a defense against any alleged infringement of letters patent, copyright, or trademark, or as a defense against liability for such infringement.

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p1-12 DOI: 10.1361/pnafn2003p001

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

1 An Introduction to Nitriding

THE NITRIDING PROCESS, first developed in the early 1900s, continues to play an important role in many industrial applications. Along with the derivative nitrocarburizing process, nitriding often is used in the manufacture of aircraft, bearings, automotive components, textile machinery, and turbine generation systems. Though wrapped in a bit of “alchemical mystery,” it remains the simplest of the case hardening techniques. The secret of the nitriding process is that it does not require a phase change from ferrite to austenite, nor does it require a further change from austenite to martensite. In other words, the steel remains in the ferrite phase (or cementite, depending on alloy composition) during the complete procedure. This means that the molecular structure of the ferrite (body-centered cubic, or bcc, lattice) does not change its configuration or grow into the face-centered cubic (fcc) lattice characteristic of austenite, as occurs in more conventional methods such as carburizing. Furthermore, because only free cooling takes place, rather than rapid cooling or quenching, no subsequent transformation from austenite to martensite occurs. Again, there is no molecular size change and, more importantly, no dimensional change, only slight growth due to the volumetric change of the steel surface caused by the nitrogen diffusion. What can (and does) produce distortion are the induced surface stresses being released by the heat of the process, causing movement in the form of twisting and bending.

Metallurgical Considerations and Process Requirements Nitriding is a ferritic thermochemical method of diffusing nascent nitrogen into the surface of steels and cast irons. This diffusion process is based on the solubility of nitrogen in iron, as shown in the iron-nitrogen equilibrium diagram (Fig. 1).

2 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Iron-nitrogen equilibrium diagram. The δ-phase, not shown on this diagram, exists from 11.0 to 11.35% N at temperatures below approximately 500 °C (930 °F).

The solubility limit of nitrogen in iron is temperature dependent, and at 450 °C (840 °F) the iron-base alloy will absorb up to 5.7 to 6.1% of N. Beyond this, the surface phase formation on alloy steels tends to be predominantly epsilon (ε) phase. This is strongly influenced by the carbon content of the steel; the greater the carbon content, the more potential for the ε phase to form. As the temperature is further increased to the gamma prime (γ′) phase temperature at 490 °C (914 °F), the “window” or limit of solubility begins to decrease at a temperature of approximately 680 °C (1256 °F). The equilibrium diagram shows that control of the nitrogen diffusion is critical to process success (Fig. 1). A number of operating process parameters must be adhered to and controlled in order to successfully carry out the nitriding process. Most of these parameters can be controlled with relatively simple instrumentation and methods. Examples of process parameters for gas nitriding include: • • • • • •

Furnace temperature Process control (see discussion below) Time Gas flow Gas activity control Process chamber maintenance

All these factors help to reduce distortion during the process, with the exception of induced residual stresses. Another benefit of nitriding is that it acts as a stabilizing process by providing an additional temper to the processed steel.

Chapter 1: An Introduction to Nitriding

Control of the process parameters is necessary to ensure formation of an acceptable metallurgical case. Without control, repeatability of the metallurgical requirements cannot be guaranteed. The process control factors are those elements that will ensure a controlled process and acceptable results: • • • • • • •

Total surface area to be nitrided Process pressure inside the sealed process chamber Gas delivery pressure system into the sealed process chamber Exhaust gas system from the sealed process chamber Control of the preheat treatment procedure prior to nitriding, including stress relief and prehardening and tempering Quality and integrity of the steel surface precleaning prior to nitriding Consistent steel chemistry to maximize “nitridability”

The Pioneering Work of Machlet In the early years of the 20th century, Adolph Machlet worked as a metallurgical engineer for the American Gas Company in Elizabeth, NJ. He recognized that the surface hardening technique of carburizing led to distortion problems due to extended periods at elevated temperatures, followed by severe quenching into either water or oil. Through experimentation, Machlet soon discovered that nitrogen was very soluble in iron. Nitrogen diffusion produced a relatively hard surface in simple plain irons or low-alloy steels and significantly improved corrosion resistance. This was accomplished without subjecting the steel to elevated temperatures and, more importantly, without cooling the steel rapidly to achieve a hard wearing surface. It could now cool freely within the process chamber, while still under the protection of the nitrogen-based atmosphere, thus reducing the risk of distortion yet still producing a hard, wear-resistant surface with good corrosion resistance. Ammonia was decomposed, or “cracked,” by heat to liberate the nascent nitrogen necessary for the process. It was not long before Machlet realized that he needed to control the decomposition accurately. He did this by using hydrogen as a dilutant gas to reduce the amount of available nascent nitrogen, thus controlling to some extent the formed case metallurgy. His reasoning behind the control of the process gas was recognition of what is now known as the “white layer” or “compound zone.” Figure 2 shows a simple construction of the nitrided case. It should be noted that this schematic is not to scale. The first patent for the development of the nitriding process was applied for in March 1908 in Elizabeth, NJ. The patent was finally approved in June 1913, some five years after the initial application. Machlet had been working for a number of years on the process prior to his patent application

/ 3

4 / Practical Nitriding and Ferritic Nitrocarburizing

Compound zone, dual phase Diffusion zone consisting of formed nitrides Transition zone from diffusion zone to core material

Core material

Fig. 2

Schematic of a typical nitrided case structure

and continued to develop both the new process and his understanding of the resulting process metallurgy. The patent was for “The Nitrogenization of Iron and Steel in an Ammonia Gas Atmosphere into which an Excess of Hydrogen Has Been Introduced” (Ref 1). Although Machlet’s development and patenting of the new nitriding procedure was technologically important, his work remained largely unrecognized and faded into obscurity. Even today, very few nitriding practitioners know who he was and what he accomplished. Most metallurgists who are familiar with the nitriding process know the work of the German researcher Adolph Fry, who is recognized as the “father of nitriding.” While Fry’s work was more publicized and his methods were taught at many fine metallurgical academic institutions, it was Machlet who first pioneered the nitriding process.

Parallel Work in Europe Adolph Fry. In Germany, a parallel research program was under way at the Krupp Steel Works in Essen. This program was headed by Dr. Adolph Fry in 1906. Like Machlet, Fry recognized that nitrogen was very soluble in iron at an elevated temperature. He also recognized very early in his work that alloying elements strongly influenced metallurgical and performance results. Fry first applied for his patent in 1921, three years after the First World War ended. His patent was granted in March 1924 (Ref 2). He used a technique similar to that of Machlet, where the nitrogen source had to be cracked by heat to liberate nitrogen for reaction and diffusion. Like Machlet, Fry used ammonia as the source gas, but he did not use hydrogen as a dilutant gas. Thus was developed the single-stage gas nitriding process as it is known today. Fry then investigated the effects of alloying elements on surface hardness. He discovered that the nitriding process produced a high surface hardness only on steels containing chromium, molybdenum, aluminum,

Chapter 1: An Introduction to Nitriding

/ 5

vanadium, and tungsten, all of which form what are known as “stable nitrides.” He also discovered the critical nature of process temperature in terms of case depth and surface metallurgy. Processing the steel at higher temperatures placed the surface at risk to form what is known today as “nitride networks” (a saturated solution of nitrogen in the immediate surface of the formed case). Because steels with higher alloy contents were not readily available for nitriding, Fry became responsible for developing a group of steels for Krupp known as the “Nitralloy” group. These steels, specifically designed as nitriding steels, soon became internationally recognized. Even today the Nitralloy steels are specified. British Standard Nitriding Steels. Shortly thereafter in the late 1920s, a company in Sheffield, England, also began work on developing a group of nitriding steels under the licensed guidance of Krupp Steels. These steels were also marketed under the brand name of Nitralloy. The company was Thomas Firth and John Brown Steelworks, more commonly known as Firth Brown Steels. The steels from Firth Brown were known as the “LK” group, designated by British Standard 970 as En 40 A, En 40 B, En 40 C, En 41 A, and En 41 B. Developed for nitriding applications, these were chromium-molybdenum steels (see Table 1 for chemical compositions). The En 41 series contained aluminum, which produced a much higher surface hardness after nitriding. Aluminum has a strong affinity for nitrogen, forming very hard aluminum nitrides that are quite stable in amounts up to 1.0% Al. Much above 1.0%, aluminum has no effect on the resultant nitriding hardness. Differences Between the U.S. and German Processes. The principal differences between the process developed in the United States and that developed in Germany were that: •

•

The U.S. process used hydrogen as a dilutant gas to control the nitriding potential of both the gas and steel, which in turn controlled the final surface metallurgy. The Germans manipulated the process through alloying and improved on such aspects as core hardness and tensile strength.

Table 1

British standard nitriding steels Composition, %

Designation(a)

En 40 A En 40 B En 40 C En 41 A En 41 B

C

Si

Mn

P

Cr

Mo

Ni

V

Al

0.20–0.35 0.20–0.30 0.30–0.50 0.25–0.35 0.25–0.45

0.10–0.3 0.10–0.35 0.10–0.35 0.10–0.35 0.10–0.35

0.40–0.55 0.40–0.65 0.40–0.80 0.65 max 0.65 max

0.05 max 0.05 max 0.05 max 0.05 max 0.05 max

2.90–4.00 2.90–3.50 2.90–3.50 1.40–1.80 1.40–1.80

0.60–0.80 0.40–0.70 0.70–1.20 0.10–0.25 0.10–0.25

0.40 max 0.40 max 0.40 max 0.40 max 0.40 max

... 0.10–0.30 0.10–0.30 ... ...

... ... ... 0.90–1.30 0.90–1.30

(a) The international designation for En 40 A, B, and C is 31 CrMoV 9. En 41 A and B are designated 34 CrAlMo 5. max, maximum

6 / Practical Nitriding and Ferritic Nitrocarburizing

Machlet’s process was not widely accepted in the United States, as it was perceived to have little if any commercial value to U.S. industry. By contrast, the Germans exploited Fry’s process in the early years following WWI. The German process enjoyed great success throughout Europe in the aircraft, textile, railroad, automotive, and machine tool industries. During the mid to late 1920s, information about Fry’s success began to filter through to American industrialists, prompting the Society of Manufacturing Engineers (SME) to take a strong interest in the German developments. This led to SME sending Dr. Zay Jeffries from Cleveland, OH, to Germany to visit Krupp Steel and Dr. Fry in 1926. It was at this meeting that Jeffries suggested to Fry that he attend the forthcoming annual SME conference in Chicago and present a paper on the process techniques and applications. Fry could not attend, so his friend and colleague Pierre Aubert made a presentation on his behalf. The presentation helped bring about commercialization of the process in the United States.

Developments in the United States Following the presentation of Fry’s work at the 1927 SME conference, American metallurgists began exploring nitriding processing parameters and the effects of alloying on the nitriding response of steels. Some of the more notable studies are described later in this chapter. McQuaid and Ketcham. Metallurgists H.W. McQuaid and W.J. Ketcham at the Timken Detroit Axle Company in Detroit, MI conducted a series of investigations to evaluate the new nitriding process. The studies were completed during a two-year period, which concluded with a presentation of their findings in 1928 (Ref 3). In general, the investigatory work focused on process temperature. The temperatures selected ranged from 540 to 650 °C (1000 to 1200 °F). The upper temperature was significantly lower than the temperatures employed by Machlet, which ranged from 480 to 980 °C (900 to 1800 °F). McQuaid and Ketcham concluded that higher nitriding temperatures had an effect on core hardness of alloy steels but little effect on the ability to nitride at those temperatures. They also found that higher process temperatures increased the risk of forming nitride networks, particularly at corners, due to the higher solubility of nitrogen in iron. When present, nitride networks cause premature failure at the steel surface by cracking and exfoliation. McQuaid and Ketcham began an exhaustive series of investigations into the new process of nitriding as developed by Machlet and Fry. The studies included: • • • •

Influence of temperature on both case formation and case depth Influence of alloying elements in the newly developed Nitralloy steels Influence of temperature on growth and distortion Influence of time on case depth distortion and growth

Chapter 1: An Introduction to Nitriding

• •

Effects of the ammonia/hydrogen relationship and dilution by hydrogen Effects of slow and rapid cooling, such as controlled cooling in the process retort by the introduction of air and rapid cooling in water

They concluded that nitriding was much easier to control than carburizing. They also found that the corrosion properties of low-alloy and alloy steels were much improved while undergoing salt spray tests and that practically any steel can be nitrided, including plain carbon steel and pure iron. McQuaid and Ketcham were also the first early metallurgists to study the white layer or compound zone. They concluded that the “white structure” is composed of a nitride, either iron nitrides or a complex nitride layer, involving both iron and alloying elements. A further conclusion was that the white layer or compound zone was extremely hard but very brittle and that the layer should be avoided if possible (though no specific guidelines were offered). They also studied the effect of decarburization on nitrogen diffusion and the mechanical strength of the nitrided case. Their results showed that the steel to be nitrided should clearly be free of surface decarburization; otherwise, the nitrided surface will exfoliate and peel away from the substrate. They concluded that rough machining or some other operation to ensure complete removal of any decarburized surface layer should be performed before carrying out any nitride operation. Robert Sergeson was associated with the research laboratories of the Central Alloy Steel Corporation in Canton, Ohio. He presented a paper in July 1929 that reviewed the work of Dr. Fry on steels containing chromium, aluminum, molybdenum, vanadium, and tungsten (Ref 4). In unison with McQuaid and Ketcham, Sergeson concluded that process chemistry and process control in nitriding were much simpler than in carburizing. He also reviewed the effect of reheating on the case after nitriding and found that, with increasing temperature, case hardness stability was much better than for carburized and quenched alloy steel. He noted that the surface hardness value for a chromium-aluminum steel began to decrease at only 525 °C (1000 °F), and only slightly. He worked with many more steels and compared the effect of temperature on both nitrided alloy steels and carburized and quenched alloy steels, yielding similar results. The process equipment that he used for his nitriding experiments was not unlike many modern gas nitriding furnaces, despite their improved materials of construction and computerized process control (Fig. 3). Sergeson examined the effect of both temperature and process gas flow on alloy steels and found that if the ammonia gas flow rate was increased at 510 °C (950 °F), little difference resulted in the immediate surface hardness and case depth. He also found that as process temperature increased, case depth increased but surface hardness decreased. His work covered alloy steels with chromium and aluminum and investigated the effects of varying aluminum and nickel contents. He concluded that nickel was not a nitride-forming element, but that it tended to retard

/ 7

8 / Practical Nitriding and Ferritic Nitrocarburizing

Door lift mechanism Fan drive motor

Air circulating fan

Refractory insulation

Furnace door Furnace thermocouple

Exhaust ammonia gas outlet tube

Nitride process chamber

To atmosphere exhaust Process delivery gas (ammonia)

Process chamber thermocouple tube Ammonia gas inlet tube

Load preparation table

Fig. 3

Schematic of a simple ammonia gas nitriding furnace

the nascent nitrogen diffusion if present in significant quantities. More detailed information on alloying effects can be found in Chapter 12, “Steels for Nitriding.” V.O. Homerberg and J.P. Walsted. Professor Homerberg was an associate professor of metallurgy at the Massachusetts Institute of Technology and consulted for the Ludlum Steel Company. Mr. Walsted was an instructor at the same university at the time that they presented their findings on the nitriding process (Ref 5). They studied the effects of temperature up to 750 °C (1400 °F), with its resulting increase in case depth but reduction of surface hardness. In addition, they studied the effects of decarburization on a nitrided surface and concluded that surfaces must be free of decarburization prior to nitriding. They reviewed Fry’s process technique and the decomposition of ammonia under heat. Once again, the equipment used for their experiments was not unlike the furnaces of today (with the exception, of course, of improved engineering materials of construction and furnace aesthetics).

Other Early Developments The Floe Process. During the early days of nitriding process technology, a persistent phenomenon was observed: an ever-present white layer on the nitrided steel surface. The white layer was identified as a multi-

Chapter 1: An Introduction to Nitriding

phase compound layer of ε and γ′ phases. Much recognition was given to Dr. Carl F. Floe of the Massachusetts Institute of Technology, who not only performed major research regarding identification of the layer and its characteristics, but also developed a process technique to reduce the layer thickness (Ref 6). Today that technique is known as the Floe process, or the two-stage process. The Floe process is carried out as two distinct events. The first portion of the cycle is accomplished as a normal nitriding cycle at a temperature of about 500 °C (930 °F) with 15 to 30% dissociation of the ammonia (i.e., an atmosphere that contains 70 to 85% ammonia). This will produce the nitrogen-rich compound at the surface. Once the cycle is complete, the furnace temperature is increased to approximately 560 °C (1030 °F), with gas dissociation increased to 75 to 85% (i.e., an atmosphere that contains 15 to 25% ammonia). Very careful gas flow control of the ammonia and its dissociation must be maintained during the second stage of the process. The two-stage process is used to reduce formation of the compound zone only; it serves no other purpose. Salt Bath Nitriding. Shortly after the development of gas nitriding, alternative methods of nitriding were sought. One such method was the use of molten salt as a nitrogen source. The salt bath process uses the principle of the decomposition of cyanide to cyanate and the liberation of nitrogen within the salt for diffusion into the steel surface. Salt bath nitriding is described in greater detail in Chapter 6. The ion, or plasma, nitriding process, which is based on the familiar chemistry of gas nitriding, uses a plasma discharge of reaction gases both to heat the steel surface and to supply nitrogen ions for nitriding (see Chapter 8 for details). The process dates back to the work of a German physicist, Dr. Wehnheldt, who in 1932 developed what he called the “glow discharge” method of nitriding. Wehnheldt encountered severe problems with the control of the glow discharge. He then partnered with a Swiss physicist and entrepreneur Dr. Bernhard Berghaus. Together they stabilized the process and later formed the company Klockner Ionen GmbH, specializing in the manufacture of ion nitriding equipment. Although the ion nitriding process developed by Wehnheldt and Berghaus was used successfully by German industrialists during World War II, it was not used extensively because it was considered too complex, too expensive, and too unreliable to guarantee consistent and repeatable results. Not until the 1970s did the process gain industrial acceptance, particularly in Europe. The significance of the glow discharge process was that it did not rely on the decomposition or cracking of a gas to liberate nascent nitrogen on the steel surface. The process was based on the ionization of a single molecular gas, which is nitrogen, and the liberation of nitrogen ions. The process offered a shorter cycle time due to the steel surface preparation and the gas ionization. Nitriding was not now restricted to steels that

/ 9

10

/ Practical Nitriding and Ferritic Nitrocarburizing

required specific nitride-forming elements. Today ion nitriding is carried out on virtually all steels and cast irons as well as refractory metals, aluminum (not yet on a commercial basis), and sintered ferrous materials. A schematic of an ion nitriding furnace layout is shown in Fig. 4. Ion Nitriding Production in the U.S. After WWII and through the late 1950s, the General Electric Company of Lynn, Massachusetts, operated a laboratory known as the Electromechanical Engineering and Physics Unit. Dr. Claude Jones and Dr. Derek Sturges, along with Stuart Martin, developed the first ion nitriding unit in the United States and applied the process to a variety of materials and parts. Their ion nitriding units met all normal nitriding standards and were accepted by the U.S. Navy. Summaries of the properties, applications, and advantages associated with the process were published in 1964 (Ref 7) and 1973 (Ref 8). Other Uses of Plasma Technology. Ion nitriding is not the only heat treatment process that utilizes the glow discharge phenomenon. Simply put, if one uses the appropriate process gases and the proper furnace, the glow discharge technique can be applied to plasma-assisted ferritic nitrocarburizing, carburizing, carbonitriding, and chemical vapor deposition. These plasma-assisted processes are described in various publications, including Heat Treating, Volume 4 and Surface Engineering, Volume 5 of ASM Handbook. Plasma technology is not new. One has only to observe the Northern Lights to witness a natural plasma. Lightning is also a natural plasma.

Vacuum process vessel

–

Work piece Power source

Process gas manifold (nitrogen, argon)

+

Vacuum pump

Fig. 4

Simple schematic of the layout of an early plasma (ion) nitriding furnace system

Chapter 1: An Introduction to Nitriding

Neon signs are a plasma glow, and “lumina storms” found in gift shops are just a few of the many examples of plasma technology at work.

Current Status of Nitriding Technology The success of any heat treatment is measured by hardness. However, hardness is relevant to the materials application and its mechanical requirements. Nitriding often is applied to low-alloy steels to “harden” the steel and improve corrosion resistance. In addition to conventional nitriding, the following processes have been developed: • • • •

Oxynitride process, during which a controlled postoxidation treatment is carried out to further enhance the surface corrosion resistance Ferritic nitrocarburizing (a controlled process using nitrogen and carbon to enhance surface characteristics of low-alloy steels) Derivatives of the two previous processes Controlled nitriding, which is a further development of traditional gas nitriding in which all the process parameters are computer controlled

Nitriding has reached maturity and become an accepted, though sometimes misunderstood, process. Both the gas and salt systems have run an almost parallel course since the early part of the 20th century. The process has found its place in both low- and high-tech applications and is becoming better understood by process technicians, metallurgists, applications engineers, furnace designers, and academics. Many developments in process techniques are being driven by environmental concerns and legislation. This has resulted in the introduction of more effective, efficient, and economical methods and equipment. Improvements can be seen in the development of gaseous methods, salt bath methods, fluidized-bed methods, and plasma processing techniques. REFERENCES 1. A. Machlet, U.S. Patent 1,092,925, 24 June 1913 2. A. Fry, U.S. Patent 1,487,554, 18 March 1924 3. H.W. McQuaid and W.J. Ketcham, Some Practical Aspects of the Nitriding Process, reprinted from Trans. ASST, Vol 14, 1928, Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 1–25 4. R. Sergeson, Investigation in Nitriding, reprinted from American Society for Steel Treaters (ASST) Nitriding Symposium, 1929, in Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 26–55 5. V.O. Homerberg and J.P. Walsted, A Study of the Nitriding Process— Part I, reprinted from American Society for Steel Treaters (ASST) Nitriding Symposium, 1929, in Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 56–99

/ 11

12

/ Practical Nitriding and Ferritic Nitrocarburizing

6. C.F. Floe, A Study of the Nitriding Process Effect of Ammonia Dissociation on Case Depth and Structure, reprinted from Trans. ASM, Vol 32, 1944, Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 144–171 7. C.K. Jones and S.W. Martin, Nitriding, Sintering and Brazing in Glow Discharge, Met. Prog., Feb 1964, p 94–98 8. C.K. Jones, D.J. Sturges, and S.W. Martin, Glow-Discharge Nitriding in Production, reprinted from Met. Prog., Dec 1973, Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 186–187

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p13-22 DOI: 10.1361/pnafn2003p013

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

2

CHAPTER

Why Nitride? THE UNIQUE ADVANTAGES of the nitriding process were recognized by the Germans in the early 1920s. It was used in applications that required: • • • • •

High torque High wear resistance Abrasive wear resistance Corrosion resistance High surface compressive strength

Early on, the process did not gain much recognition in the United States because of its moderate hardness values for plain carbon, cast iron, and low-alloy steels. The very long nitriding process cycle times required to reach the same case depths achieved by more conventional methods such as carburizing were considered a disadvantage. For example, to achieve a case hardening depth of 1.0 mm (0.040 in.), the nitriding process requires 90 h, compared to 4.5 h for carburizing (Fig. 1, 2). As described in Chapter 1, the patent for gas nitriding was first applied for by Adolph Machlet and was for the nitrogenization of iron and steel in

1700

4.5 h 927

800

1475

Temperature,°C

Temperature,°F

Total case depth required = 1.0 mm (0.040 in.)

Quench

Time

Fig. 1

Example of carburizing, followed by quenching, to produce a total case depth of 1.0 mm (0.040 in.)

Total case depth required = 1.0 mm (0.040 in.) 90 h 925

495 Free cool

Temperature,°C

/ Practical Nitriding and Ferritic Nitrocarburizing

Temperature,°F

14

Time

Fig. 2

Example of nitriding at 495 °C (925 °F), followed by free cooling, to produce a total case depth of 1.0 mm (0.040 in.) on a simple nitriding steel

an ammonia gas atmosphere diluted by hydrogen. The hardness results that he achieved were not high by the standards of today or even by the German standards of the day. Hardness results generally measure the success of the process and are expected to be in the region of approximately 60 to 64 HRC. Early surface hardness values obtained by Machlet were in the region of 30 to 35 HRC, considered too low in terms of wear properties. Keep in mind, however, that hardness is relative to the wear characteristics of the steel part being treated. What was not recognized was the excellent corrosion resistance that nitriding imparted to low-alloy steels and cast irons. The Germans owed their success to Adolph Fry’s work at Krupp Steel, where he developed the special Nitralloy steels that became synonymous with the nitriding process (and which exhibited considerably higher hardness values than those obtained by Machlet). German industrialists began to control the steel analysis for nitridable steels, licensing it to various steelmakers in other countries. The attitude toward the use of the process in America began to change after the 1927 presentation of Fry’s paper at the SME conference in Chicago.

Key Process Considerations Several factors helped nitriding to gain acceptance: • • • • •

Compared to other case hardening methods, nitriding is a relatively low-temperature process. Nitriding is relatively easy to control in terms of process parameters. It produces enhanced corrosion resistance in low-alloy and low-carbon steels. Core hardness is not significantly affected, due to prehardening and tempering. No quenching is required, thus reducing distortion.

Low-Temperature Process. The nitriding process requires a relatively low temperature compared to the more widely recognized surface treatment methods. The temperature employed is in the region of 500 °C (925 °F),

Chapter 2: Why Nitride?

although it varies according to the steel being treated. Temperature selection is based on the final tempering temperature of the steel during its preheat-treatment procedure. Should the nitriding temperature be at or above the tempering temperature, then the core hardness of the pretreated steel will diminish according to time and temperature. It is most important to maintain the core hardness of the pretreated steel to provide adequate support to the diffused case. Without good core support, the case can fail. Surface treatment processes that require higher temperatures than the nitriding process requires include: • •

Carburizing, which employs a temperature in the region of 970 °C (1775 °F) (Fig. 3a) Carbonitriding, which employs a temperature in the region of 870 °C (1600 °F) (Fig. 3b)

Figure 4 compares the temperature ranges used by various diffusion surface hardening techniques, along with case depth characteristics. Case depths accomplished by carburizing are usually considerably deeper than those accomplished with nitriding. Some carburized case depths can be up

2.5 h 925

1475

800

Quench

Temperature,°C

Temperature,°F

1700

Time

(a)

Temperature,°F

870

1475

(b)

800

Temperature,°C

2h 1600

Time

Processing time-temperature plots. (a) Illustration of carburizing to produce a total case depth of 0.70 mm (0.028 in.) at 925 °C (1700 °F). (b) Illustration of carbonitriding to produce a total case depth of 0.38 mm (0.015 in.) at 870 °C (1600 °F)

Fig. 3

/ 15

16

/ Practical Nitriding and Ferritic Nitrocarburizing

Thermochemical diffusion techniques Cabontride

Carburize Pack Gas Salt Ion Gas Diffuses carbon into the steel surface Process temperatures 1600-1950°F (870-1065°C) Case depth: medium

Fig. 4

Ferritic nitrocarburize

Salt Ion Gas

Diffuses carbon and nitrogen into the steel surface Process temperature 1550-1650°F (845-900°C) Case depth: shallow

Boronize

Salt Ion Pack

Diffuses carbon nitrogen, sulfur, oxygen (individually or combined) into the steel surface Process temperatures 1050-1300°F (565-705°C) Case depth: shallow

Nitride Gas Pack Gas Salt Ion

Diffuses boron into the steel surface Process temperatures 1400-2000°F (760-1095°C) Case depth: shallow

Diffuses nitrogen into the steel surface Process temperatures 600-1020°F (315-550°C) Case depth: shallow

Comparison of various diffusion surface hardening techniques. Source: Ref 1

to approximately 6.35 mm (0.250 in.) (requiring an extremely long cycle at the process temperature). The downside of the deep case, of course, is the occurrence of grain growth, which is inevitable due to the extended cycle time at elevated temperatures for the carburizing procedure. Nitriding takes place in the ferrite region on the iron-carbon equilibrium diagram (Fig. 5), which means that grain size in both the surface and the core is not affected. However, the same deep case achieved by carburizing cannot be accomplished by nitriding. Once the cycle time extends beyond 90 h, there is no commercial or metallurgical advantage to be gained. Nascent nitrogen has a strong affinity for iron and steel at elevated temperatures and will readily diffuse. The higher the temperature to which the steel is elevated, the faster and deeper the nitrogen diffusion occurs. However, caution must be exercised in relation to: • • • • •

Temperature selection Gas dissociation Surface area of treated work Steel chemistry Quality and type of formed case that is required

If process temperature selection is too high, then a saturated solution of nitrogen in iron will occur, which can cause the problem known as nitride networking. This condition will embrittle sharp corners, leading to spalling or even exfoliation. No Quench Requirement. With conventional surface diffusion process techniques such as carburizing and carbonitriding, the steel must be at a suitable austenitizing temperature (depending on the steel composition) after carburizing, followed by quenching (or rapid cooling) to transform the austenite phase (face-centered cubic, or fcc, lattice) into martensite (bodycentered tetragonal, or bct, lattice) (Fig. 6). One should also be cognizant of

Chapter 2: Why Nitride?

1800

3270

1700

3090

1500 (δ-Fe)

1400 1394°C 1300

Temperature,°C

1200 1100 1000 900

Solubility of graphite in liquid iron

Liquid

1227°C

4.26% 2.08% 1154°C

(γ-Fe) Austenite

Austenite + cementite

2730 2550 2370 2190

2.11% 1148°C 4.30%

0.68% 912°C A3 Acm

800 770°C 700 0.77% 600

2910

1538°C 1495°C

6.69% Cementite (Fe3C)

2010 1830 1650

738°C

1470

Temperature,°F

1600

1290

A1(727°C)

1110

500

930

(α-Fe) Ferrite 400

Ferrite + cementite

300

750 570

200

390

100

210

0 30 Fe 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 Carbon, wt%

Fig. 5

Iron-carbon equilibrium diagram. The nitriding process is carried out at temperatures below the A1 line.

Face centered cubic structure Additional carbon dissolves into structure Rapid cooling

Heating to high temperature

Iron atoms Carbon atoms Room temperature body centered cubic structure

Fig. 6

Room temperature body centered tetragonal structure

Crystal lattice changes that take place during high-temperature heat treatment processes such as carburizing. Ferrite is bcc structure; austenite, fcc; martensite, bct. Source: Ref 2

/ 17

/ Practical Nitriding and Ferritic Nitrocarburizing

the size of the lattice structure between the two phases of austenite and martensite. Remember that austenite is an enforced condition in an alloy carburizing steel, as is martensite. Martensite cannot be created without the transformation from austenite by rapid cooling. Also, in order to form martensite it is necessary to have carbon present in sufficient amounts when the steel is rapidly cooled (Fig. 7). A rapid cooling rate is unnecessary for nitrided steel after completion of the process cycle. The process chamber free cools under the ammonia atmosphere down to a suitable temperature of approximately 200 °C (400 °F) and then is purged with clean, dry nitrogen. Unlike the ferritic nitrocarburizing, carbonitriding, and carburizing procedures, the nitriding process does not involve a critical cooling rate. Minimal Distortion. Distortion is one of the most persistent problems facing heat treaters and engineers. Distortion manifests itself during the final heat treatment process in the form of either:

Austenite (fcc) 14 atoms

Martensite start

Quench

Ferrite (bcc) 9 atoms

Martensite (bct) 9 atoms

Time

Schematic illustration of the phase transformation taking place when hardening steel with sufficient carbon present. Crystal lattice: bcc, body-centered cubic; fcc, face-centered cubic; bct, body-centered tetragonal

Fig. 7

Martensite start line

Austenite and phase change

He at (stres up s reli eve)

•

Shape distortion: a change in geometrical form, such as curving, twisting, or bending Size distortion: a change in workpiece volume due to either growth or shrinkage

Temperature

•

Temperature

18

Quench (thermal shock) Phase change (stress induced)

Time

Fig. 8

Phase changes in relation to thermally induced stress on quenching

Chapter 2: Why Nitride?

( st

H res e a t u p sr elie ve )

Nitride

wn l do C oo h a s e p ( n o a nge ) ch

Temperature

Distortion can be minimized but not eliminated. Stresses are induced into the steel through rolling and forging as well as machining procedures. These induced stresses will be manifested at the final heat treatment (Fig. 8). As steel is heated, the heat-up cycle becomes a stress relieving process, and the stresses begin to be relieved from the steel. If distortion is to be kept to a minimum, then it is necessary to introduce an interim stressrelieving procedure after the preheat treatment and rough machining and before the final machining and nitriding process. The stress relief process will act as a stabilizing process on the final nitriding process. Another method of stabilizing is to cryogenically treat the steel (particularly an alloy or tool steel). This will transform any retained austenite into untempered martensite that will be tempered by the nitriding process, ensuring better dimensional stability during and after nitriding (Fig. 9). If the heat treater exercises careful control over the heating rate during the heat-up cycle, stress relieving can be minimized. The reverse side of the coin becomes apparent after the appropriate soaking period has been completed and it becomes necessary to cool the steel rapidly, as during hardening of either a through-hardened steel or a carburized or carbonitrided steel. Thermal stress patterns are then induced into the steel due to geometric section changes, differential cooling rates, and phase changes in the steel.

(No phase change) (Additional temper) (Dimensionally stabilized if retained austenite is present) Time

Fig. 9

Benefits of the nitriding process

Before nitriding

Fig. 10

Effect of growth due to nitriding

After nitriding

/ 19

20

/ Practical Nitriding and Ferritic Nitrocarburizing

In addition, when another element or combination of elements is added to create changes in surface chemistry, volume changes will occur in the form of size change and uniform growth. For example, a ring will grow in terms of outside and inside diameters, the net effect being that the bore diameter decreases (Fig. 10). Nitriding is a relatively low-temperature process and produces a shallower case than does carburizing. Most importantly, no quench is involved. On completion of the process, the steel can be cooled down naturally under its atmosphere or force-cooled using clean, dry nitrogen. Generally the steel is allowed to cool naturally, thus reducing the risk of thermal gradients due to sectional changes. If the engineer can live with extended cycle times, the nitriding process can offer many distinct advantages: • •

Reduced distortion Reduced final machining time

• •

Improved part cleanliness Improved dimensional stability

If one considers the cost of scrap due to distortion and time spent straightening, nitriding is not an expensive process compared to carburizing. However, in terms of high-volume production requirements (e.g., automotive gears), nitriding does not always present a viable option. The nitrided case usually exhibits greater dimensional stability simply because there is no opportunity for retained austenite to form, as can occur in carburizing and quenching. Over time, retained austenite will decompose to untempered martensite. In carburizing, retained austenite will leave mixed phases in the formed case, which can lead to dimensional instability. The nitriding process does not promote dimensional instability. It also acts as a stress relieving procedure. However, growth most certainly takes place during the nitriding process, due primarily to the diffusion of nitrogen into the steel surface. The amount of growth is influenced by: • • • • • • • •

Time Temperature Steel chemistry Gas flow Gas dissociation Steel surface condition prior to the nitriding process Surface metallurgy (compound layer thickness) Total case depth

Control of the growth aspect of nitrided steel is discussed in Chapter 11, “Distortion.” High Hardness Values. It has often been said that the nitriding process can only be applied to special steels that contain specific alloying ele-

Chapter 2: Why Nitride?

ments with an affinity for nitrogen and which form stable nitrides. A surface-treated steel often is considered inadequate if a high hardness value has not been achieved. Pure iron and low-alloy steels will nitride. However, they will exhibit maximum hardness values of about 35 HRC (by gas nitriding). In the early days of process development, Fry developed the Nitralloy group of special alloy steels that produce high hardness values after nitriding. The Nitralloy steels contain alloying elements such as chromium, molybdenum, vanadium, tungsten, and aluminum. Because these steels tended to be more expensive than the more conventional case hardening steels, reluctance toward their use quickly developed. This attitude began to dissipate once larger lots of these steels were produced and their benefits (minimal growth and distortion, high hardness) were recognized. The high hardness values achieved in the Nitralloy steels are due to the affinity of the alloying elements to form stable nitrides at designated process temperatures. The resulting hardness value is a function of the amount of these elements present. Considerably higher hardness values are exhibited with steels containing up to approximately 1 to 3% Al. Above 3%, there is no effect on hardness. Although this discussion has centered on the Nitralloy steels, steels that contain the same elements either individually or collectively will nitride. This includes stainless steels, tool steels, and alloy steels. Resistance to Oxidation. Compared to steels that have undergone traditional case hardening techniques, nitrided steels offer improved corrosion and oxidation resistance. The nitrided surface of an alloy steel or tool exhibits increased resistance to saltwater corrosion, moisture, and water. However, this does not apply to stainless steels; in fact, their resistance to corrosion will be reduced. This is because chromium has an affinity for oxygen, readily forming chrome oxide on the stainless steel surface. The chrome oxide acts as a barrier to nitriding. For diffusion to occur, the surface must be passivated, thus reducing corrosion resistance. Although nitriding improves the corrosion properties of alloy steels, the improvement is not permanent. Surface degradation or pitting will eventually occur, albeit not as rapidly as might occur had the steel not been nitrided. The core properties of nitrided alloy steels usually do not change, but this does not apply to lower-alloy steels. Nitriding normally is performed at a temperature below that of the final tempering temperature of the steel after the prehardening sequence. Care must be taken during the prehardening and tempering procedure to minimize surface decarburization; any surface decarburization must be removed prior to nitriding. It is usually good practice to have an adequate core hardness value in order to support the final nitrided case and a tempered martensite core.

/ 21

22

/ Practical Nitriding and Ferritic Nitrocarburizing

If the steel is nitrided at a temperature below that of the final tempering temperature, then the core hardness value will not be affected. A process temperature approximately 55 °C (100 °F) below the final tempering temperature generally is selected.

Concluding Remarks The nitriding process was first put to commercial use in the automotive industry on transmission gears, particularly spiral bevel pinions and spiral bevel gears because of their natural tendency to try to “straighten” at the heat treatment process temperature. These types of gears were notorious for distortion on the gear teeth. Current spiral bevel machining design practice is to cut distortion into the gear. This technique requires an intimate knowledge of the steel combined with extremely good control of the heat treatment process. Nitriding has become more widely accepted because of its ability to serve many applications that previously were not considered possible or even worthwhile. Engineers and metallurgists use nitriding creatively, making it a viable and commercially acceptable process. As our understanding of the benefits associated with the process evolves, nitriding will continue to grow in use and popularity. REFERENCES 1. D. Pye, Diffusion Surface Treatment Techniques: A Review, Ind. Heat., March 2001, p 39–44 2. K.G. Budinski, Diffusion Processes, Surface Engineering for Wear Resistance, Prentice Hall, 1988, p 78–119

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p23-30 DOI: 10.1361/pnafn2003p023

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

3

How Does the Nitriding Process Work? SEVERAL PROCESS PARAMETERS must be considered in order to ensure nitriding success in terms of metallurgy and distortion: • • • •

Nitrogen source Heat Time Steel composition

In gas nitriding the nitrogen source is almost always derived from the decomposition (or dissociation) of ammonia gas supplied via an external bulk storage system or individual bottles connected to a manifold (Fig. 1). Ammonia gas begins to decompose when heat is applied, usually from an external source within the furnace. At the usual nitriding temperatures of 500 to 570 °C (930 to 1060 °F), ammonia is in an unstable thermodynamic state and decomposes in the following manner: 2NH3 ↔ 2N + 3H2

(Eq 1)

D

A

Fig. 1

B

C

Simple schematic arrangement of an ammonia gas nitriding system. A, bulk storage tank; B, gas nitriding furnace; C, gas dissociation test station; D, exhaust to atmosphere. Source: Pye Metallurgical Consulting Inc.

24

/ Practical Nitriding and Ferritic Nitrocarburizing

Typically, three reactions take place at the steel surface when the steel is at the set process temperature: NH3 → 3H + N 2N → N2 2H → H2

(Eq 2) (Eq 3) (Eq 4)

The atomic nitrogen and hydrogen components shown in Eq 2 are unstable and will unite with other like atoms to form molecules as shown in Eq 3 and 4. It is while they are in the atomic state that diffusion takes place. The released nitrogen diffuses into the steel at the nitriding processing temperature, but very slowly, to the point where it is not economically practical or effective. The temperature of 500 °C (930 °F) is considered to be an “economical” temperature. Nascent nitrogen has an affinity for steel and iron and will readily diffuse into both materials at elevated temperatures. The higher the temperature, the faster and deeper the nitrogen diffusion. An economical temperature is one that produces an optimum case depth while not adversely affecting the core properties of the treated steel.

The Liberation of Nitrogen (Ref 1) Nitrogen decomposes or dissociates in accordance with Eq 1. At the instant of decomposition, the liberated nitrogen will exist as nascent or atomic nitrogen and as such can be absorbed by the steel. Nitrogen has an atomic diameter of 0.142 nm and is dissolved in iron in interstitial positions in octahedral voids of the cubic lattice that have a maximum diameter of 0.038 nm in bcc alpha (α) iron and a maximum diameter of 0.104 nm in fcc gamma (γ) iron. Nitriding of pure iron at temperatures up to 590 °C (1094 °F) with an increasing nitrogen content, according to the binary Fe-N phase diagram (see Fig. 1 in Chapter 1), leads to formation of the following phases: • • •

Body-centered cubic α iron, which dissolves 0.001 wt% N at room temperature and 0.115% N at 590 °C (1095 °F) Face-centered cubic γ′ nitride, Fe4N, which dissolves 5.7 to 6.1% N Hexagonal (epsilon) (ε) nitride, Fe2-3N, which exists in the range of 8 to 11 wt% N

Orthorhombic zeta (ζ), Fe2N, forms at temperatures below 500 °C (930 °F) and nitrogen contents exceeding 11 wt%—conditions that are not used in nitriding practice. The process of the smaller nitrogen atoms passing between the ironbase crystals as heat is applied up to a suitable process temperature is known as “interstitial diffusion.” This process is shown schematically in Fig. 2.

Chapter 3: How Does the Nitriding Process Work?

Fig. 2

Schematic of interstitial diffusion during the nitriding process. The work is heated to the nitriding temperature with ammonia flowing into the retort. The ammonia gas dissociates to nitrogen and hydrogen at the part surface. The nitrogen diffuses into the work in atomic form, and the hydrogen becomes a part of the furnace atmosphere.

Dissociation of the Gas at the Selected Nitriding Temperature (Ref 1) At temperatures around 500 °C (930 °F), the stability of ammonia is questionable, leading to dissociation rates higher than 98% and thus to formation of a protective gas without any nitriding effect. Despite the beneficial catalytic effect of the workpiece surfaces and the furnace wall, the dissociation of ammonia is an extremely slow process. Therefore, ammonia-based nitriding atmospheres for steel treatment rarely contain less than 20% ammonia and frequently up to 50%, and thus their degree of dissociation is far from equilibrium. The remaining ammonia content is decisive for the effect of nitriding, during which nitrogen diffuses into the steel according to the reaction (Ref 2): NH3 → N(α) + (3/2)H2

(Eq 5)

which occurs at the boundary layer. The nitrogen activity, which is the driving force in the mass transfer, can be calculated according to (Ref 3): aN = KN

pNH3 p3/2 H

(Eq 6)

where KN is the equilibrium constant of nitrogen, and pNH3 and p3/2 H are the partial pressures of ammonia and hydrogen, respectively. The nitrogen transfer is comparatively low, and the release of hydrogen from the ammonia molecules determines the process rate. Therefore, as mentioned earlier, nitriding times are rather long, up to 120 h.

/ 25

26

/ Practical Nitriding and Ferritic Nitrocarburizing

The nitriding effect of a nitriding atmosphere is defined by its degree of dissociation, with high degrees of dissociation always indicating a near-equilibrium state where the nitriding effect is low. In processing, it is common to use a constant degree of dissociation (e.g., 30%), but sometimes a two-stage procedure involving variations in temperature and dissociation degree (the Floe process, described in Chapter 1) is followed. However, the measured ammonia content is not equivalent to the actual degree of dissociation. In ammonia dissociation according to Eq 1, two molecules of ammonia dissociate to one molecule of nitrogen and three molecules of hydrogen. This increase in volume dilutes the ammonia content, as do additional gases, and must be taken into account in determining the actual degree of dissociation. The nitriding effect can be determined more easily by means of the nitriding potential: Np =

pNH3 p3/2 H2

(Eq 7)

The nitriding potential is generally used for describing the nitridability of an ammonia atmosphere. Its control allows the formation of predictable case structures and depths for a range of steels, including some tool and stainless grades (Ref 4). Higher Np values produce higher surface concentrations of nitrogen and steeper concentration gradients. Lower potentials allow the development of nitrided cases with no brittle-compound (white) layer in high-alloy steels.

Why Ammonia Is Used Because the nitriding process is carried out only in the α-iron phase, 0.115 wt% N will be dissolved at 590 °C (1095 °F). At 500 °C (930 °F), the percentage by weight of dissolved nitrogen will drop accordingly to approximately 0.099 wt%. Thus, low solubility values of nitrogen in α-iron, combined with the high partial pressure values for gaseous nitriding that are needed for nitrogen absorption, make use of bottled nitrogen impossible. Gaseous molecular nitrogen will not dissociate into nascent nitrogen when cracked by heat at process temperature, and thus no diffusion of molecular nitrogen will occur. The steel surface, in conjunction with heat, will act as a catalyst for the decomposition of ammonia into nascent nitrogen for diffusion into the steel surface. The carbon in the steel plays a part in the reaction because the carbon is present in solution with the iron in austenite. Nitrogen has a stabilizing effect on austenite, which will transform very slowly. Thus when martensite is formed as a result of carbon in austenite in the new form of Fe-N, and at an elevated temperature, the resulting structure means that the

Chapter 3: How Does the Nitriding Process Work?

nitriding is limited to the ferrite region on the iron-carbon diagram. The bcc lattice structure remains unchanged. Growth, however, will be experienced due to an increase in the volume of the immediate steel surface and by the nitrogen diffusion (Fig. 3). What this suggests is that the observable growth will be primarily to surface volume rather than to the core, as no phase changes occur in either the core or the immediate surface. Nascent nitrogen is being diffused at the surface and dispersed into the immediate surface, increasing the surface volume. The atomic nitrogen and hydrogen as described in Eq 1 is considered unstable and will begin to unite with like atoms to form a molecule as shown in Eq 2 and 3. During the atomic state is when diffusion into the steel begins to take place. Because gaseous molecular nitrogen will not dissociate into nascent nitrogen when cracked by heat at process temperature, no diffusion of molecular nitrogen will occur. The steel surface, in conjunction with heat, will act as a catalyst for the decomposition of ammonia into nascent nitrogen and for diffusion into the steel surface. Provided that all the appropriate conditions are met, nascent nitrogen will begin to diffuse into the steel surface at a rate determined by the process temperature. In other words, the process temperature can be as low as 350 °C (600 °F) and will react with the alloying elements to begin to form the solid solution of stable nitrides. Conversely, the temperature can be as high as 600 °C (1050 °F). However, although the reaction of the nascent nitrogen will still occur, it will also strongly affect the thickness of the surface compound layer and give rise to the risk of nitride networking.

Distortion Do not be misled into thinking that the nitriding process causes absolutely no distortion. It is a question of the definition of distortion. Distortion resulting from the nitriding process takes the form of uniform growth

Growth Growth Growth Growth Growth Growth

Fig. 3

Illustration of the growth in volume due to nitriding. The amount of growth will be determined by the time at the selected process temperature.

/ 27

28

/ Practical Nitriding and Ferritic Nitrocarburizing

at the immediate surface. The amount of growth is determined by the following factors and will be seen as a dimensional change: • • • •

Steel chemistry Gas dissociation or gas ratios Process temperature selection Time at process temperature which influences the thickness of the compound layer

Distortion (twisting and bending that leads to shape change) can result from: • • • •

•

Mixed residual phases due to an excessively slow cooling rate on prehardening (presence of retained austenite) Residual stress patterns due to machining and the omission of intermediate stress relieving Stacking of components atop one another in the process chamber Selection of the nitriding temperature when residual phases are present. If the process temperature is higher than the tempering temperature, any residual phases could still be transformed. Omission of a stabilizing process prior to the nitriding procedure (cryogenic treatments)

Preheat Treatment The steel must be preheated to establish core properties that will support the nitrided case during its operation. If the core is not preheat treated and is left in, for example, the annealed condition, nitriding will be very difficult to accomplish. Even if it were possible to nitride an annealed steel, the workpiece would not perform well within its operating environment simply because a load on the case could exceed the core strength of the steel. This is one of the reasons why it is necessary to preheat treat the steel (harden and temper to produce tempered martensite). Annealed steel is in the ferritic condition; that is, the atomic structure of the ferrite phase is in the bcc lattice structure. On heating the steel for the hardening operation, the phase will change to austenite, which is a 14-atom fcc lattice configuration (see Fig. 6 and 7 in Chapter 2). On quenching for the hardening operation, the lattice structure changes to that of martensite, which has reverted back to a 9-atom structure, but now in the bct configuration. This now means that the steel is in the phase of fresh martensite, which is its most unstable condition. The steel now must be tempered to reduce the core hardness back to the appropriate hardness value that will best support the nitrided case during operation.

Chapter 3: How Does the Nitriding Process Work?

If during the hardening operation austenite remains with the untempered martensite, that retained austenite must be transformed to untempered martensite. This can be achieved during either the tempering or the nitriding operation. Remember, the nitriding process temperature is slightly lower than the primary tempering temperature of the steel. The tempering procedure will decompose the retained austenite into a tempered martensite condition. In addition, the nitriding process temperature will also act as a stabilizing procedure for the steel and enhance dimensional stability. What hardness will be the correct hardness? The answer requires determining what type of loading will be placed on the case. The core hardness will also be determined by the steel chemistry, the austenitizing temperature, and the final tempering temperature. REFERENCES 1. J. Grosch, Heat Treatment With Gaseous Atmospheres, Steel Heat Treatment Handbook, G.E. Totten and M.A.W. Howes, Ed., Marcel Dekker, 1997, p 663–719 2. C.A. Stickels and C.M. Mack, Overview of Carburizing Processes and Modeling, Carburizing: Processing and Performance, G. Krauss, Ed., ASM International, 1989, p 1–9 3. B. Edenhofer and H. Pfau, Self-Adaptive Carbon Profile Regulation in Carburizing, Heat Treatment and Surface Engineering: New Technology and Practical Applications, G. Krauss, Ed., ASM International, 1988, p 85–88 4. G.J. Tymowski, W.K. Liliental, and C.D. Morawski, Take the Guesswork out of Nitriding, Advanced Materials & Processes, Dec 1994, p 52–54

/ 29

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p31-38 DOI: 10.1361/pnafn2003p031

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

4

Microstructures of Nitrided Iron and Steel FORMATION OF THE NITRIDED CASE begins through a series of nucleated growth areas on the steel surface. These nucleating growth areas will eventually become what is known as the “compound layer” or, more commonly, the “white layer.” This layer is usually very hard and brittle and comprises two intermixed phases. The layer does not diffuse into the steel, but remains on the immediate surface and grows thicker with time, temperature, and gas composition (Fig. 1). The carbon in the steel changes the morphology of the nucleation process, thus causing a mixed phase formation at the steel surface, with the diffused and reacted nitrides forming beneath the surface. The region immediately beneath the white layer is called the “diffusion zone.” This region is made up of stable nitrides formed by the reaction of nitrogen with nitride-forming elements. The area below the diffusion zone is the core of the steel, which consists of tempered martensite. All three regions—the white layer, diffusion zone, and core—are shown in Fig. 2. The nitrides begin their formation by the nucleation of γ′ at the immediate steel surface interface with the nitriding atmosphere. This nucleation Compound zone, dual phase Diffusion zone consisting of formed nitrides Transition zone from diffusion zone to core material

Core material

Fig. 1

Typical nitrided case

32 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 2

Typical nitrided case structure showing the white layer (top), the diffusion zone, and the core below the diffusion zone. Source: Ref 1

process progresses and continues until the subsequent nucleation of ε at the steel surface interface (Fig. 3). Note that the nitrogen diffusion is much slower in the compound layer than in the steel substrate.

Influence of Carbon on the Compound Zone Carbon manipulates the amount of γ′ and ε formed within the surface structure. In a typical nitriding steel with a carbon content of approximately 0.4 wt%, the formation of γ′ to ε will be roughly equal using the gas nitriding process (Ref 1). The higher the carbon content of the steel, the greater the ε-phase in the compound layer. The lower the carbon content, the greater the γ′-phase. The thickness of the compound zone is a function of time, temperature, and pressure (usually atmospheric pressure). The amount of carbon in the steel has a small effect on thickness. Again, however, carbon content considerably affects the composition of the compound layer, determining whether the layer will be predominantly γ ′, ε, or equal amounts of each phase. If the compound layer thickness is a critical issue, then the steel must be selected carefully to accomplish the required surface metallurgy.

Controlling Compound Zone Thickness The thickness of the compound layer is controllable to a large extent by manipulating the process technique. As discussed subsequently, it can be controlled by dilution, the two-stage Floe process, or by ion nitriding.

Chapter 4: Microstructures of Nitrided Iron and Steel / 33

Fig. 3

Schematic showing the nucleation of γ ′- and ε-nitrides on iron. Source: Ref 2

Whether the compound zone is detrimental to the formed case will depend entirely on the application of the part and its working environment. This can only be established by knowing how the compound zone will perform under operating conditions. Is it advantageous or detrimental to form an either γ′-rich or ε-rich compound zone or to have equal amounts of each phase in the zone, or even to have only a minimal compound layer thickness? Compound layer formation is perhaps the most controversial issue related to the nitriding process and has been the subject of many debates, discussions, and research papers (Ref 3). See, for example, the bibliography published in conjunction with Ref 4, which includes 33 references that describe the influence of compound layer thickness, phase composition, chemical composition, and microstructure on the behavior of nitrided steels. Dilution. When hydrogen is added to the flow of ammonia during gas nitriding, the nitrogen component of the ammonia is diluted. This reduces the amount of available nascent nitrogen for diffusion into the steel surface.

34 / Practical Nitriding and Ferritic Nitrocarburizing

Controlling the nitrogen availability allows control of the compound layer thickness and, to a large extent, the layer composition. Two-Stage Process. As the name implies, the process developed by Dr. Carl Floe in 1942 (see Chapter 1 and Ref 5) involves a two-step procedure: one step at the normal nitriding temperature (500 °C, or 930 °F) with normal ammonia gas dissociation (between 15 and 30%), and a second step at a higher temperature between 540 and 565 °C (975 and 1050 °F) with a gas dissociation rate between 75 and 80%. The net effect is a much thinner compound layer. The principle of the procedure is to reduce the amount of available nitrogen for surface diffusion and to ensure a rapid diffusion by raising the process temperature. The temperature for the second stage must be carefully selected. Selection of higher process temperatures poses the risk of grain-boundary networking with iron nitrides at the periphery of the grain boundaries, leading to premature component failure at sharp corners (Fig. 4). When considering use of the two-stage process, it may be prudent not to employ the higher temperature. The higher temperature does not influence the final hardness value, only the diffusion rate. Ion nitriding is gaining much recognition. The process relies on the creation of a gaseous plasma under vacuum conditions. The process gases can be selected in whatever ratio suits the required surface metallurgy. In other words, the formation of the compound layer can be single phase, dual phase, or diffusion zone only. The surface metallurgy can be manipulated to suit both the application and the steel. Ion nitriding has opened the door to many applications that were not possible with conventional nitriding techniques.

Fig. 4

A sharp corner profile, illustrating the effects of nitrogen enrichment (nitride networking) at the corner

Chapter 4: Microstructures of Nitrided Iron and Steel / 35

What Happens Below the Compound Zone? The compound layer will always be thicker on plain carbon steel than on alloy steel for the same cycle time and temperature. However, the metallurgy will be completely different in terms of both surface compound layer and diffusion zone. Beneath the compound zone at nitriding temperature, the nitrogen dissolves into the α-iron and also reacts with certain alloying elements—such as aluminum, molybdenum, chromium, tungsten, vanadium, and silicon—if they are present. All of these alloying elements form nitrides in steel. This area of nitride formation below the compound zone is known as the diffusion zone (Fig. 1, 2). The nascent nitrogen will begin to react with those elements that will readily form stable nitrides. The nitrides exhibit very high hardness values, particularly the aluminum nitrides.

Can Plain Carbon Steel Be Nitrided? Contrary to popular belief, plain carbon steel or even plain iron can be nitrided (Fig. 5). However, the compound zone will be much thicker than for alloyed steel with equal carbon content. This is because the nitrides formed by the alloying elements will contain more nitrogen than those formed with iron, or iron in the plain carbon steel. In addition, the hardness values will be much lower. Plain carbon steels that have been pretreated to a tempered martensite structure typically will exhibit hardness values of 400 to 700 HV (49 to 60 HRC) under a 200 gf microhardness load. Alloy steels can exhibit hardness values from 700 to 1000 HV (60 to 69 HRC) under the same load. Surface hardness values for stainless steels, particularly when surface chemistry reactions are controlled (as in the plasma nitriding process), can reach 1500 HV (78 HRC).

Fig. 5

The metallographic appearance of AISI 1015 (UNS G10150) steel after a 2 h vacuum nitrocarburizing treatment in an ammonia/methane mixture with 1% oxygen addition

36 / Practical Nitriding and Ferritic Nitrocarburizing

Only in a plain carbon steel (one without nitride-forming elements) will a supersaturated solution of α-iron occur. It is quite possible for the iron nitrides to segregate on slow cooling from the nitriding temperature or via an aging process at a low-temperature set point. Another often overlooked feature of the nitriding process (even on a plain iron or plain carbon steel) is enhancement of the steel surface corrosion characteristics. By contrast, the corrosion resistance of stainless steels will be reduced, due to breakdown of the surface chrome oxide barrier to enable nitrogen diffusion into the steel. Once the chrome oxide barrier is depleted, the ability of the stainless steel to resist corrosion is reduced.

Calculating the Compound Zone Thickness Formation and thickness of the compound zone is influenced by: • • •

Time Temperature Gas composition

• •

Preheat temperature Pressure

A simplified Harris formula can be used as a rough guide for calculating the compound zone thickness: Compound zone thickness = √t × f

where t is process time at temperature and f is factor by temperature. This formula assumes a full ammonia atmosphere with no dilution gas, a dissociation rate of 30%, and a gas turnover rate of 5 to 1 of the nitriding chamber volume. If the nitrogen is added to the ammonia, thus potentially increasing the nitriding potential of the ammonia, or if hydrogen is added to the ammonia, then the nitriding potential is reduced and the gas turnover rate of 5 to 1 will increase according to the dilution volume increase.

Other Factors Affecting Surface Case Formation The steel being processed must be completely free of surface contaminants. This means that the work surfaces should be “clinically” clean—a condition commonly achieved by vapor degreasing the surface after hardening and tempering and prior to nitriding. Surface contaminants, which can cause formation of a nonuniform, or “spotty,” case, include: • • •

Cutting fluids Oils for surface protection Fingerprints

• •

Paint Decarburization

Cutting fluids can contain such compounds as chlorides and sulfides. When heat is applied to the steel at the start of the nitriding process,

Chapter 4: Microstructures of Nitrided Iron and Steel / 37

decomposition will occur, causing formation of a prohibitive surface barrier that prevents nitrogen diffusion. Oils should also be removed from the steel surface so that they do not decompose and leave a carbon residue. The carbon will form a barrier at its point of deposition on the surface, thus resisting the nitriding effect and resulting in a nonuniform nitrided case. Fingerprints are formed by body oils, which are hydrocarbon based. Once again, when heat initially is applied to the steel, the oil will decompose, leaving behind a carbon residue on the steel surface. The material handler should wear lint-free cotton gloves, or other gloves that will not react or leave a deposit on the steel surface. Paint residue or even marking ink will set up a nitride-resistant barrier on the steel surface. Once again, absolute surface cleanliness is mandatory. Decarburization is a surface problem. The presence of a decarburized layer will inhibit the uniform formation of a good nitrided surface. If decarburization has occurred from mill rolling and insufficient machine cleaning, or if the preheat treatment operation was carried out in a decarburizing atmosphere, then a nonuniform case will form. The result is usually visible in the form of an “orange peel” effect on the steel surface, or the case will exfoliate from the core steel. Hot-rolled steel should be machined with at least 10% of the bar diameter or 10% of the thickness removed. If the steel is in the form of rolled plate, it is imperative that the machining be uniform. Nonuniform machining will only contribute to the possibility of a partial decarburized layer on one side and will also induce a residual stress pattern that will relieve itself during the nitriding procedure. REFERENCES 1. D. Pye, Nitriding Techniques and Methods, Steel Heat Treatment Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker, 1997, p 721–764 2. M.A.J. Somers and E.J. Mittemeijer, Härt.-Tech. Mitt., 46: 375, 1991 3. T. Bell, B.J. Birch, V. Korotchenko, and S.P. Evans, “Controlled Nitriding in Ammonia-Hydrogen Mixtures,” presented as Heat Treatment 1973 (United Kingdom), The Metals Society 4. F.T. Hoffman and P. Mayr, Nitriding and Nitrocarburizing, Friction, Lubrication, and Wear Technology, Vol 18, ASM Handbook, ASM International, 1992, p 878–883 5. C.E. Floe, A Study of the Nitriding Process Effect of Ammonia Dissociation on Case Depth and Structure, ASM Conference Transactions, 1944

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p39-52 DOI: 10.1361/pnafn2003p039

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

5

Furnace Equipment and Control Systems THE NITRIDING FURNACE is simple in design. Other than the materials used, little has changed in furnace construction. Principal design changes involve the control of process parameters, with developments made in the use of computerized systems employing programmable logic controllers (PLC) for time, temperature, and gas flow control (Fig. 1). Good gas circulation is necessary to prevent process gas stagnation within the process chamber (also called the retort).

Fig. 1

Typical PC/PLC screen configuration for nitriding furnace control. Courtesy of Plateg USA, Meadville, PA

40 / Practical Nitriding and Ferritic Nitrocarburizing

Essential Furnace Design Criteria A well-designed gas nitriding furnace provides accurate, uniform control of process temperature, gas flow, and gas circulation. The furnace should be equipped with a good, simple mechanical handling system to maneuver the process chamber in or out of the furnace.

Accurate Temperature Control As the steel is likely to be prehardened and tempered, the tempering temperature would be higher than the nitriding process temperature. It is necessary to maintain temperature uniformity throughout the furnace heating chamber within at least ±5 °C (10 °F). That generally is accomplished by an internal air-circulating fan, although the circulation system can be external to the process chamber. Remember, nitriding cycles are usually lengthy, especially for deep case depths. Large process temperature fluctuations must not occur during the process, particularly any upward swing in temperature. A sudden temperature increase will lead to the possibility of reduced core hardness, as well as nitride networking (usually on corners) due to the higher solubility of nascent nitrogen at higher temperatures (Fig. 2).

Gas Circulation in the Process Chamber Good gas circulation is necessary to prevent gas stagnation and to maintain an approximate 5:1 to 6:1 gas exchange ratio. This, of course, will depend on the gas dissociation required of the process. Two methods of gas circulation can be employed: • •

(a)

Overpressure circulation Internal fan circulation

(b)

Fig. 2 Micrographs of white nitride layers developed on vacuum-melted AMS 6470 steel. (a) White layer 0.033 mm (0.0013 in.) thick formed after single-stage nitriding at 525 °C (975 °F) for 60 h with 28% dissociation. Buildup of white layer at corner was 0.084 mm (0.0033 in.). (b) White layer 0.020 mm (0.0008 in.) thick formed after double-stage nitriding at 525 °C (975 °F) for 9.5 h with 25 to 28% dissociation, then at 550 °C (1025 °F) for 50.5 h and 80 to 84% dissociation. Buildup of white layer at corner was 0.050 mm (0.00020 in.). 2% nital etch; 150×. Source: Ref 1

Chapter 5: Furnace Equipment and Control Systems / 41

Overpressure circulation simply means ammonia gas flows in and out due to the pressure difference. This process relies on gas delivery pressure and, most importantly, on part geometry. Complex geometries can cause gas stagnation, leading to the possibility of nonuniform case formation and sometimes even no case formation. Examples of complex geometries include bushes, sleeves, and cavities. Good gas circulation is imperative to prevent the possibility of stagnation in blind holes, cavities, and shielded areas within the process chamber (i.e., areas where one part is shielding another part). If gas stagnation occurs and nascent nitrogen is not replenished within the hole or cavity, then a shallow nitrided case may form. The overpressure method of gas circulation requires careful loading of components. Internal fan circulation is unquestionably the best method for preventing gas stagnation (Fig. 3). Gas circulation by a fan nearly guarantees complete accessibility of the process gas throughout the process chamber. Another benefit of the internal fan system is good temperature uniformity. Uniform temperatures are required in any heat treatment process, especially those involving aerospace and aircraft components. Otherwise, a nonuniform case will develop, thus setting up the probability of a stress pattern in the nitrided component. Internal fan circulation is not necessarily restricted to the bell-type furnace shown in Fig. 3; a horizontal furnace or even a pit-type furnace design can be used.

Heating bell

Retort Cooling bell Work basket

Exhaust fan

Heating elements

Work support

Ammonia supply

Exhaust Circulating fan Oil seal Bell-type furnace with heating bell

Fig. 3

Bell-type furnace with cooling bell

Schematic of bell-type furnace containing an internal fan for gas circulation

42 / Practical Nitriding and Ferritic Nitrocarburizing

Types of Nitriding Furnaces Any type of furnace can be used for the nitriding process, provided that it has the integrity to maintain an atmosphere without leaking and meet the required process control conditions. This means that even if a furnace is not gastight, it can still be used with an inner-sealed retort or container. Nitriding furnace designs include: • • • • •

Box temper furnace Vertical air-circulating furnace Lift-off bell furnace Fluidized-bed furnace Salt bath furnace

The only restriction is that the furnace should have a method of isolating the process gas from the shop environment. This is accomplished by processing the work in a sealed (gastight) furnace or in a sealed process retort or chamber.

Insulation A furnace that is repeatedly heated up and down through a temperature range both uses and loses energy. As much of the energy (heat) as possible needs to be contained within the process chamber, and external heat losses need to be kept to a safe working minimum temperature. Therefore, the furnace must be insulated. Insulation for the reduction of furnace heat losses can take the form of: • • •

Insulating refractory brick Low thermal mass ceramic fiber insulating material Vacuum insulation

Some of the major advantages of a nitriding furnace with regard to insulation and heating elements are discussed below. There is no ideal insulation material. The type of insulation depends on whether a process retort is used and how the retort goes into the furnace. Poor mechanical handling may damage the heating system and the insulation materials. Relatively Low Temperatures. Because nitriding furnaces usually are not subjected to high temperature levels, wear and tear on the insulation and furnace elements is very low. The heating elements are more often than not of a higher watt density than is necessary, thus ensuring long life. If the furnace is gas fired, the radiant tubes are usually overdesigned for the application. Long Useful Life. The furnace is not subjected to adverse temperature fluctuations, thus prolonging the life of the heating system and insulation material. The author personally knows of a used vertical air-circulating

Chapter 5: Furnace Equipment and Control Systems / 43

furnace that in 1963 was converted into a nitriding furnace, with new refractory brick and heating elements added. The furnace is still in operation 40 years later, with no further element or refractory replacement. If the refractory lining and heating elements are subjected to mechanical damage, of course, then their life expectancy will be reduced. However, unless a nitriding furnace is subjected to poor mechanical handling, it requires low maintenance. Energy Costs. The quality and thickness of the insulation will determine the energy cost of the cycle. The insulation design should take into account heat losses to the outer furnace casing and the insulation thickness and quality. Higher quality insulation may have a higher initial cost, but that cost will be recovered in lower operating costs.

Determining Appropriate Furnace Design Furnace configuration and design is influenced by: • •

• •

Part design: Is the part a tooling item, dies for pressure die casting, or production work? Part loading: How large are the parts to be treated, and what is the maximum load that is to be charged into the process chamber? This is important, as it will determine the heat/energy requirements for the furnace, chamber, and load. Frequency of operation: Will the furnace be operated on a jobbing basis or will there be guaranteed loads with regular frequency of loading? Furnace availability: Will the furnace be used for other operations such as tempering, subcritical annealing, or stress relieving?

Retort Construction Steels. Heat-resistant type 309 or 310 stainless steel is quite suitable as a retort material. The most satisfactory material is Inconel; however, it is expensive. The primary requirement is that the material must not catalyze ammonia dissociation at the internal surfaces of the process chamber. Materials that fall into this category are mild or plain carbon steel, and types 302 and 304 stainless steel. If used for the process chamber, mild steels and plain carbon steels will draw away the nitriding reaction from the workpiece surface to the point that minimal nitriding, if any, will take place. They also lack the necessary mechanical strength at the process operating temperature, quickly distorting and failing prematurely. Types 302 and 304 stainless steel also lack mechanical strength and will distort, as well as deteriorate rapidly on the inside due to corrosion resulting from the breakdown of surface chrome oxide caused by the nitriding reaction. Temperature-resistant glass, in theory, is probably the best retort construction material because of its nonreactive nature with ammonia gas.

44 / Practical Nitriding and Ferritic Nitrocarburizing

However, its fragility would result in mechanical damage, making glass impractical for use in a production nitriding environment. Insulating refractory firebrick can be used to line the process chamber, provided that it is dense and heat resistant, with low porosity and low iron content. The mortared joints must be very thin and contain minimal porosity; otherwise, the ammonia process gas will begin to leak. If the iron content of the brick is high, the iron will react with the ammonia to form iron nitrides and begin to erode from the joint, particularly if there is a fan circulation system in the process chamber. Enameling. The container can be enameled but, like temperatureresistant glass, is subject to chipping by mechanical damage and is not a practical solution for the nitriding process chamber. Deoxidized copper can be used, but lacks strength at high temperatures. The copper will not take part in the process reaction, being completely impervious to nascent nitrogen or anhydrous ammonia. Copper is used as a masking medium to prevent nitriding from taking place on selected portions of steel parts.

Retort Maintenance Even if stainless steel or Inconel is used, the retort will not be maintenance free. No matter what the material of construction, the internal retort walls require a periodic program of regeneration. Regeneration of the retort is necessary when gas dissociation becomes difficult to control and when workpieces exhibit inconsistent metallurgical results. This indicates that the free nitrogen is reacting with the internal walls of the process chamber, due to deterioration of their chrome oxide surfaces. Regneration is simply accomplished by heating the empty process chamber to about 900 °C (1650 °F) and holding at that temperature for a few hours. This procedure will “burn” out the nitrogen and cause the chrome to oxidize, thus forming chrome oxide. After the process chamber cools, the surfaces are cleaned by either shot-blasting or sandblasting. The frequency of regeneration correlates with retort usage and will vary from plant to plant. Once every 3 to 5 years is typical. Trays and fixtures require the same considerations as the process retort in terms of construction materials and maintenance. Each time the nitride operation is run, the trays and fixtures are exposed to the atmosphere and the process temperature. Therefore, a planned maintenance procedure must be in place in terms of crack repair to support webs on the trays.

Sealing the Retort to Prevent Ammonia Leaks Figure 4 shows a nitriding installation that uses a simple seal that fits on the retort flange. Mild steel or stainless steel studs, nuts, and washers secure the sealed top cover. If mild steel fasteners are used, dip the thread

Chapter 5: Furnace Equipment and Control Systems / 45

Dissociator

Dissociation Pipette Seal

Seal

Ammonia Tank

Work Basket Exhaust

Heating Elements Nitriding Retort

Fan Gas Inlet Gas Outlet

Fan Motor

Nitriding Furnace

Fig. 4

Nitriding process retort with a seal that fits on the retort flange

into a mixture of oil and flake graphite or molybdenum disulfide to ensure easy removal after processing. Otherwise, the threads of both the stud and nut will seize solid by the end of the process cycle. Other sealing methods include oil seals (commonly used on the lift-off bell furnace configuration), clamp arrangements, and vacuum seals. Great care must be taken to ensure that the process chamber is gastight. Checking and Identifying Seal Leaks. The new seal will usually leak slightly on the first cycle without serious consequences. After that, the seal has “bedded” in. Ammonia leaks can be identified by using a lighted sulfur pipe cleaner (such as those used by a pipe smoker) that has been dipped into molten sulfur. Once lit and manually moved around the flange perimeter, a cloud

46 / Practical Nitriding and Ferritic Nitrocarburizing

of dense white ammonium sulfide smoke will appear at any leak source. Tighten the appropriate studs or ensure that the clamps are secure, and check the integrity of the seal once again.

Safety Precautions When Using Ammonia Ammonia is an extremely pungent gas, which can make breathing impossible if large volumes of ammonia escape. Even if small volumes of ammonia gas leak from the process chamber, people near the leak source may experience loss of breath and extremely painful eye irritation. It is imperative to read the Material Safety Data Sheet or International Chemical Safety Card when using ammonia gas. Ammonia gas is also very flammable. If small leaks occur in the seal of an internal process chamber, the ammonia can ignite and burn vigorously especially in the presence of oxygen. If part of the sealing method is made from aluminum, then there is a strong possibility that the aluminum portion of the seal will melt, thus increasing the fire risk. Large concentrations of anhydrous ammonia can cause serious injury, even death, due to violent explosions if the leaking ammonia is exposed to a naked flame. Great care must be taken to eliminate the presence of oxygen and an ignition source. Other important safety precautions include (Ref 2): •

•

•

• •

• •

•

Do not allow ammonia to come into contact with mercury; under some conditions it can produce both a toxic and violently explosive gas mixture. Ammonia gas that may have leaked from the process retort or gas storage bottle can be violently explosive and flammable. Ensure adequate air ventilation around the nitriding furnace. Keep a rebreathing apparatus system in the process vicinity. If anhydrous ammonia leaks it is violently pungent and can only be approached while wearing a rebreathing apparatus. Avoid the ammonia gas coming into contact with the skin or sores, as this will be very painful. Fit an afterburner system or scrubber or both on the exhaust line so that the ammonia is decomposed before being dispersed into the atmosphere. Ensure that everyone concerned with the nitriding process knows where the main safety shutoff valve is located. Ensure that the safety shutoff valve is in position and can be turned with relative ease. If there are intermediate valves on the line and it is a key valve, make sure that the key is chained to the shutoff valve. If the process uses only the dissociation burette measuring system for process gas control, check city regulations regarding the disposal of water contaminated by ammonia. Do not put the ammonia water down a city drain.

Chapter 5: Furnace Equipment and Control Systems / 47

•

• • •

•

• •

If a leak does occur on an exposed delivery pipe, remember that water will absorb 70 times its own volume of ammonia. In other words, ammonia is very soluble in water. Simply wrap soaking wet towels or rags over the pipe to absorb the ammonia until professional help is available. Ammonia is also a very dense gas and will settle to the lowest point. The area can simply be sprayed with a water hose; however, advise city authorities if the ammonia water goes into a drain. If a bad ammonia leak occurs, evacuate the area immediately, advise the appropriate authorities, and then ventilate the area thoroughly. Clearly mark all exit paths and exit doors. Be sure that all personnel understand completely how to evacuate the affected work area. A safety shower and eyewash station should be close to the furnace operating area. If a person comes in contact with large volumes of ammonia gas, it is necessary to flood the body with an excess amount of water. If no safety shower is available, then wash the affected part with large volumes of water from the nearest source. The same applies to eye contact with ammonia; the eyes must be flushed with clean, uncontaminated water. If no eyewash station is available, then use a squeeze bottle filled with water to flush the eyes. Make sure that all personnel know how to contact both a responsible person and, if necessary, a physician. Maintain up-to-date telephone numbers and post them in an accessible and convenient place. Do not, under any circumstances, try to neutralize the ammonia leak or spill with an acid. A violent reaction can occur. Most importantly, read the Material Safety Data sheets. It is in your own interest and that of everyone else who works with you to be familiar with the safety aspects of handling ammonia gas.

Caution Regarding Safety When Using Ammonia. This section is intended only to alert the reader to the dangers of ammonia. It does not claim to address all possible safety concerns. The user, in conjunction with equipment and material suppliers and appropriate government agencies, must develop, institute, and maintain health and safety practices sufficient and in compliance with regulatory requirements.

Furnace Heating A furnace can be heated electrically or by natural gas. The choice of heating medium depends on: • • •

Heating medium availability, efficiency, and cost Maintenance costs Replacement material costs

48 / Practical Nitriding and Ferritic Nitrocarburizing

Electric heating elements usually are made of a nickel-chromium (Nichrome) material and are not normally exposed to the ammonia atmosphere. If ammonia exposure occurs, the element will corrode rapidly. Electric heating elements usually are encapsulated or located in the furnace that holds the process retort. If the furnace is heated by natural gas, the burners usually are located in a remote heater box and the heated combustion air is distributed to the process retort. Alternatively, the combustion products can heat a series of external radiant tubes that surround the process retort. Calculation of the heat input requirement uses this formula (Ref 3):

kW·h =

Steel specific heat × Gross load mass × Temperature rise 3412

where the specific heat of steel is usually considered to be 0.15 Btu/lb·°F; the load mass (in pounds) equals the total load, including retort fixturing; the temperature rise equals the required temperature minus the ambient temperature (in °F); and 3412 is the conversion factor from Btu’s (British thermal units) to kW·h. Note that the calculated figure is defined as “that energy required to heat the complete load up from ambient temperature to the process operating temperature in an hour.” The calculated figure can then be further divided by the number of hours in which it is both practical and economical to raise the furnace temperature. Remember, the calculation must be for the gross load going into the furnace. Normally the furnace designer will calculate the heat input calculation based on the maximum furnace operating temperature. This will account for the total weight to be heated, including: • • •

Workload weight Total weight of the process chamber Total weight of the fixtures and work support materials

Bear in mind that the calculation is for the energy input only and does not account in any way for heat losses through the furnace insulation or through apertures such as seals, thermocouple holes, fan-drive shaft holes, or any other aperture. The calculation for gas heating energy requirements for gross workload is (Ref 4): Btu’s = Specific heat × Load mass × Temperature

Once again, the calculated value refers only to the energy required to heat the furnace in 1 h. To heat in 1 h may be both costly and impractical, mak-

Chapter 5: Furnace Equipment and Control Systems / 49

ing it necessary to divide the resulting value by whatever time factor is convenient. The calculated value does not account for furnace heat losses through refractories or apertures cut in the furnace. The “add on” kW or Btu’s will depend on the refractory value of the furnace insulation material and total surface area of the furnace and apertures.

Process Control and Instrumentation Good process control and management improves: • • • • • •

Process repeatability and economics Metallurgical requirements Operator interface Recordkeeping Tracking of trends Utilization of process service requirements

Some factors, such as temperature control, can use conventional instrumentation such as thermocouples and a time-temperature recorder. Advanced control instrumentation uses a computer for process control and a dual PC/PLC system (see Fig. 1). This system arrangement must conform to ISO 9000 and its derivatives for the requirement of: • • •

Data acquisition, logging, and storage Process control, including time, temperature, gas flow, and dilution gas if applicable Historical trends, maintenance, and so on

It is interesting to note that the hardware for the PC/PLC combination is usually not as expensive as for more conventional control methods— except, of course, for the cost of programming. The level of programming depends on the particular set of process considerations.

Temperature Control Temperature control is best achieved with a thermocouple attached to the workpiece, or placed as close to the workpiece as possible. After determination of the number of control zones, the furnace will require at least two thermocouples per zone: a set-point temperature thermocouple and an over-temperature thermocouple. The over-temperature thermocouple is necessary to prevent temperature overshoots and furnace temperature “runaways.” If the over-temperature thermocouple is removable, always remember to reload it. Fitting a Load Thermocouple into the Retort. A hole is cut into the retort and a pipe with a single gastight sealed end is welded into place in the box. The thermocouple is inserted into the open end of the tube. The

50 / Practical Nitriding and Ferritic Nitrocarburizing

tube should extend as far as necessary into the retort to show the best temperature control point during the process cycle. The thermocouple should be calibrated for the control instrument electromagnetic field (EMF) requirement.

Monitoring Gas Dissociation Gas dissociation is simply the amount of decomposition of the ammonia gas (Ref 5). During the nitriding process, ammonia gas is continually decomposing due to the catalytic effect of the steel and the retort. The gas breaks down into its elemental form of hydrogen and nitrogen, and thus the exhaust gases consist of ammonia and hydrogen: (2NH3 + H2 + N2)

Of the above three gases, only ammonia is soluble in water, to the point that water will absorb as much as 70 times its own volume of ammonia (Ref 6). The simple method of measuring the dissociation is to use a pipette (burette) as shown in Fig. 5. To make a measurement, the ammonia gas in the nitriding box is first admitted into A by opening taps C and D. After the air has been expelled, taps C and D are closed. During the measurement, tap E is opened and the water immediately absorbs the undissociated ammonia. The water takes up precisely the volume previously occupied by the ammonia, but the remaining N2-H2 gas (dissociated ammonia) does not dissove in water. This is a simple and effective method of measuring the gas dissociation. Other methods include using an oxygen-measuring unit and recalibrating to feed in percent dissociation of ammonia or hydrogen. The rate of dissociation at which the process should operate is usually around 30%. This, of course, will depend on whether a single-stage or

Water

B E

0 25

NH3 + N2 + H2

N2 + H2

50 75

25 N2 + H2

50 H2O

50

75

75 H2O

A

100 25%

D

0

25

C H2O

Fig. 5

0

N2 + H2

100 50%

100 75%

Height of water column in graduated chamber (A) for 25%, 50%, and 75% degree of dissociation

Dissociation pipette (burette) schematic. See text for discussion.

Chapter 5: Furnace Equipment and Control Systems / 51

two-stage process is being conducted. For the two-stage (Floe) process, the ammonia gas dissociation is at a lower value, generally (but not always) around 15%.

Oxygen Probe Control Many advances have been made in the use of the oxygen probe control in the nitriding process. L. Sproge of Sweden and S. Midea of the United States, both of the Aga Gas Company, reported their computer control model in a paper that discusses the equilibrium of hydrogen and oxygen with water vapor (Ref 7). In order for the control system to be effective, the flow of the process gas components must be accurately metered using mass flow controllers. These units will meter very precise gas flows to the process.

Nitriding Sensors Development of the nitriding sensor has significantly contributed to nitriding process control (Ref 8, 9). This system has been designed using the principle of a solid-state electrolyte, measuring variations in the magnetic and electrical properties of a steel as temperature increases and surface chemistry changes. The sensor is placed in the furnace processing chamber and then calibrated according to the desired nitriding result, the parameters of the parts being treated, and the surface area of the workpieces. The control system allows the technician to control the thickness of the compound layer and case depth. If water vapor or oxygen is present within the nitriding atmosphere, the oxygen partial pressure can be used as an indirect measure of the nitriding potential of the process atmosphere. The resulting generated signal from the sensor indicates the partial pressure of the oxygen and the degree of dissociation occurring. Initial part cleanliness is essential, because any surface contamination will affect the sensor results. REFERENCES 1. Gas Nitriding, Heat Treating, Cleaning, and Finishing, Vol 2, 8th ed., Metals Handbook, American Society for Metals, 1964, p 149–163 2. “Safety Requirements for the Storage and Handling of Anhydrous Ammonia,” ANSI K61.1, American National Standards Institute 3. G. Totten, G.R. Garsombke, D. Pye, and R.W. Reynoldson, Heat Treatment Equipment, Steel Heat Treatment Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker, 1997, p 293–482 4. D. Pye, “Understanding Nitriding and Ferritic Nitro Carburizing,” course notes, Pye Metallurgical Consulting 5. J. Grosch, Heat Treatment with Gaseous Atmospheres, Steel Heat Treatment Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker, 1997, p 663–720

52 / Practical Nitriding and Ferritic Nitrocarburizing

6. V.O. Homerberg and J.O. Walstead, A Study of the Nitriding Process: Part 1, ASST Nitriding Symposium, American Society for Steel Treaters, 1929 7. L. Sproge and S. Midea, Analysis and Control of Nitriding and Nitrocarburizing Atmospheres, Carburizing and Nitriding with Atmospheres, ASM International, 1995, p 303–307 8. B. Edenhofer and J.W. Bouwman, Vacuum Heat Treatment, Steel Heat Treatment Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker, 1997, p 483–526 9. Marathon Sensors Inc., Cincinnati, OH, www.marathonsensors.com

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p53-64 DOI: 10.1361/pnafn2003p053

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

6 Salt Bath Nitriding

AN ALTERNATIVE to the gas nitriding process was sought in the mid1930s that would produce a more uniform and better metallurgically formed case. Notable chemical companies conducted many experiments to find an appropriate alternative method. A liquid, researchers thought, would fulfill the uniformity requirement through surface contact of the liquid to the steel. The depth and quality of case would be determined by the chemical composition of the liquid. A heat source would be necessary to drive the nitrogen into the steel surface. A cyanide-based salt, it was discovered, fulfilled the requirements of the process, and salt bath nitriding began. Salt bath nitriding is in essence the same process as gas nitriding; only the medium is different. Salt bath nitriding offers extremely good case depth uniformity. Salt bath nitriding utilizes the melting of a salt containing a rich nitrogen source. When heat is applied from either an internal or external source, the salt melts and liberates nitrogen to the steel for diffusion. When a steel workpiece is introduced into the salt and heated up to a temperature in the molten salt, similar reactions begin to occur as in gas nitriding; that is, controlled amounts of nitrogen are released to diffuse into the steel surface.

Salts Used and Process Advantages Nitriding Salts. Early on, a cyanide-based salt, such as NaCN, was used with other salts. A typical nitriding salt mixture would be (Ref 1): • •

NaCN, 30% Na2CO3 or K2CO3, 25%

• •

KCl, 39% Moisture, 2%

A proprietary nitriding salt composition would be (Ref 1): • •

NaCN, 60% K2CO3, 15%

• •

KCl, 24% Moisture, 1%

54 / Practical Nitriding and Ferritic Nitrocarburizing

Advantages. The salt bath nitriding technique gained popularity in the early 1950s because it required a lower capital investment than gas nitriding. Operating costs also were considerably lower, making it a costeffective method of nitriding steel. An added benefit was that salt bath nitriding was carried out at a slightly higher temperature than gas nitriding: 565 to 585 °C (1050 to 1085°F). This meant faster processing time. To summarize, the advantages of using salt bath nitriding equipment were: • • • • • •

Relatively low operating cost Low maintenance Easy operation, requiring a lower skill level Small batch-type furnace generally used, occupying less floor space Easy startup, easy shutdown Slightly faster diffusion cycle times

Types of Salt Bath Nitriding Processes Since the early development of the salt bath nitriding process, many derivatives have appeared (Ref 1). The two major European companies that pioneered the salt bath process were the Cassel Division of Imperial Chemical Industries (ICI) in England and the Durfferit Division of Degussa in Germany. The early process was developed by ICI, which was eventually acquired by Degussa in the late 1970s. The salt, known by ICI as N.S.450, relied on the decomposition of cyanide to cyanate. In Germany, Degussa developed the two-component process of forming both nitrides and some carbides in the steel surface (i.e., nitrocarburizing). The German process was introduced as the Tufftride process. Tufftride became popular in Europe in the early 1950s and was introduced to the United States in the early 1960s. It was quickly accepted as a reliable procedure and identified by the Society of Automotive Engineers (SAE) designation of AMS 2755. The process simply involves: • • • • •

Preheating Immersion into the molten salt bath Cooling in still air or a suitable quench medium Postwashing Optional polishing or oiling

The surface finish produced by the process is usually matte black, which is corrosion resistant. Because of emerging environmental concerns, Degussa pioneered the use of a low-cyanide salt. Kolene Corporation developed a low-cyanide

Chapter 6: Salt Bath Nitriding / 55

salt process, now known as the Melonite process (Ref 2), which offered an alternative to the Tufftride process. The process uses a cyanide-free salt, the composition of which is specified under the SAE designation AMS 2753. Another derivative of the Melonite process by Degussa is known as the QPQ (“Quench-Polish-Quench”) process (QPQ is a trademark of the Kolene Corp.). Both the Melonite process and the QPQ process require the part to be quenched into a molten oxidizing bath to neutralize any residual cyanide that might be present. After the quench procedure, the part is mechanically polished, followed by a further resurface oxidation process.

Salt Bath Nitriding Equipment and Procedure Salt bath nitriding can be conducted in either a batch system or a continuous system. Either equipment system can be electrically or gas heated (Fig. 1). The process time is generally around 60 to 90 min at temperature (part temperature), followed by a quench, then into the oxidizing bath operating at a process temperature of 400 °C (750 °F), followed by cooling with a water quench. At this point the process may be considered complete, or the part can be mechanically polished. After polishing, the part can be immersed in another oxidizing bath to complete the procedure. The metallurgical results are the same as with the conventional gas nitriding process: A compound layer is formed at the steel surface with a diffusion zone area immediately below the compound layer. The surface compound layer thickness will be determined by the steel composition. Usually case depths are confined, but not limited, to shallow depths of approximately 0.13 mm (0.005 in.). Analysis of the compound layer on plain carbon steel showed it to contain a nitrogen concentration of around 6%, with very small amounts of carbon. If the procedure used is the Melonite process, which includes a post-oxidize treatment, then a shallow surface oxide layer will be seen microscopically on the immediate surface of the part (Fig. 2, 3).

Using a New Salt Bath A newly made-up nitriding salt bath should not be used immediately on completion of the melt procedure. The bath first must be “aged” or “conditioned.” If the newly molten salt bath is used as soon as the salt is molten, the salt will immediately begin a vigorous surface reaction on the steel, attacking the surface and causing a form of pitting known as orange peel. This means that the bath is vigorously decarburizing the surface of the steel.

56 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Principal furnace types for liquid salt bath nitriding. (a) and (b) Externally heated. (c) and (d) Internally heated, with immersed alloy electrodes and metal liner or submerged electrodes with ceramic tile lining

The bath must be raised to operating temperature and “aged” by oxidizing its cyanide content through decomposition of the cyanide to cyanate. The aging period depends on the surface area of the salt bath. For example, a bath that is 750 mm (30 in.) in diameter by 750 mm (30 in.) deep will take approximately 12 to 16 h to age. It is important to note that no work should be processed through the bath during the aging process.

Chapter 6: Salt Bath Nitriding / 57

Fig. 2

Ferritic nodular iron, salt bath nitrided 90 min at 580 °C (1075 °F), oxidizing molten salt quenched. 500×, nital etch. Courtesy of Kolene Corp.

Fig. 3

SAE 5115 (UNS G51150), chromium-manganese low-carbon steel, salt bath nitrided 90 min at 580 °C (1075 °F), oxidizing molten salt quenched. 500×, nital etch. Courtesy of Kolene Corp.

As with gas nitriding, workpieces must be cleaned prior to salt bath nitriding to prevent any oil, grease, paint, and so forth from contaminating the bath. In addition, the steel surfaces must be free of oxides or other contaminants. A cyanate decomposition percentage up to 25% normally will produce acceptable surface composition and hardness results. If the carbonate content is high, the temperature of the bath should be lowered to around 440

58 / Practical Nitriding and Ferritic Nitrocarburizing

to 455 °C (825 to 850 °F). This will cause precipitation of the carbonate, which can then be scooped out as sludge from the bottom of the bath. The salts do not last indefinitely. If used on a daily basis, the bath should be discarded once every 3 to 31/2 months and replaced with a new bath. Again, care must be taken to age the new bath.

Bath Replacement The bath is ready for replacement if it is becoming difficult to maintain a cyanate level around 25% or if signs of corrosion are evident on the steel surface. Be careful not to confuse the corrosion with that caused by postwashing of the steel. The cyanide to cyanate control level is not the same for all steels processed through the nitriding bath and will vary for various steels. Sodium cyanide (NaCN) levels are normally maintained as follows: • • •

High-speed steels: 15% minimum NaCN Hot-work tool steel: 20% minimum NaCN Alloy steels: Approximately 25% NaCN

Bath Testing and Analysis The bath should be sampled, and the sample should be titrated against a standard solution of silver nitrate (AgNO3) (Ref 3). The test methods described in the following paragraphs are simple titration methods of analysis and not complex. It is essential, however, before testing that the equipment used in the sampling be thoroughly clean and that the salt sample taken from the bath also be clean and free of contaminants such as sludge or graphite, which could be on the molten salt surface.

Materials and Equipment The following is a list of materials and equipment required for titration testing of the nitriding salt bath: • • • • • • • •

One burette calibrated to read directly in percentage of NaCN. Each graduation on the burette is equal to 0.5 mL (0.5 cm3) One sample spoon designed to hold approximately 1 g of powdered sample One mortar and pestle for crushing the sample One bottle of chemical reagent (0.2 N AgNO3) Two 250 mL glass beakers and two glass stirring rods One wide-mouth 500 mL bottle One once-dropping bottle of indicating solution (10% KI) One vial of lead carbonate powder (2PbCO3·Pb (OH)2)

Chapter 6: Salt Bath Nitriding / 59

It is advisable to maintain a shift logbook of titrations of the nitriding bath to observe any anomalies that might be taking place.

Analysis Procedure The procedure for performing the analysis consists of eight steps: 1. Sampling the liquid nitriding bath: Plunge a clean, dry steel rod into the molten operating bath and remove immediately. A film of salt will freeze and adhere to the rod. Scrape off and collect about 10 g in a clean mortar. Crush to a fine powder. 2. Measuring sample: Measure out 1 g. Transfer the sample from the spoon to the 250 mL glass beaker and fill to the halfway mark with distilled water, preferably warm. 3. Dissolving sample: It is preferable to bring the solution to a gentle boil; however, it can be dissolved by stirring. Add a small pinch of lead carbonate (1/4 to 1/2 g). Disregard the small amount of black residue. Allow the solution to settle or precipitate the lead sulfide (PbS). 4. Decanting (satisfactory for ordinary shop practice): Permit the solution to settle for 5 to 10 min until all solids precipitate to the bottom of the beaker. Decant the clear liquid into the second beaker and proceed with titration. 5. Filtration (for maximum accuracy): Filter through 11 cm No. 2 Whatman filter paper or the equivalent. To the filtrated solution, add 3 or 4 drops of indicating solution (10% KI). 6. Operation of the burette: By pressing the bulb, you automatically fill the burette. The burette will always fill to the zero mark and any excess will return to the bottle. The tube should always be filled before each test. Pressing the inch clamp will release the chemical reagent through the glass delivery tip into the beaker. 7. Titration: Titrate by slowly adding the chemical reagent solution from the burette until the contents of the beaker turn cloudy. At first, a slight discoloration or cloudiness may form. Ignore this. Continue titrating until the solution is entirely cloudy and opalescent (resembling the color of yellow lemonade). This is the “end point.” 8. Determining % NaCN in the bath (1 g sample): The reading on the graduated burette is the percentage of NaCN in the salt bath tested.

Important Factors for Successful Results Wash all grease and oils from parts to be nitrided. Oils may contain sulfur, which is detrimental and will reduce efficiency. If work is washed in an alkaline cleaner containing sodium hydroxide or similar alkali, the work should be rinsed in hot distilled water and dried before nitriding.

60 / Practical Nitriding and Ferritic Nitrocarburizing

Determination of Sodium Carbonate and Sodium Cyanate Sodium Carbonate. The determination of sodium carbonate in the bath consists of the following steps: • • • •

•

Accurately weigh a 0.5 g sample of the bath, transfer to a 300 mL Erlenmeyer flask, and add 100 mL of distilled water. After the sample has dissolved, add 25 mL of a neutral 19% barium chloride solution. Stop flask tightly, and let stand for 2 h. Filter and wash the precipitate with neutral 1% barium chloride until the washings give no color with phenolphthalein. Place the filter paper and precipitate in the original flask. Add 50 mL of distilled water and a measured excess of standard N/10 (0.1 N) hydrochloric acid (HCl). Boil the solution for 5 min and allow to cool. Add 4 to 5 drops methyl red indicator and titrate with standard N/10 (0.1 N) sodium hydroxide (NaOH).

The following formula can then be used to determine the sodium carbonate percentage:

%NA2 CO3 =

(VHCl · normality HCl – VNaOH · normality NaOH) · 5.3 sample weight

where VHCl and VNaOH are the volumes in mL and sample weight is measured in grams. The normality of a solution is the number of gram equivalent weights of solute per liter. Sodium Cyanate. The procedure for determination of sodium cyanate in the bath consists of the following steps: •

•

• •

Accurately weigh an approximate 1 to 2 g sample of the bath and transfer to a 650 mL Kjeldahl flask. Use a 1 g sample when checking a nitriding bath and a 2 g sample when checking a carburizing bath. Add 100 mL ammonia-free distilled water, 2 or 3 selenized Hengar granules, and 5 mL of concentrated sulfuric acid. Digest under the hood or in a Kjeldahl digesting unit until the volume reaches approximately 25 mL. Allow to cool. This digestion drives off all the cyanide nitrogen as hydrogen cyanide and fixes the cyanate nitrogen as ammonium bisulfate or ammonium sulfate. Add 350 mL of ammonia-free distilled water and 50 mL of ammoniafree 72% sodium hydroxide to the digested sample in the Kjeldahl flask. Immediately connect to a Kjeldahl distilling unit, being sure to include a Kjeldahl trap before entering the condenser. Place the tip of the delivery tube into a 500 mL Erlenmeyer flask containing 50 mL of saturated boric acid solution with methyl red indicator added. The tip of the delivery tube must be below the surface of the boric acid solution.

Chapter 6: Salt Bath Nitriding / 61

• •

Distill off 150 to 200 mL of liquid, disconnect, and wash down the condenser with ammonia-free distilled water. Titrate the distillate with standard N/10 (0.1 N ) hydrochloric acid.

The following formula can then be used to determine the sodium cyanate percentage:

% NaCNO =

VHCl · normality HCl · 6.5 sample weight

where VHCl is the volume in mL and sample weight is measured in grams. The normality of a solution is the number of gram equivalent weights of solute per liter. Note: To avoid running appreciable blanks, reagent-grade chemicals and ammonia-free water should be used for all solutions.

Bath Maintenance Maintenance of the salt bath and related equipment is critical and quite simple. The procedures can be broken down into daily, weekly, and monthly activities (Ref 1): Daily maintenance consists of: • • •

Analyzing the bath for cyanide and cyanate content at the start of each shift Checking that the temperature control instrument is in proper working order Checking the cleanliness of the bath by using a perforated scoop and desludging the bath before the start of the shift Weekly maintenance consists of:

• •

Checking the cleanliness of the bath around the top (that is, the part above salt level) Removing the salt pot and checking the integrity of the external surfaces, especially if the salt bath is gas fired. Watch for signs of excess scale and “ballooning” at the bottom of the pot, which means that the pot wall is thinning and the salt is too heavy for the pot at that wall thickness. Monthly maintenance consists of:

•

Checking the gas burner train if the nitride bath is gas fired. Make sure that the burner train linkages are secure and that modulating valves (if fitted) are operational.

62 / Practical Nitriding and Ferritic Nitrocarburizing

• • • •

•

Determining that the burner ignition system is operational and that the flame rod is clean Checking that no traces of nitriding salt are present on the cleanout port Checking that the elements are fully immersed if the furnace is electrically heated using an immersed heating element Checking that the elements show no signs of deterioration or indications of hot spots if using an externally heated salt bath. Deterioration of the elements is easy to see. Checking that the ammeter phases are in balance. If there is an imbalance, the elements should be checked for uniform wear or potential breakages.

Operating the Salt Bath As discussed, furnace operating personnel must ensure that the bath is fully aged, regenerated, and of the correct salt analysis. Once these parameters have been established, the bath is ready for operation. As with any salt bath procedure, the steel workpiece must be thoroughly preheated and free of any surface moisture—either on the steel, the support fixture, or the fastening wire—before immersing it into the molten salt. On immersion, the workpiece will freeze the immediate salt surrounding it and cause the salt to solidify on the steel surface. The “cocoon” of salt surrounding the steel acts as an insulator to shield the steel from excessive heat. This will reduce the risk of distortion through thermal shock. Due to its nature of being molten when under heat, the salt will slowly oxidize at the salt/air interface and cause a breakdown of the cyanide to carbonate. This will slowly begin to affect the internal surfaces of the pot, thus reducing pot life.

Safety Precautions Safety precautions are as important with low-temperature salts as with high-temperature salts (Ref 3). Some of the more important considerations are: • • • • •

Always preheat to both reduce thermal shock and remove moisture from the steel part surface. Do not mix nitrate salts with cyanide-based salts, as this poses a serious risk of violent explosion from the salt bath. Always wear the appropriate safety clothing, such as safety shoes, safety gloves that cover the forearm, a face shield, and an apron. Make sure that the ventilation of the bath is operational. Have a cyanide antidote kit available if cyanide salts are ingested orally. These salts are highly poisonous.

Chapter 6: Salt Bath Nitriding / 63

• • •

Make sure that secure storage is available for the storage of cyanidebased salts. Do not allow unauthorized persons access to cyanide-based salts. Keep a logbook of salt usage.

Design Parameters for Furnace Equipment The furnace equipment for salt bath heat treatment is simple in design. The same parameters apply to the design of the salt bath process as to gas nitriding. However, the salt bath furnace tends to be a more compact unit that occupies less floor space than its gas nitriding counterpart. In addition to batch-type equipment, continuous systems are available for salt bath nitriding. Most systems are encapsulated, with a protective front and an observation panel. REFERENCES 1. Liquid Nitriding, Heat Treating, Vol 4, ASM Handbook, ASM International, 1991, p 410–419 2. J. Easterday, Salt Bath Nitriding Proceedings (Detroit), Kolene Corp., Oct 1995 3. Cassel Manual of Heat Treatment and Case Hardening, 7th ed., Imperial Chemical Industries, Ltd., Jan 1964

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p65-70 DOI: 10.1361/pnafn2003p065

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

7

Control of the Compound Zone or White Layer THE COMPOUND ZONE is more commonly known as the white layer, simply because when the nitrided sample is sectioned through the case, and then polished and etched with a standard solution of nital (2 to 5% nitric acid and alcohol), the immediate surface etches out white in appearance above the nitrided case. The zone is called “compound” due to the presence of more than one phase (Fig. 1). Two phases generally are present in the compound zone: the epsilon (ε) phase, which has a chemical formula of Fe2-3N, and the gamma prime (γ ′ ) phase, which has a chemical formula of Fe4N (Fig. 2). Depending on which phase dominates, spalling or chipping can occur during service (Ref 2). Choice of

Fig. 1

Formation of the nitrided case. Courtesy of Pye Metallurgical Consulting, Inc.

66

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 2

Formation of the compound zone. Source: Ref 1

the necessary surface metallurgical phase should be determined by the part application.

A Test to Determine the Presence of the White Layer A drop of cupric ammonium chloride, Cu(NH4Cl)2, as a spot test on the surface of a nitrided part will indicate the presence of the white layer. If there is a white layer (or any remaining white layer after grinding or chemical finishing), the drop will deposit copper onto the surface. If there is no white layer, no copper will be deposited (Ref 3).

Reduction of the Compound Zone by the Two-Stage Process The first method of reducing the compound zone was developed by Dr. Carl Floe of the Massachusetts Institute of Technology (see Chapter 1 for historical background). He developed what was originally known as the Floe process, now more commonly called the two-stage or sometimes the double-stage process. The purpose of the two-stage process is to reduce the thickness of the compound zone on the immediate surface of the steel by reducing the nitriding potential at an elevated temperature. The first stage of the process ensures rapid formation of the white layer, and the second stage arrests the formation of the white layer without allowing the diffusion zone to be denitrided. The two-stage nitriding process is a relatively simple procedure (Fig. 3). Nitriding is carried out using ammonia gas with a dissociation of approximately 30% at 495 °C (925 °F). This is followed by raising the temperature during the last third of the cycle to 565 °C (1050 °F) with a dissociation rate of 75 to 80%. In some applications, formation of the surface compound zone is desirable. If the layer is not required, it can be removed mechanically (by grinding), or chemically (using a cyanide solution developed by Bell Helicopters) (Ref 4).

Chapter 7: Control of the Compound Zone or White Layer

Fig. 3

Typical two-stage process

Other Methods for Controlling Compound Zone Formation Nitriding Potential. Control of the compound zone can be achieved by controlling the nitriding potential of the process gas. In a 1973 paper, B.J. Lightfoot and D.H. Jack of the University of Leeds (Ref 5) suggested that at nitriding potentials that completely avoid white layer formation, only steels that contain aluminum will nitride satisfactorily. Steels that depend on chromium as the primary nitride-forming element will nitride very slowly. The white layer constituents must be stable to provide a fine dispersion of chromium nitrides in ferrite. Strict control of the nitriding potential restricts white layer growth. Building on Adolph Machlet’s early work with dilution techniques (see the later discussion in this chapter), Thomas Bell of the University of Liverpool began work in the early 1970s on dilution of the ammonia gas by hydrogen (Ref 2). His objective was to establish the mechanism of compound layer formation and to determine how the layer thickness can be managed. By monitoring the nitriding atmosphere (an ammonia-hydrogen mixture), and knowing the input gas (whether ammonia to enrich the nitrogen content or hydrogen to dilute the nitrogen content), he discovered that the nitriding potential can be controlled to produce components with a white layer thickness of no more than 0.004 mm (0.00016 in.). The dilution process was first developed by Machlet when he patented his process in 1913 (see Chapter 1 for further details). The patent was granted for the nitrogenization of low-alloy steels and cast irons, using ammonia gas diluted with hydrogen. Machlet had observed compound layer formation without realizing its significance, yet recognized that the layer could be controlled by dilutant gases such as hydrogen. The sole

/ 67

68

/ Practical Nitriding and Ferritic Nitrocarburizing

purpose of the dilution procedure, using hydrogen, is to reduce the nitriding potential of the ammonia gas. Some nitriding potential will remain, however, and thus a very shallow compound layer may form on the immediate surface, albeit microns thick. Careful control of the dilution gas in relation to the process gas is necessary in order to ensure the required surface metallurgy. Based on this method, a Canadian company (Nitrex Metal, Inc.) manufactures computercontrolled furnace equipment to control compound layer formation. The equipment seems to be enjoying good success (Ref 6). Ion Nitriding. By ionizing molecular gases and preparing the steel surface in a completely different manner compared to the gas nitriding procedure, the compound layer can be controlled to the point of elimination. The procedure relies on the ability of the control parameters to change as required by: • • •

Process gas ratios Process operating pressure levels Process temperature

With the exception of plain low-carbon steels, most steels can be satisfactorily ion nitrided. The surface metallurgy will be filled with iron nitrides. The principles of the ion nitriding process are discussed in Chapter 8.

Case Depth of Nitriding Determination of case depth in relation to the time required for diffusion of nitrogen into the steel surface is a contentious, often-discussed subject. Formation of both the compound layer and the diffusion layer is based on: • • • • •

Time Temperature Gas composition Steel analysis Steel surface condition

The nitriding procedure is based on diffusion at a selected temperature in relation to: • • • •

Potential for nitride networking Growth Distortion Corrosion resistance

The preceding considerations are factors in determining temperature. But what of the required time at temperature to establish the desired total

Chapter 7: Control of the Compound Zone or White Layer

case depth? In the late 1930s and early 1940s, F.E. Harris determined a simple formula for the effect of diffusion at temperature (Ref 7). The formula is based on the square root of time at a particular temperature multiplied by a factor for that temperature: Case depth = K√t

where the case depth is in inches, t is in hours, and K is found in the following table: Temperature °C

°F

Temperature factor (K)

460 470 475 480 500 510 515 525 540

865 875 885 900 930 950 960 975 1000

0.00150 0.00155 0.00172 0.00195 0.00210 0.00217 0.00230 0.00243 0.00262

These temperature factors are based on a commercially available nitriding steel without the addition of aluminum and do not account for what might be considered abnormally high alloying contents (such as might be found in stainless steels). The factors also do not account for surface condition and variations in the particular cast/melt. The factor values are approximate and should be used only as a guide. Similar guides for estimating case depth for plasma nitriding and ferritic nitrocarburizing can be found in Chapters 12 and 21, respectively. Remember, the higher the alloying content of the steel is, the slower the diffusion rate of nitrogen into the steel surface. The lower the alloying content is, the faster the diffusion rate. Temperature uniformity within the process retort is critical to uniform case depth as well as uniform case metallurgy. If temperature variations occur within the process retort, both the compound layer and the case depth also can vary. REFERENCES 1. M.A.J. Somers and E.J. Mittemeijer, Härterei-Technische Mitteilungen, Vol 47 (No. 5), 1992 2. T. Bell et al., Controlled Nitriding in Ammonia-Hydrogen Mixtures, Heat Treatment ’73, The Metals Society, Dec 1973, reprinted in Source Book on Nitriding, American Society for Metals, 1977, p 259–265

/ 69

70

/ Practical Nitriding and Ferritic Nitrocarburizing

3. D.B. Clayton and K. Sachs, Reduction of “White Layer” on the Surface of Nitrided Components, Heat Treatment ’76, The Metals Society, May 1976, reprinted in Source Book on Nitriding, American Society for Metals, 1977, p 242–247 4. D.A. Dashfield, Nitriding Problems and Their Solutions, Met. Prog., Feb 1964, p 88–93 5. B.J. Lightfoot and D.H. Jack, Kinetics of Nitriding With and Without White-Layer Formation, Heat Treatment ’73, The Metals Society, Dec 1973, reprinted in Source Book on Nitriding, American Society for Metals, 1977, p 248–254 6. W.K. Liliental, G.J. Tymowski, and C.D. Morawski, Typical Nitriding Faults and Their Prevention Through the Controlled Gas Nitriding Process, Ind. Heat., Jan 1995, p 39–44 7. F.E. Harris, Case Depth, Met. Prog., Aug 1943, p 265–272

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p71-88 DOI: 10.1361/pnafn2003p071

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

8 Ion Nitriding

GAS IONIZATION is a method of causing a gas to develop an electrical imbalance. At a particular pressure and electrical potential, the gas will glow—not unlike the gas used in neon light tubes. The phenomenon of gas ionization was first investigated in the 1800s, when studies made in the upper reaches of the Arctic Circle in Norway showed the aurora borealis to be lowpressure ionized air that exhibits a characteristic spectrum of the rarer gases present in the upper atmosphere. Observable in an indefinite region of the night sky around the magnetic pole, the ionization is influenced by the magnetic disturbances of the sun and terrestrial magnetism from the earth. The “dancing” movement of the aurora is caused by upper atmosphere winds. The plasma technique is based on this natural phenomenon, and the nitriding process that utilizes it is known as ion nitriding. Ion nitriding is also called: • • •

Plasma nitriding Glow discharge nitriding Plasma ion nitriding

Derivatives of the three processes listed previously include: • • • •

Continuous direct current (dc) plasma nitriding Pulsed dc plasma nitriding Hot-wall furnace system Cold-wall furnace system

History of Ion Nitriding The plasma technique was first put to use as a metallurgical processing tool by Dr. Wehnheldt, a German physicist, but he was unable to control it as a nitriding process due to the instability of the glow discharge. This type of instability can be seen in “lumina storm” novelty lamps. The discharge dances away from the center ball and around the inner glass case.

72

/ Practical Nitriding and Ferritic Nitrocarburizing

Shortly after Wehnheldt’s discovery, he met the Swiss physicist and entrepreneur Dr. Bernhard Berghaus (Ref 1). Together they were able to develop, control, and market the process as a gas nitriding alternative that offered acceptable control of the compound zone. Development of the plasma relies on: • • •

Low (less than atmospheric) pressure Voltage Gas composition

The plasma technique arrived in the United States during the 1950s. One of the first U.S. companies to recognize the usefulness of the process was General Electric. G.E. engineers Dr. Claude Jones, Derek Sturges, and Stuart Martin successfully investigated the glow discharge method of nitriding and were able to use the process on a wide variety of materials and components (Ref 2). In Germany, the work of Wehnheldt and Berghaus led to formation of the company Klockner Ionen, which commercialized the process. The company designed and built ion nitriding equipment, and licensed other international companies to build the equipment and develop the process as well. During the mid-1970s, scientists at the University of Aachen in Germany worked on better methods of controlling the glow discharge and other associated phenomena such as arc discharging. The procedure developed at Aachen was that of pulsed dc current technology, which simply means interrupted power to the point of power shutdown. This technique offered many advantages to process engineers in terms of control of the nitriding procedure.

How the Ion Nitriding Process Works The process is based on the phenomenon of current flowing between two electrodes placed in a sealed gaseous environment. The gas within the tube acts as an electrical conductor and carries the current from one side to the other as it would if it were a wire conductor (Fig. 1). The gaseous atoms become excited and are propelled along a very short “mean free path” and collide with one another. When this occurs, energy is released and a “glow” occurs—hence the name “glow discharge nitriding.” The color of the glow is determined by the type of gas used. At normal atmospheric temperature and pressure, the resulting energy is too low to have any significant use as an energy source in terms of providing heat. As the pressure is decreased to the region of 0.1 Pa (10–5psi), the “molecular mean free path” is increased. The energy release on molecular impact is great but is infrequent due to the long molecular mean free path that each gas molecule has to move to find another molecule to impact. Thus the plasma can then generate heat, but in an amount insufficient to heat the workpiece surface for plasma processing. In other words,

Chapter 8: Ion Nitriding

Fig. 1

Influence of pressure on the glow discharge. (a) A sealed glass tube containing a gas at normal atmospheric temperature and pressure (100 kPa, or 15 psi) conducts an electrical current. The gas glows brightly but does not release much usable energy. (b) At low pressure, for example, (0.013 Pa, or 2 × 10–6 psi), very few gas molecules exist and collide infrequently, releasing low energy. The glow appears almost like a fog. (c) At higher pressure, say 1.3 to 13 Pa (2 × 10–4 to 2 × 10–3 psi), the gas molecules move freely and impact frequently, releasing usable plasma energy that glows brightly and crisply.

because of the infrequency of molecular impact, the resulting energy still cannot be used as a heating medium. There is no ideal pressure value, but there is a range in which the pressure can be adjusted to suit the operating parameters of the particular material and geometry. That pressure range is given to be 50 to 500 Pa (0.007 to 0.07 psi). Control of the process chamber pressure will determine the area of glow on the steel surface. If the pressure is too high (i.e., toward atmospheric pressure), then the glow seam that surrounds the part will become intermittent. Where there is no glow on the steel surface, there will be no nitride formation. This means that the voltage has failed to support the plasma ignition at that particular pressure. In addition, hightemperature areas can occur at sharp corners, often resulting in localized overheating and even burning. If the pressure is too low (i.e., toward high vacuum), then the glow seam around the part will be very “foggy,” resulting in ineffective nitriding of the internal surface areas of holes. Pressure is one of the principal areas of control. If the pressure conditions are correct and the voltage is too high, then a discharge will occur much like a lightning strike. This occurrence is known as the arc discharge region. If the voltage is too low and the pressure too high, the glow will disappear and nothing will happen.

/ 73

74

/ Practical Nitriding and Ferritic Nitrocarburizing

Glow Discharge Characteristics To understand the principles of the glow discharge, it is necessary to refer to the Paschen curve (Fig. 2). The Paschen curve is a comparison of input voltage in relation to current density of the steel part surface, and the various events that take place depending on voltage in relation to current density. By understanding the Paschen curve, one can determine process voltage requirements. Nonmaintained Region. The nonmaintained region of the Paschen curve states that if a voltage is applied to a gas, then the electrons within that gas can be charged to the point where an electrical ignition of the gas will occur. This can be likened to the spark that occurs when an automobile spark plug is charged with high voltage. The air between the plug gap is electrically charged to the point where a spontaneous spark will occur. In the process chamber the electrons from the ionized, ignited gas will be accelerated toward the cathode from the anode. Once the molecular collision begins to occur due to the gas ionization, energy (heat) will be generated at the work surface. When the process voltage is increased, then an appropriate increase will occur in the current density. The gas ionization will progress into the next phase of the Paschen curve. Self-Maintained Region (Townsend Discharge). The self-maintained region on the curve is the area in which more electrons can be released by further gas ionization, which will perpetuate even further gas ionization. The area can be considered to be a chain reaction.

Fig. 2

Paschen curve showing the relationship between voltage and current and the various glow discharge characteristics

Chapter 8: Ion Nitriding

Transition Region (Corona). The current density will increase if the current limiting resistance is reduced, thus causing a voltage drop between the cathodic workpiece and the anodic internal vessel wall. Within this area, voltage stability cannot be maintained. Subnormal Glow Discharge. In this region the glow discharge is beginning to ignite and a very fuzzy glow will be seen. Normal Glow Region. It is at this point on the Paschen curve that a glow seam will completely cover the work. Its thickness will be determined by the chamber vacuum pressure and the process voltage. Abnormal Glow Region. It is in this region that the glow seam will completely cover the steel work surface uniformly, following the geometric profile of the workpiece. One must now adjust the process pressure to ensure penetration of blind holes and cavities. However, the ratio of hole diameter to hole depth must be considered: 1× hole diameter : 4× hole depth

The blind hole depth can equal four diameters. This ratio will follow for small holes down to a diameter of approximately 4 mm (1/8 in.). However, the ratio will double if the hole is a straight-through hole to: 1× hole diameter : 8× hole depth

A through-hole depth may equal eight diameters. The ratio is valid for small hole penetrations, but for large holes from 50 mm (2 in.), the rule will not apply; the glow seam can be “forced” down the hole by the adjustment of pressure. It is in this region of abnormal glow that plasma nitriding takes place and where ideal conditions exist. Arc Discharge Region. As the current density increases, a noticeable increase in the voltage drop will occur, causing an appreciable increase in the power density at the work surface. As power density increases, the temperature of the steel work surface rises to the point of serious overheating, resulting in metallurgical problems such as grain growth, localized melting, and pitting. As the power intensity builds, the potential for an arc will occur. This arc, visible through the process chamber sight glass, looks like a lightning strike.

Process Control Process parameters requiring good control are: • • • •

Current density Power Process chamber pressure Gas composition

/ 75

76

/ Practical Nitriding and Ferritic Nitrocarburizing

Instrumentation for achieving control of these parameters is the personal computer/programmable logic controller (PC/PLC) combination, ideal for the plasma nitriding process. Control Characteristics. When a partial pressure is created in a process chamber and a constant voltage is applied to the chamber with the appropriate cathode potential and anode, then the gas in the partial pressure will ionize. If a steel part is placed in contact with the cathode so that it is at the same electrical potential, then not only will the partial pressure gas “glow,” it also will create heat in the steel part at the cathode potential due to the kinetic energy generated by ionic bombardment. If the process gas used in the process chamber is diffusible in steel when heat is applied, then the diffusion processes can be accomplished. The diffusible gases are nitrogen, hydrogen, carbon monoxide, boron, sulfur, and methane. Elements. In the cold-wall or continuous dc type of system, there are no heating elements within the furnace, simply because the thermal energy is created by the ionic bombardment of the workpiece by the free gas ions. The workpiece is heated by kinetic energy from ionic bombardment. However, some systems do use internal elements to assist in temperature uniformity with large, densely packed loads. When using a system with internal heating elements, care must be taken in positioning the workpieces to prevent localized overheating. Otherwise, nonuniform surface metallurgy may occur across the load.

Other Uses for Plasma Processing Any thermochemical process can be accomplished by plasma processing, provided that an appropriate furnace and gas are used. Plasma can be used for other surface treatments such as diffusion and deposition techniques. Diffusion techniques include (Ref 3): • • • •

Plasma-assisted nitriding Plasma-assisted ferritic nitrocarburizing Plasma-assisted carbonitriding Plasma-assisted carburizing

Deposition techniques, more commonly known as “thin film” processes, can be accomplished through plasma generation. The thin-film deposition technique can be divided into two groups: (a) tribological (wear and corrosion resistance) and (b) decorative coatings. Thin-film deposition techniques deposit a thin metallic film onto a carefully prepared steel surface, resulting in hardness and corrosion resistance. The thin-film coating is usually (but not always) preceded by a plasma nitriding procedure, which produces a hard substrate to which the thinfilm material can bond. An example of thin-film processing is plasmaassisted chemical vapor deposition (PA-CVD) using chromium, tungsten,

Chapter 8: Ion Nitriding

aluminum, titanium, metallic-carbon combinations, or other materials. Techniques are available for creating duplex thin-film deposits.

What Happens in the Ion Nitriding Process The process makes use of the familiar formula for ammonia: 2NH3 = N2 + 3H2

On decomposition, ammonia breaks down to its elemental forms of nitrogen and hydrogen. The reverse will happen on cooling the decomposed gas. In ion nitriding, unlike gas nitriding where ammonia gas is used, the gases are brought in as separate gases: N2 + H2

Because the gases are not in a combined form, the metallurgist is able to vary the nitriding potential by varying the proportions of the individual gases. By varying the hydrogen-to-nitrogen ratio of the elemental gases, the compound zone (white layer) formation can be controlled. When the nitrogen gas is introduced into the process chamber, the gas will be ionized: e– → N2 = N– + N

Gas Ratios When ammonia gas is decomposed under heat, it will decompose into the following elemental gas ratios (Fig. 3): 1 Nitrogen : 3 Hydrogen → N2 + 3H2 2NH3 ←

Fig. 3

Illustration of the ammonia molecule 2NH3 and its decomposition

/ 77

78

/ Practical Nitriding and Ferritic Nitrocarburizing

This formula for ammonia shows three hydrogen molecules to one nitrogen molecule, and the ratio is a fixed ratio. If the ratio of nitrogen gas to hydrogen gas is varied, any ratio can be selected to create any particular surface metallurgy. The ratio can be a low nitrogen to hydrogen ratio, or a high nitrogen to hydrogen ratio, depending on the surface metallurgy requirements. In other words, a fixed gas chemistry produces a fixed surface metallurgy, and a variable gas chemistry allows a variable surface metallurgy. Variable gas proportions permit creation of the appropriate surface metallurgy that will best suit both the application and the steel being processed. This means that metals such as the following can be treated with ease: • • • • • • • •

Low-carbon steels Pure iron Austenitic stainless steels Martensitic stainless steels Powder metallurgy ferrous materials High-strength low-alloy (HSLA) steels Tool steels Refractory metals

Reactions at the Steel Surface During iron nitriding, four reactions will occur at the surface of the material being treated. Reaction I. Ionized and neutral nitrogen atoms are produced by energized electrons: e– → N2 = N+ + N + 2e–

Reaction 2. Iron and other contaminants are removed from the surface of the work by an action known as sputtering. The impact of the nitrogen ions bombarding the work surface dislodges the contaminants, which are removed by the vacuum pumping system. Contaminant removal can be loosely described as atomic cleaning and allows nitrogen to diffuse into the work surface: N+ → Work surface = Sputtered Fe and sputtered contamination

Reaction 3. As a result of the impact of the sputtered iron atoms, case formation begins at the work surface of iron nitrides: Sputtered Fe + N = FeN

At this point, intensive surface cleaning of the workpiece occurs due to the sputtering action on the surface. This can be likened to atomic shot

Chapter 8: Ion Nitriding

blasting where the carrier medium is the air blast and the cleaning medium is the steel shot. In this instance the gas ion is like the steel shot, and the electrical imbalance is like the air blast. The surface becomes atomically cleaned. Reaction 4. At the work surface the breakdown of FeN begins under the influence of continual plasma bombardment. The plasma causes instability of the FeN, which begins to break down into the ε-phase, followed by the γ′-phase and an iron/nitrogen compound zone (Fig. 4): FeN → Fe2N + N Fe2N → Fe3N + N (ε-phase) Fe3N → Fe4N + N (γ′-phase) Fe4N → Fe + N (iron/nitrogen compound zone)

Fig. 4

Glow discharge ion nitriding mechanisms (Koelbel’s models). Note the voltage profile on top. The potential drop is greatest near the workpiece so this is where the ions have the most kinetic energy and this is where the plasma will glow brightest.

/ 79

80

/ Practical Nitriding and Ferritic Nitrocarburizing

Thus, if one can control the reactions at the steel surface, one can control the surface metallurgy. Hardness profiles show control of the formation of the compound zone and the diffusion zone.

Surface Stability Each of the surface layers achieves excellent dimensional stability simply because the temperature to generate plasma is not dependent upon a conventional heat source. The plasma energy is the heat source, and the process temperature can be adjusted to suit the steel by manipulation of plasma voltage and pressure. However, use of a lower process temperature will extend the cycle time for diffusion.

“Corner Effect” and Nitride Networking The corner effect is a phenomenon that can occur in any type of diffusion heat treatment process, including plasma nitriding, gas nitriding, and carburizing (Fig. 5). Because nitrogen is diffusing from all angles of the corner, a normal reaction of nitrogen saturation occurs at the corner (particularly on gear teeth). If allowed to proceed unchecked, a supersaturated solution of nitrogen can form at the corners, leading to nitride networking throughout the corner region. The net result is that the corner can become very brittle and prematurely fail by chipping or spalling (Fig. 6, 7). If the

Fig. 5

Illustration of the corner effect due to the multidirection of nitrogen into the steel surface. Nitride networking also is shown.

Chapter 8: Ion Nitriding

Fig. 6

Chipping at the pressure point of a gear tooth. Courtesy of Pye Metallurgical Consulting, Inc.

Fig. 7

Extrusion die with surface exfoliation. Courtesy of Pye Metallurgical Consulting, Inc.

nitrogen is controlled to reduce the nitriding potential to suit the part application, then the risk of nitride networking is considerably reduced or nearly eliminated.

Degradation of Surface Finish Sputter cleaning and ion nitriding generally enhance the surface condition of the workpiece. Ionic bombardment of the steel surface reduces turning lines and high points. Of course, care must be exercised when using denser gases such as argon for ionic bombardment. Argon will tend to etch the surface to the point where the surface will become pitted. When aggressive sputter cleaning is necessary, it is strongly recommended that the argon be blended with hydrogen to a ratio of 95% H2 to 5% Ar up to a maximum of 90% H2 to 10% Ar.

/ 81

82

/ Practical Nitriding and Ferritic Nitrocarburizing

Control of the Compound Zone Plasma nitriding can control compound zone formation on many different steels and engineered components, including: • • • • • • • • • • •

High-speed steel cutters Aluminum extrusion dies Hot forging dies Broaches Plastic extrusion screws and barrels Aluminum shot sleeves Aluminum die casting dies Hydroforming tools Autobody blanking dies Helical gears, spiral bevel pinions, flat bevel gears, and spur gears Auto engine valves

By manipulating the compound zone formation, wear-resistant, impactresistant, and corrosion-resistant surfaces can be created. Gas nitriding uses anhydrous ammonia (2NH3) as the nitrogen source, which has a given ratio of 1:3 (one part nitrogen to three parts hydrogen). The limitation of the fixed ratio is that the ammonia will produce a mixed phase compound zone of both ε and γ′ (usually around 50% each). High internal stresses result from the different phase volumes, which means that the crystal interfaces are inherently weak. The thicker the compound zone, the weaker it becomes, causing weak crystal boundaries within the zone to fail even under small loads. This is particularly evident on parts such as H13 hot-work tool steel for aluminum extrusion dies (Fig. 8a). If the compound zone is allowed to develop into a thick layer, the die will bend when loaded by the aluminum billet (Fig. 8b). As the die bends, the loaded face becomes concave and the back face becomes convex. The compound zone on the loaded face will undergo compression and crack. This also will occur if the effective case depth of the nitrided layer exceeds 0.25 mm (0.010 in.). A similar effect will occur on an H13 hot-work press forging die if the case is too deep and the compound layer too thick. As the press loads the die face, the surface of the die becomes compressed and the formed case deforms at the surface, leading to surface cracking and exfoliation. This is not serious if the exfoliation occurs on areas of the die that are not in use, but when exfoliation occurs on a working face it becomes a major problem. Cracks can initiate quite easily due to impact loading or incorrect preheat treatment. If the core hardness mechanical properties are lower than the load conditions of the treated steel, then the core is likely to fail to support the nitrided case on the surface, no matter how well the nitriding has

Chapter 8: Ion Nitriding

Fig. 8

Cross section through an extrusion die. (a) Bearing face and relief clearance area. (b) Aluminum billet for extrusion against the die face, which is soft in the core. Side A will compress and crack; B will stretch and tear on the case.

been done. Therefore, the thinner the compound zone (to the point of elimination), the more ductile the steel and the better its fatigue properties. Ion nitriding allows control of the thickness of the compound zone.

Process Gases Metallurgical-grade nitrogen and high-purity hydrogen are the gases primarily used for ion nitriding. Argon can be used, but only to assist in component cleaning (known as sputter cleaning) before the nitriding sequence. Methane can also be used to deliver controlled amounts of carbon to influence control of the ε-phase in the compound zone. Once again, care should be exercised in using methane; too much carbon can actively promote a dominant ε compound phase. Accurate control of the gas delivery into the process retort ensures accurate control of the nitriding metallurgy. Precise metering of each gas is accomplished by mass flow controllers.

/ 83

84

/ Practical Nitriding and Ferritic Nitrocarburizing

Process Parameters In conventional gas nitriding, time, temperature, and dissociation (or salt analysis for salt bath nitriding) are the key process parameters that must be controlled. In ion nitriding, the control parameters include: • • • • • • • •

Time Temperature Pressure Current density Amperage Voltage Gas flow Gas ratio

Depending on the type of plasma technology employed, other parameters can be controlled to ensure repeatable process conditions. Temperature uniformity during the nitriding process is critical. Without it, the formed case will be nonuniform. Some latitude in temperature tolerance can be allowed, but usually no more than ±5.5 °C (10 °F) of the process set point. Remember, the higher the process temperature, the greater the risk of nitride networks due to the potential for an excess of nitrogen in iron. This also means that nonuniform case depths will occur across the temperature range within the process retort. For example, a temperature difference of 35 °C (60 °F) over a 24 h cycle period can result in a case depth variation of up to 0.038 mm (0.0015 in.). Obviously, the longer the process cycle time, the greater the case depth variation. This principle applies to gas, salt bath, as well as plasma processing techniques.

Plasma Generation Philosophies There are two methods of plasma generation: continuous dc and pulsed dc. Continuous dc means that plasma is generated with a particular dc current based on a given work surface area. The voltage requirement will vary with the pressure (Fig. 9). With this type of system, the process heat generation normally is derived from the kinetic energy created on ionic bombardment. This means that the furnace does not have heating elements as are often seen in conventional furnaces. Pulsed dc can be likened to entering a darkened room and switching on the light. On leaving the room, the light is switched off. The peak voltage from the power source is constant, but the duration of the power to the light is variable. Pulsed dc power supplies have the ability to switch the power on and off as required by component geometry. The pulsed dc peak voltage can be varied according to the part and chamber configuration (Fig. 10). The equipment will be discussed in greater detail in Chapter 9.

Chapter 8: Ion Nitriding

Fig. 9

Continuous dc power plasma nitriding

Fig. 10

Pulsed dc plasma showing Pmax, Pmin, and Ptemp (where P = power) in relation to power input versus time. Note that the voltage can be adjusted, as can the duration of the pulse. Courtesy of Plateg USA

Advantages Here are a few of the advantages of the plasma generation technique for nitriding: • • • • • • • • •

Based on known technology Environmentally friendly gases No fire risk Shorter cycle times Better furnace utilization during continuous production Very low operating costs Minimal operator involvement Low process gas consumption Repeatable process parameters

/ 85

86

/ Practical Nitriding and Ferritic Nitrocarburizing

• • • •

Repeatable metallurgy Ability to treat almost any steel Ease of selective nitriding Low-maintenance equipment

Environmental Impact. Because ion nitriding gases are used in such small volumes and are nontoxic and environmentally friendly, the process will cause no harm to the air or operational personnel. The process can be successfully used in cell manufacturing techniques. The volumes of gases used in the process are considerably less than those used in gas nitriding. There is no burnoff or gas scrubbing to consider when using the plasma nitriding process.

Oxynitriding The oxynitriding process has grown in popularity over the past two decades, especially in pulsed plasma nitride processing. The pulsed plasma nitride unit offers the capability of a controlled backfill of moisture in the form of either a vapor or oxygen-bearing gas. The purpose of the oxynitriding process is to form a controlled oxide layer on the surface of the treated steel. Once the oxide barrier has been formed, there is a resistance to corrosion. The degree of corrosion resistance will be determined by the thickness of the oxide layer, which in turn is determined by both time and temperature. The oxynitriding process can be performed in gas, salt, or plasma. The procedure is done on completion of the nitride cycle, when the control program moves into the cooldown mode. It is during the cooldown procedure that oxygen is fed into the process chamber (Fig. 11). The net effect is that the nitrided surface is deliberately oxidized to provide a corrosionresistant, oxygen-rich surface layer. Upon completion of the cooldown, the furnace bell is opened and the oxynitrided work is removed. Figure 12 shows a plain carbon-manganese steel piston rod that has been nitrided, followed by the controlled oxynitride procedure. The center rod shows the rod before the start of treatment. The two rods at the right have been oxynitrided, and the two rods at the left are untreated; all four have been subjected to salt spray testing. The process gases used for the controlled oxynitriding procedure are environmentally friendly and pose no threat to the ecology or the immediate environment. The surface finish of the steel after the procedure is dark blue, almost matte black. The oxidation layer is usually around 1 µm thick, but can be varied according to the cooldown time and the time held at an elevated temperature (around 900 °F, or 480 °C). Generally the procedure is to go into cooldown immediately after nitriding and commence the controlled oxidation treatment. It is a simple, effective procedure that adds no significant time to the overall process.

Chapter 8: Ion Nitriding

Fig. 11

Schematic illustration of the oxynitriding process

Fig. 12

Examples of oxynitrided piston rods. Center rod: before treatment. Two rods at left: untreated and subjected to salt spray testing. Two rods at right: treated, then subjected to salt spray testing. Material is similar to UNS G41400 and H41400 chromium-molybdenum steels. Courtesy of Plateg USA

Oxynitriding produces a pleasing cosmetic appearance and ensures a high degree of surface protection against corrosion. The procedure can be applied to items such as boring bars, cutting tool holders, broaches, drill bits, and cutting tools. In fact, oxynitriding can be applied to any workpiece subject to corrosive conditions and is not restricted to cutting tool applications.

/ 87

88

/ Practical Nitriding and Ferritic Nitrocarburizing

REFERENCES 1. F. Hombeck, Forward View of Ion Nitriding Applications, Ion Nitriding, T. Spalvins, Ed., ASM International, 1987, p 169–178 2. C.K. Jones, D.J. Sturges, and S.W. Martin, Glow Discharge Nitriding in Production, Met. Prog., Dec 1973, reprinted in Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 186–187 3. Heat Treating, Vol 4, ASM Handbook, ASM International, 1991

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p89-110 DOI: 10.1361/pnafn2003p089

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

9

Ion Nitriding Equipment ION NITRIDING EQUIPMENT falls under two categories: cold-wall continuous direct current (dc) technology and hot-wall pulsed dc technology. Both are described in this chapter along with other important considerations for ion (plasma) nitriding equipment and processing.

Cold-Wall Continuous dc Plasma Nitriding The cold-wall continuous dc plasma system is perhaps the simplest of the plasma nitriding furnace systems (Fig. 1). It consists of a simple vacuum chamber encapsulated with a water jacket such as that found in a

Fig. 1

Schematic of a typical cold-wall continuous dc plasma nitriding system. Source: Ref 1

90 / Practical Nitriding and Ferritic Nitrocarburizing

conventional vacuum furnace construction. As the furnace is essentially a vacuum furnace, it will require: • • • •

Vacuum pumping system Power source Gas supply Process controller

Process Parameters Plasma nitriding involves more process control parameters than gas nitriding does. Parameters for plasma nitriding include: • • • • • • • • • • • •

Input voltage Amperage Chamber pressure Workpiece temperature Current density Nitrogen gas flow Hydrogen gas flow Methane gas flow Oxidizing gas for oxynitriding flow rate Process time Temperature rise rate Temperature cooldown rate

These parameters, along with a few derivatives of them, are controlled by a personal computer/programmable dedicated logic controller (PC/PLC) system.

Cold-Wall Furnace The primary component of the cold-wall furnace is the furnace process chamber, constructed much the same as a conventional vacuum furnace and consisting of inner and outer vessels. The inner vessel, or vacuum vessel, is usually fabricated from stainless steel, and the outer water jacket is usually manufactured from carbon steel. A water-cooling area between the two vessels conducts any heat losses from the inner vacuum vessel to the water and to a heat exchanger. The vessel sidewall usually is fitted with a sight port for observing the plasma conditions in the work area. Through the base of the furnace are fitted the power feedthroughs, which create the cathode potential of the hearth within the furnace, as well as the thermocouple feedthrough. The power feedthrough is designed to allow continuous power flow to the cathode feedthrough.

Plasma Generator Power Pack The plasma power generator is usually a solid-state unit designed to produce continuous dc power from line voltage through to the variable

Chapter 9: Ion Nitriding Equipment / 91

voltage control to the furnace cathode power feedthroughs situated within the process chamber. Because the power feedthrough allows continuous power from the plasma generator, its insulating characteristics must be of a very high quality to insulate the cathode from the anode. Thus an electrical bias is set up between the process chamber and the cathodic workpiece. Some older power generator designs used a straight dc power system. Glow stability was sometimes a problem when plasma nitriding complex part geometries and holes, particularly blind holes. This led to the use of high-process voltages in order to address the nitriding of complex geometries, which in turn created problems arising from the high voltage in relation to the arc discharge region on the Paschen curve (discussed later in this chapter). Should the arc be initiated, there is a serious risk of both metallurgical and mechanical damage to the part, caused by overheating, possible burning, and possible stock removal at sharp corners on the workpiece. Therefore, some form of arc detection and suppression electronic-control system is mandatory. The power supply is used to set up a bias between the workpiece and the vacuum chamber wall. The disadvantage of these units is that when operating in the lower regions of the glow, it is difficult for the glow to penetrate along the form of the part being treated, particularly in parts with blind holes or complex geometries. In such cases, both pressure and voltage must be varied; however, if high voltage and pressure levels are selected, there is a serious risk that the arc discharge region could be reached and arcing could occur.

Heating Elements A cold-wall furnace normally has no heating elements. Heat into the part is generated by the kinetic energy developed by the ionic bombardment and is controlled simply by voltage and current density regulation. In some cases the furnace manufacturer will design a furnace with supplementary elements to assist the plasma heating. These elements would most likely be found within the furnace process chamber and usually are electrically isolated to prevent them from being nitrided.

Furnace Thermocouples Perhaps even more important than control of process temperature is control of part temperature. Unlike more conventional heat treatment methods, temperature generally is measured at the part rather than the chamber. In conventional heat treating, the temperature generally is measured by a thermocouple located within the process chamber. It is often incorrectly assumed that what the thermocouple is measuring is the chamber temperature. However, the thermocouple measures only the temperature at the point of the thermocouple. It does not measure temperature at any other point within the process chamber.

92 / Practical Nitriding and Ferritic Nitrocarburizing

Many factors within the furnace chamber influence temperature uniformity. Therefore, part temperature—not chamber temperature—must be measured. The part and process temperature are measured by considering the thinnest part and the thickest part within the process chamber. The part temperatures are usually held to within a tolerance band of ±5 °C (10 °F). If the thermocouple cannot be attached to the workpiece, then it should be attached to dummy test coupons that are representative of the workpiece cross-sectional area and the material being treated. The thermocouple must be isolated from the furnace anode. As in conventional process temperature measurement, the signal from the thermocouple is generated in millivolts and transmitted back to either a process controller or a data logger to record the process control parameters. The signal transmission method can be either conventional hard wire or fiber optics. Temperature uniformity throughout the process chamber is mandatory to ensure uniform surface metallurgy, surface hardness, core hardness, and case thickness. Without temperature uniformity, serious metallurgical conditions can result.

Gas Flow Good process gas-flow input control is critical. Some of the earlier systems used flowmeters. Although this method worked to some extent, it was not accurate enough. Later methods used micrometer needle valves, but these too could not offer a high degree of accuracy and repeatability. A more accurate method of controlling gas delivery is the mass-flow controller, an electronic device that allows precise flow control. Use of the mass-flow controller is not limited to plasma nitriding; it can also be used in gas nitriding, particularly when using the dilution method. Gas flow can affect nitriding quality. The required gas flow remains constant only if the work surface area also remains constant. Most heattreat shops cannot guarantee constant same-surface area loads; therefore, the gas flow requirements will vary from load to load, and will vary in relation to the type of surface metallurgy required. Depending on the process retort size, the usual gas flow consumption rate can be up to approximately 100 L/h. A 600 × 900 mm (24 × 36 in.) furnace would use up to 30 L/h maximum. The reason for low gas consumption is simply because only the gas necessary for the process is used. There is no “sweep” gas usage, as with conventional methods of nitriding.

Vacuum Pump Chamber pressure is controlled by a simple mechanical vacuum pump, a rotary-vane vacuum pump (Fig. 2, 3), or a combination system of a mechanical pump and a roots blower. With an operating pressure of approximately 10 to 500 Pa (1.4 × 10–3 to 7.3 × 10–2 psi or 0.075 to 3.75 torr), the roots blower improves the system pumping speed and compression. Lower-vacuum diffusion pumping systems and cryogenic systems would be used only

Chapter 9: Ion Nitriding Equipment / 93

Fig. 2

Schematic of a rotary-vane pump. Source: Ref 2

when a very clean inner chamber is required and low residual gas values are necessary. Pump sizing is an important aspect of furnace design, as this will determine the pumpdown time of the process chamber. There is no point in oversizing the vacuum pump to pump down in less than 10 min. The vacuum pump-out port on the furnace chamber is usually located in the hearth. The furnace hearth and the vacuum pump are connected by a flexible stainless steel connector. Maintenance of the vacuum pump is critical but simple. The vacuum pump oil must be checked weekly, and changed every 3 to 6 months. The correct gas ballast setting is mandatory. Remember, one of the process gases is hydrogen, which is a very soluble gas as well as highly flammable and explosive if mixed in the correct combination with oxygen. If the gas ballast setting is done incorrectly, an explosion or fire could result. Note: Refer to the manufacturer’s operating and maintenance manual for the correct method of setting the vacuum pump gas ballast. If the vacuum pump should fail, the unit will not plasma nitride and production will stop.

Cathode and Anode The vacuum vessel acts as the anode potential, and the furnace hearth is attached to a specially designed and insulated power feedthrough, which in turn is connected to the dc power source. The principal concern with the

94 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 3

Four stages in the cycle of a rotary-vane pump: induction, isolation, compression, and exhaust. Source: Ref 3

power feedthrough is that it be insulated with a very dense ceramic material with high insulation characteristics. The insulator is then covered with a steel cover and each cover is electrically separated.

Hot-Wall Pulsed dc Plasma Nitriding The hot-wall pulsed dc plasma nitride furnace is similar in some respects to the cold-wall continuous dc system. The major physical difference between the two is that the hot-wall system is fitted with an insulated heating bell furnace around the process chamber. This means that workpiece heating can now be separated from plasma generation, because plasma is not necessary to heat the parts. Plasma is used only for workpiece surface preparation and to ionize the process gas. Figure 4 shows a typical schematic layout of a hot-wall plasma furnace. Table 1 compares the cold-wall and hot-wall plasma nitriding systems.

Chapter 9: Ion Nitriding Equipment / 95

Fig. 4

Schematic of a hot-wall pulsed dc plasma nitriding furnace and associated equipment. Courtesy of Plateg GmbH

Table 1 Comparison between hot-wall and cold-wall plasma ion nitriding systems Question

Cold wall

Hot wall

At what temperature is plasma started?

Room temperature

At a suitable elevated temperature, usually around 200 °C (400 °F)

Why is plasma started at those temperatures?

The cold-wall furnace uses a constant dc system, which requires plasma voltages around 600 to 800 V. Mechanical and metallurgical damage to the workpiece surface may occur by processing so close to the arc discharge region.

The hot-wall furnace utilizes a partial pressure condition using hydrogen or nitrogen as a thermal conductance gas. The vacuum retort is heated only by external heaters and not by plasma voltage. This means that the input voltage is not as high (400 to 500 V), and is away from the arc discharge region.

Why is there a difference in plasma generation voltages?

To heat and maintain the workpiece temperature, the required power (kW) corresponds to a current density on the workpiece of approximately 10 A/m2 (1 A/ft2) at this partial pressure and voltage.

Because the part is preheated the power required to maintain the workpiece temperature at this partial pressure corresponds to 1–2 A/m2 (0.09–0.2 A/ft2). A lower voltage is enough to produce these currents.

How does the heatup rate compare?

The cold wall usually requires more time for heatup.

The hot-wall heatup of the port is usually about 15 times faster.

Why pulse the power input?

With a constant voltage input, there is a constant heat output. Reducing voltages to reduce temperature changes the current density and the cathode fall voltage distance (glow seam); therefore, other parameters also must be changed.

With a pulsed voltage, high voltage can be used without risk of overheating the part, or taking the part to the point of arc discharge. This means that the other parameters need not be changed.

What happens to the heat?

With a cold-wall system the released electron is hotter than the ion. This means that the electron goes back to the furnace wall (anode) and creates heat. Heatup of the wall will continue, necessitating water cooling of the wall to dissipate heat.

The hot-wall furnace combined with the pulse technology uses external blowers to prevent excessive wall heating. The wall temperature can safely rise to around 650 °C (1200 °F) without concern over heat buildup.

What happens if the glow seam must be pushed into a deep blind hole?

Increasing the operating pressure causes a corresponding increase in current density, followed by an increase in part surface temperature.

Using the hot-wall pulsed power system with the same pressure/temperature combination and the same voltage/current relationship as the cold wall, the plasma energy can be maintained by varying the duty cycle (pulse variation), even with a changing voltage and current density.

Source: Ref 4

96 / Practical Nitriding and Ferritic Nitrocarburizing

Another substantial difference is that the plasma power generation pack is now a pulsed dc generator, rather than a continuous dc unit. At first, this might not seem to be a striking development. However, the Paschen curve (Fig. 5) shows that high voltages are needed to generate a plasma glow. The use of high process voltages at ambient temperatures is quite dangerous to the workpiece because of the potential for arc discharge. Pulse technology nullifies that risk simply by pulsing the power at such a frequency as to completely interrupt the power supply, thus eliminating the arc buildup and consequently the risk of arc discharge. Hot-wall pulsed plasma nitriding equipment contains no internal heating elements. If heating elements are installed inside the process chamber, then they will be nitrided. However, supplementary heating elements are located in the external heating bell furnace. The elements can be mounted either in the insulation wall of the external bell (as with traditional methods) or directly onto the external wall of the process vessel. This is a more efficient method of preheating the process chamber and allows for better heat transfer into the process vessel (Fig. 6). Operation of the hot-wall furnace involves these steps: • • •

First, the process area is loaded onto the furnace hearth. The process chamber and bell are closed, thus sealing the process chamber from the atmosphere. The chamber is then evacuated by the mechanical vacuum pumping system down to the appropriate vacuum level.

Fig. 5

Voltage versus current characteristics (a Paschen curve) for different types of discharge in argon

Chapter 9: Ion Nitriding Equipment / 97

Fig. 6 •

•

•

•

Arrangement of heating elements for hot-wall furnaces with corresponding temperature profiles. Length in mm. Courtesy of Plateg GmbH

The process chamber is backfilled with hydrogen gas, and the external heating elements are switched on to heat the gas; the workload is heated by convection. Once the process temperature has reached approximately 230 °C (450 °F), the sputter cleaning procedure begins. Hydrogen ions atomically blast the workpiece surfaces, a more thorough and effective method than aqueous cleaning. Holding time at the sputter clean temperature is determined by the initial surface cleanliness but generally is not more than 20 to 30 min. If the workpiece surfaces are badly contaminated, a gaseous mixture of up to 10% argon and 90% hydrogen should be used. To avoid surface etching, do not use more than 10% argon in the mixture. Sputter cleaning is further discussed later in this chapter. Once the sputter cleaning operation is complete, the process chamber temperature is raised to the appropriate nitriding temperature for the required holding time and with the appropriate gas flow to achieve the required surface metallurgy.

The primary argument against use of the hot-wall system is that it requires energy to heat the process retort, and that this expenditure is unnecessary. However, no matter which surface hardening process is chosen, be it nitriding or carburizing, either a process retort or a process chamber must be heated. In plasma nitriding, it is necessary to heat both

98 / Practical Nitriding and Ferritic Nitrocarburizing

the process retort and the water contained in the water jacket if a continuous dc system is used. In the hot-wall pulsed plasma system, the energy required provides external heating to the process retort and conductive heat to the workload. This means that only the minimal energy necessary to generate the plasma glow seam is required. The continuous dc system, however, requires energy from the plasma and supplemental internal elements.

Pulsed Power Supply The power source used in the pulsed plasma nitriding system is the heart of the system. It must provide the operator accurate control of pulsing time and required process voltage in order to create the plasma necessary for surface preparation and nitriding completion with regard to both the steel and the part geometry. The plasma generator supply must: • • • •

Create the physical conditions for the abnormal glow discharge (see the Paschen curve in Fig. 5) Provide good temperature uniformity within the workload area Heat the workload Prevent arc discharge conditions

With a conventional dc power supply, fulfillment of the above conditions is somewhat limited simply because the first three conditions are linked together. When the user has a mixed load of different part geometries and sizes, it becomes difficult to handle because small parts with a high surface area-to-volume ratio can quickly overheat. The minimum power input necessary for the abnormal glow discharge has to be balanced. For example, cooling of the chamber wall may create nonuniform temperatures. The cold-wall unit is usually cooled by a water jacket. This means that heat generated in the process chamber is taken away through the water to a heat exchanger. This is a wasteful method of cooling. The hot-wall system uses a series of external blowers that draw in shop air to cool the external wall of the process vessel as required (Fig. 7). Arc formation must be detected and interrupted as soon as possible. The time between development of the arc and its discharge is a matter of milliseconds, and it will damage the workpiece through rapid overheating at the point of contact. The steel may burn, or the arc discharge can cause localized grain growth and sometimes stock removal. Figure 8 schematically shows the power characteristics of a dc power supply. The gap of the abnormal glow discharge is between the lines Pmin and Pmax, and the power of the plasma, Pplas, must be between these values. The area below the line Ptemp is equal to the energy input necessary to balance the energy losses of the system and to hold the temperature in the workload at the desired value.

Chapter 9: Ion Nitriding Equipment / 99

Fig. 7

Hot-wall plasma nitriding furnace. Arrows indicate the air blowers that cool the external process vessel wall. Courtesy of Plateg GmbH

A better solution for fulfilling the plasma conditions described earlier is to use a pulsed dc power supply with the following special conditions: •

•

• •

The power should not be of a sine-wave type. That means the power will only be reduced and not completely shut off or isolated. It is thus necessary to have a defined power-on, power-off system—that is, a defined square form so that the power jumps from zero voltage into the allowed gap of the abnormal glow discharge region. The length of the pulse should be shorter than the development time of the arc. This means less than 100 µs so that the arc is suppressed. Interruption of the arc will be possible during each pulse. The pause that follows each pulse should be short enough to allow an easy ignition for the next pulse—for example, less than 1 µs. The ratio of pulse to pause should be variable over a wide range to control the power input by the plasma in the workload so that it will be possible to use an auxiliary heating system (such as external heating) for better temperature within the process chamber.

With the pulsed dc system, the point at which Pplas occurs is only during the pulse within the gap. The sum of the areas under each pulse is equal to area under the line Ptemp. In the figure the pulse time (the width of the square wave) is constant. The pause time can vary to balance the energy losses Ptemp. By this method, the temperature adjustment is separated from

100 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 8

Power characteristics (a) Continuous dc. (b) Pulsed dc. Abnormal glow discharge (see Fig. 5) occurs between Pmin and Pmax. The power required for plasma nitriding, Pplas, is in this region. Ptemp is the time average of power required to maintain the workload at the desired temperature. In (b) the pulse widths are regulated so the area under the pulses equals the area under Ptemp. Source: Ref 5

the other process parameter. Typical pulse time values are between 5 and 100 µs, while pause time can vary between 5 to 200 µs (Ref 5). Pulsed dc power gives the furnace user three additional process control variables: • • •

Process voltage Time of power on Time of power off

The sum of the on time and off time of one pulse makes up one cycle, so the control parameters could be described as voltage, frequency, and ratio

Chapter 9: Ion Nitriding Equipment / 101

of on time to off time. This amounts to variable pulsed dc power in relation to time on and time off, as well as the ability to vary the chamber internal pressure. These additional controllable variables allow the operator to better control the process and to address such part geometries as blind holes, cavities, and complex shapes. Occurrences such as arc discharge and hollow cathode potential (discussed later in this chapter) are still a concern, but not as great as with the continuous dc system (Ref 5). The process controller will now control: • • • • • • • • • • • • • • •

Furnace temperature Part temperature Part temperature ramp rate to process temperature Plasma process voltage Partial pressure control Current density Nitrogen gas flow Hydrogen gas flow Methane gas flow (if used) Argon gas (if used) Oxygen gas (if used) Process cycle time Pulse power time on Pulse power time off Cooldown rate

Because of the numerous process variables available, greater use can be made of PC/PLC computer technology. Considering the state-of-the-art communication technology in industry today, the complete furnace system and any furnace problems can be remotely monitored, saving costly downtime, and enabling the equipment supplier to better serve the user, and complying with International Organization for Standardization (ISO) specifications for process and record keeping. In addition, the system is almost completely self-diagnostic and self-managing and is able to compare current process information with historical data.

Work Cooling after Plasma Nitriding There are five methods of work cooling from the process temperature to an acceptable exposure temperature after plasma nitriding. Selection depends on the primary design of the furnace system. The cooling methods described in this section apply to both cold-wall and hot-wall furnaces. Free-Cool Method. Free cooling is achieved by simply turning off the plasma power and allowing the work to cool from the process temperature down to below 150 °C (300 °F) under partial pressure conditions. The time required depends on workpiece mass and surface area. The disadvantage of this method is that it is very inefficient. There is insufficient

102 / Practical Nitriding and Ferritic Nitrocarburizing

process gas within the furnace chamber under vacuum conditions to allow effective cooling with convection methods of heat transfer. Cooling under Partial Pressure and Convective Gas Conditions. This method involves shutting off the plasma power and backfilling the chamber to a partial pressure with a nonreactive gas—usually clean, dry nitrogen (not metallurgical grade). The work will cool down somewhat faster than with the vacuum free-cooling method because there is a convection gas within the process work chamber. However, there is no means of distributing acquired heat from the workpiece into heated gas and discharging it to the atmosphere. Cooling under Positive Pressure. This method of cooling relies on sufficient nitrogen being introduced into the work chamber so that the chamber pressure exceeds atmospheric pressure. Gas agitation is now by an internal recirculation fan. This method is also more efficient than the free-cooling method. Cooling Using a Combination of Nitrogen Backfilled Gas in Conjunction with a Water-Cooled Heat Exchanger. This method of cooling is the most efficient method available to the plasma nitride furnace user. The furnace is fitted with a finned-tube copper heat exchanger with ambient-temperature water passing through the heat exchange coils as the backfilled gas passes over the heat exchanger. Movement of the cooling gas is made possible by the internal recirculation fan. Postoxidation Treatment. The fifth option is to cool down to a specific temperature and conduct a postoxidation treatment (oxynitriding) to enhance the surface corrosion properties of the workpiece. Postcooling. Once cooling has taken place, the process chamber can then be opened safely without risk of oxidation or discoloration of the work surface. The workpiece should be a matte gray color. However, if a postoxidation treatment has been carried out, the color of the workpiece surface will be blue to almost black, depending on the postoxidation treatment temperature.

Other Considerations for Ion Nitriding Equipment and Processing Other important considerations associated with ion nitriding include the hollow cathode effect, sputter cleaning, furnace loading, pressure/voltage relationships, workpiece masking, and furnace configuration options.

Hollow Cathode Hollow cathode is an area of low vacuum pressure where the plasma glow seam does not follow the precise contour of the part being treated. For example, if the plasma glow seam dips slightly into a blind hole, free electrons (energy) are trapped in the area beneath the glow seam. The glow seam holds the free electrons within the hole. The electrons as energy then begin to migrate through the wall of the hole. This raises the temperature of

Chapter 9: Ion Nitriding Equipment / 103

the steel, which can lead to overheating and sometimes burning (Fig. 9). If the parts are placed too close together, a similar effect can take place and cause localized overheating or sometimes burning (Fig. 10).

Supplementary Internal Heating Elements Supplementary internal heating elements are sometimes used in the cold-wall system. However, it must be remembered that everything within the vacuum chamber will be nitrided, including the heating elements. The hot-wall furnace operates in a completely different manner initially. The furnace is evacuated by the vacuum pump to the appropriate vacuum level and backfilled with hydrogen to a partial pressure. The external hot-wall heaters are started and the part is heated by convection gases and not by plasma energy. At ambient temperatures and high plasma

Fig. 9

Illustration of trapped free electrons in a blind hole having the potential to overheat the corners

Fig. 10

Parts too close together cause the “hollow cathode” effect, leading to possible overheating and burning.

104 / Practical Nitriding and Ferritic Nitrocarburizing

generation voltages, plasma energy is dangerous, as is the case with the cold-wall system. With the hot-wall system, when the temperature reaches approximately 230 to 260 °C (450 to 500 °F), the plasma generation voltage is started and the part is then sputter cleaned under hydrogen gas up to the appropriate nitriding temperature.

Sputter Cleaning Sputter cleaning can be likened to “atomic shot blasting,” that is, cleaning by ionic bombardment. This is the procedure used to preclean the work surfaces prior to nitriding. The sputter gas typically is hydrogen, the lightest gas, which cleans as well as acts as a reducing gas. Any surface oxide will be reduced by the hydrogen to the base metal. If the workpiece surface is seriously contaminated, the hydrogen can be mixed with argon to increase the gas density and its cleaning ability. Again, apply caution when using argon to ensure that the mixture is not so severe as to cause surface etching of the steel component (maximum ratio 90% hydrogen, 10% argon). The sputtering time depends on the prior surface condition of the steel being treated. Generally, one would introduce several temperature steps above the initial convective heating at 230, 370, and 450 °C (450, 700, and 850 °F), then up to the final selected nitriding temperature. The holding time at the selected sputter cleaning stage would be approximately 10 min (depending, of course, on how much cleaning is necessary). Note that as the temperature increases, the sputter cleaning continues (Fig. 11).

Fig. 11

Time versus temperature curve illustrating the sputter cleaning process as the temperature increases to the processing temperature

Chapter 9: Ion Nitriding Equipment / 105

Power Source for Sputter Cleaning. A power unit such as one that generates a pulsed dc voltage is used. Remember that with pulse processing, two variables are introduced: variable voltage and variable pulse time. These can be adjusted in relation to the steel part geometry.

Loading of the Furnace Furnace loading is not complicated if the operator has a basic understanding of the process. The furnace can be loaded with a static fixture that will accommodate the particular parts to be nitrided. This assumes that the parts are of similar geometry and similar case-depth requirement. The real secret of parts loading is to be aware of the potential for the hollow cathode effect. However, the part geometry can be mixed, provided that the workpiece materials are similar in both chemical analysis and case-depth requirements. If different case-depths are required and the steel analyses are radically different, it is not possible to mix the load.

Pressure in Relation to Voltage It is most important to have good control over both the vacuum level and the voltage as these two process parameters will strongly influence the plasma glow-seam coverage of the part (or, more accurately, they will determine the point of plasma ignition on the part). If the pressure is too high with a normal process voltage for the part geometry, the part will begin to lose the plasma glow-seam coverage. If this occurs, the area without plasma ignition will not nitride (Fig. 12a and b).

Masking A part can be masked quite simply by remembering the following: What plasma can see, it will nitride. Mechanical masking involves wiring steel shim stock material to the part. The thickness of the shim stock is usually about 0.05 to 0.1 mm (0.002 to 0.004 in.). If the part is placed on the furnace hearth, the side of the part that is in contact with the hearth will not nitride. If it is necessary to nitride its underside, then the part can be mounted on steel points (Fig. 13). If two parts are placed directly on top of each other, the two contacting faces will not nitride and will remain soft. Holes can be masked by inserting a simple steel plug. For a threaded hole, the first two or three turns of a stud can be screwed into it. There is no need to coat the stud threads. Proprietary “paints” are available for masking that will resist the effects of sputtering (unlike early paints that sputtered off the surface in particles that made their way to the inner vessel wall, changing the electrical characteristics of the anode over time). Nitriding stop-off paints will resist the effects of sputtering if the proper application methods are followed. Complete coverage is mandatory, and there should be no brush marks. Copper plating is commonly used to mask workpieces undergoing traditional gas nitriding. In plasma nitriding, however, the deposited plate can

106 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 12

Glow-seam coverage dependence on voltage. (a) Workpiece during plasma nitriding with continuous dc glow discharge in the normal region (see Fig. 5). The total dc power input is not high enough to cover the complete workpiece surface. Only the areas covered by the glow will be nitrided. Nitriding at the uncovered areas will be reduced or absent. (b) Workpiece during nitriding with pulsed dc glow discharge. By using the pulsed dc with a repetition frequency of about 10 kHz, the complete surface is covered and uniform nitriding results. The average power input is the same as in (a). The peak power of each pulse is higher so that the region of an abnormal glow (Fig. 5) is applied during this pulse. Pulsed technology allows complete coverage of the surface with high peak powers, but low average power input, so that workpiece overheating can be avoided. Courtesy of Plateg GmbH.

be sputtered off during the cleaning process and transported to the anode vessel, where they will coat the inner surface and affect its electrical characteristics. Thus, copper plating is not recommended for plasma nitriding. Additional information on masking prior to plasma nitriding operations can be found in Chapter 15, “Stop-Off Procedures for Selective Nitriding.”

Configurations of Plasma Nitriding Units Plasma nitriders generally are built in the vertical configuration (Fig. 14), though some are built in the horizontal or pit-type configuration. The choice depends on: • • • • •

Available floor space Available roof height Production requirements Part geometry Available furnace design

Chapter 9: Ion Nitriding Equipment / 107

Fig. 13

Workpiece in furnace. (a) Component mounted on support points. (b) Detail showing the effect of support on local case formation

Fig. 14

Vertically configured plasma system. This system has two chambers or bells, so one can operate while the other is being loaded or unloaded. Courtesy of Plateg GmbH

Summary: Advantages of Plasma Nitriding Plasma nitriding, particularly pulsed plasma nitriding, has become a mature technology capable of processing workloads efficiently and repeatably. Here are some of the advantages of the pulsed plasma process: •

Environmental benefits: The process is clean and nontoxic and produces no disposable effluent.

108 / Practical Nitriding and Ferritic Nitrocarburizing

•

• •

•

• •

•

•

•

No fire risk: Hydrogen, used for sputter cleaning and as a dilution gas, presents no fire risk whatsoever because of the minuscule amounts of oxygen remaining within the process chamber under vacuum conditions. No obnoxious smells: Neither nitrogen nor hydrogen will produce any offensive smells, nor any adverse skin reactions. Minimal distortion: Because the process can control the type of surface metallurgy created on the workpiece, compound zone thickness can be better controlled, leading to lower overall size growth. In addition, better advantage can be taken of the lower process temperatures. The process temperature will act as a stabilizer, or an additional tempering procedure. Thus any retained austenite that might be present as a result of the prehardening and tempering procedure will be decomposed, leading to better dimensional stability. Clean work: The work surface is usually very clean. However, oxygen from the air can discolor the work surface if the vessel is opened at too high a temperature. The work surface can also be discolored by an overly intensive sputter-cleaning program. Repeatable results: Consistent and repeatable results can be achieved that suit the application. Elimination of nitride networks: Problems associated with nitride networks can be overcome by manipulating the nitriding potential. This is achieved simply by reducing the amount of required nitrogen for the process. Process management: The system is almost self-managing, taking full advantage of PC/PLC combinations for better control. Preventive maintenance can be better planned, and remote troubleshooting and process control are possible. If a power failure occurs, the system can be programmed to restart when power resumes. Operating costs: Plasma nitriding process requires minimal operator supervision (load/unload/program/initiate), provides good utilization of floor space, and results in reduced energy costs because of shorter cycle times compared to traditional nitriding. The capital cost of the equipment is higher than for gas nitriding but is offset by better plant utilization due to faster process cycles and more repeatable metallurgy. Integration into cell manufacture: The pulsed plasma nitride furnace can be integrated into the manufacturing line. Figure 15 shows a precision gear manufacturing facility that has successfully integrated heat-treatment production requirements into the gear cutting line. A variety of small helical and spiral gear pinions and planetary gears are nitrided. This system can also be built into a robotic handling system, completely automating the process line (though only with a single product line, generally automotive components). Such automated systems can considerably reduce operating costs.

Chapter 9: Ion Nitriding Equipment / 109

Fig. 15

Single bell unit suitable for integration into a gear manufacturing production line. Courtesy of Plateg GmbH

REFERENCES 1. D. Pye, Nitriding Techniques and Methods, Steel Heat Treatment Handbook, G.E. Totten and M.A.W. Howes, Ed., Marcel Dekker, 1997, p 744 2. N. Harris, Oil Sealed Mechanical Pumps, Modern Vacuum Practice, McGraw Hill, 1989, p 71 3. N. Harris, Oil Sealed Mechanical Pumps, Modern Vacuum Practice, McGraw Hill, 1989, p 72 4. D. Pye, “Practical Nitriding” course notes, 1986 5. R. Gruen, Pulse Plasma Treatment: The Innovation for Ion Nitriding, Ion Nitriding Proceedings, ASM International, 1987, p 143–147

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p111-118 DOI: 10.1361/pnafn2003p111

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

10 Nitriding in Fluidized Beds

A FLUIDIZED-BED FURNACE SYSTEM can be used for the gas nitriding process. The fluidized-bed furnace is a unique metallurgical processing tool that enables the user to complete most heat treatment processes, including surface treatments. The discussion here focuses on fluidized-bed nitriding. Previous chapters have discussed gas nitriding, salt bath nitriding, and plasma ion nitriding. Each of these procedures is conducted at a process temperature of approximately 500 °C (930 °F). The fluidized-bed furnace uses ammonia gas for its nitrogen source, whereas the salt bath uses cyanide. Fluidized-bed nitriding is similar in process technique to gas nitriding and similar in the method of heat transfer to salt bath nitriding. The fluidized bed exhibits the same characteristics as a liquid, with one exception: A fluidized bed is not wet. The technique of fluidization involves the disturbance of a bed of finely divided particles, which behave as a liquid would behave. This is accomplished by passing a gas at sufficient volume and pressure so as to separate microscopically the fine particles. If the volume and pressure of the gas are too great, then the fine particles will be carried in the gas stream and leave the bed. Thus, gas pressure and volume are critical. The bed does not require large volumes of gas, only that necessary to separate the particles (Ref 1) (Fig. 1). The particles in this instance are aluminum oxide. Good heat transfer takes place from the heating medium to the aluminum oxide particles, which in turn transfer the heat to the workpiece. When the bed is in operation, its surface of finely divided aluminum oxide particles bubbles just like water bubbles when air passes through it (Fig. 2).

Heating Method The methods of heating a fluidized bed are very similar to those used to heat an atmosphere-type furnace or a salt bath. The heating system can be electrical or gas, and internal or external.

112

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Various types of contact in fluidized beds. Source: Ref 1

Fig. 2

Liquid-like behavior of gas fluidized beds. Source: Ref 1

External Resistance Heating. The aluminum oxide is usually contained in the heat-resisting material located inside the furnace casing, with the heating elements mounted onto the furnace installation material (Fig. 3). Several gases can be used to fluidize the aluminum oxide particles: • • •

Nitrogen Methane Endothermic gas

• •

Ammonia Gas mixtures such as methane and nitrogen or ammonia and methane

Internal Resistance Heating. This method of heating (Fig. 4) is very similar to that of the immersed electrodes in a salt bath. The difference in this instance is that the internal heating elements are usually sheathed. It is a very simple method of heating the fluidized bed and provides good heat transfer between the heating elements and the aluminum oxide. Gas-Heated Fluidized Beds. Gas heating is usually less expensive than electrical heating (depending, of course, on locality and geography).

Chapter 10: Nitriding in Fluidized Beds

Fig. 3

External resistance heating. (a) Fluidized-bed furnace with external heating by electrical resistance elements: (1) pivoting cover in two parts; (2) insulation; (3) refractory material; (4) fluidized bed; (5) resistance elements; (6) intake for fluidized gas (air or nitrogen); (7) parts to be treated. (b) Recirculation of fluidizing gas in a fluidized-bed externally heated by electrical resistance. Source: Ref 1

Once again, heating of the fluidized-bed process chamber can be either external or internal. External Combustion Heating. When the fluidized-bed container is heated by an external gas combustion system, the container is manufactured from the same heat-resisting material as it would be if manufactured

/ 113

114

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 4

Fluidized-bed furnace with internal heating by electrical resistance elements: (1) pivoting cover in two parts; (2) insulation; (3) refractory material; (4) fluidized bed; (5) heating elements; (6) intake for fluidizing gas; (7) parts to be treated. Source: Ref 1

for an electrical heating system (Fig. 5). The fluidized-bed container is contained within the furnace combustion chamber, which in turn is surrounded by the installation material and the furnace casing. Many different types of combustion burners can be utilized for the heating system, including recuperator type (Fig. 6).

Nitriding in the Fluidized-Bed Furnace The ability of the fluidized bed to recover to the process temperature after the workload has been removed from the furnace and a new load introduced makes it an attractive choice for nitriding. The bed does not require conditioning or purging when changing from one atmosphere system to another, making it productive and economical in terms of operating costs and throughput. The fluidized bed also permits easy change from one process system to another (e.g., from nitriding to ferritic nitrocarburizing, or from carburizing to carbonitriding) (Ref 2). The one major disadvantage is that it requires a fairly high volume of reactive gas to complete the process. Temperature control is exactly the same as for a conventional heat treating furnace. The bed temperature uniformity is usually well within ±5 °C (10 °F). Atmosphere control, particularly for nitriding where gas dissociation is critical, is somewhat difficult. Fluidized-bed nitriding tends to rely

Chapter 10: Nitriding in Fluidized Beds

Fig. 5

Immersed-element ceramic retort fluidized-bed. Source: Ref 1

Fig. 6

External gas-heated fluidized bed with recuperator. The use of regenerative burners where the exhaust gas temperature is only 200 °C (390 °F) achieves efficiencies similar to those of electrically heated furnaces. Source: Ref 1

on volume of ammonia gas in relation to workpiece surface area. Because the atmosphere gas control is variable, the compound layer can be controlled—a critical factor in surface performance. As with the conventional ammonia gas nitriding furnace, many grades of steel can be processed effectively using the fluidized-bed technique. The types of steels that can be treated, including all the stainless steel grades, will be described in Chapter 12. Stainless steels particularly require good gas flow control and the appropriate gas dissociation. The fluidized process cycle times, as in the gas nitriding process, are governed by the laws of diffusion. In other words, the diffusion rate of nascent nitrogen into the steel surface is the same for fluidized-bed nitriding as

/ 115

116

/ Practical Nitriding and Ferritic Nitrocarburizing

it is for ammonia gas nitriding. The floor-to-floor time is quicker for fluidized-bed nitriding due to the faster initial recovery time (Fig. 7). Figure 8 shows typical processing times for nitride case depths.

Oxynitriding Oxynitriding can be accomplished in the fluidized-bed furnace much like it is accomplished in gas nitriding. This means that on completion of the nitriding cycle, controlled amounts of moisture are added to the process chamber. The oxynitriding process forms a very thin surface oxide layer on the immediate surface of the workpiece. This deliberately oxidized surface layer is resistant to some aspects of corrosion (though not all). In general, the process is used for applications where an expensive material such as stainless steel is being replaced by low-carbon steel with an enhanced surface condition. The oxynitriding process is gaining in popularity, particularly in the automotive industry, in both Europe and the United States.

Fig. 7

Heating rates to 1000 °C (1830 °F) for cylinders of varying diameter in different types of heat treatment furnaces. Source: Ref 1

Chapter 10: Nitriding in Fluidized Beds

Fig. 8

Total nitride case depth versus time in a fluidized bed at 525 °C (975 °F). Source: Ref 1

Operating the Fluid Bed for Nitriding Startup of the fluid bed should follow the furnace operation manual of the furnace manufacturer. Once the fluidized bed has been brought to the process temperature, the bed is activated with the process gas, or a gas with a sufficiently high flow rate to “unlock” the bed of aluminum oxide. This means that the bed is no longer “slumped,” but activated and bubbling. Great care must be taken not to have the gas flow rate so high that it will blow the aluminum oxide out of the bed. Any previous atmosphere in the bed (e.g., carburizing) can be changed over quickly to nitriding. This is a major advantage of using a fluidized bed instead of a salt bath. The only changeover necessary is the process gas, making the fluidized-bed furnace more productive and much easier to handle. The least amount of process gas necessary to complete the nitriding should be used. Fluidized-bed nitriding is no faster than conventional gas nitriding, because both use a gas nitriding furnace. The major benefit is that the floor-to-floor time is faster, since the bed recovery time is shorter than with a conventional furnace. The resulting metallurgy of the two processes should be the same (Tables 1, 2).

Measurement of the Gas Dissociation Measurement of the gas process dissociation is not the same as for gas nitriding. The gas dissociation measurement of ammonia in the conventional gas nitriding system is based on the solubility of ammonia in water. Water will absorb 70 times its own volume of ammonia, providing an effective method of measuring the gaseous activity within the nitriding process chamber. With a fluidized bed, the method is effective only if the exhausted ammonia can be captured and the dissociation measured. Such measurement is

/ 117

118

/ Practical Nitriding and Ferritic Nitrocarburizing

Table 1

Recommended fluidized-bed nitriding procedures Recommended time

Material

Carbon and low-alloy steels Tool and die steels (structural) Tool steels (cutting) Corrosion- and heat-resistant steels Ductile, malleable, and gray cast iron Powder metal products (ferrous)

min

max

1h 30 min 5 min 1h 1h 30 min

2h 3h 1h 2h 4h 2h

Temperature °C

°F

580± 5 540– 580 540– 580 580± 5 580± 5 580± 5

1075 ±10 1000 –1075 1000 –1075 1075 ±10 1075 ±10 1075 ±10

Source: Ref 1

Table 2

Depth of compound layer after fluidized-bed nitriding Case depth mm

Material

Carbon and low-alloy steels Tool and die steels (structural) Tool and die steels (cutting) Corrosion- and heat-resistant steels Ductile, malleable, and gray cast iron Powder metal products (ferrous)

min

0.0038 0.003 0.003 0.0038 0.0038 0.0038

in. max

0.03 0.013 0.03 0.03 0.0

min

0.00015 0.0001 0.0001 0.00015 0.00015 0.00015

max

0.001 0.0005 ... 0.001 0.001 0.001

Source: Ref 1

based on process gas flow in relation to work surface area being treated. This means that if the work surface area is not constant (as in a typical commercial heat treatment shop), and if the gas flow rate remains constant for each load of work processed, then the gas dissociation will be different for each load. Studies on various materials and various load surface areas must be made to ensure a reasonable chance of consistent and repeatable nitriding. ACKNOWLEDGMENT Grateful acknowledgment is given to Ray Reynoldson of Quality Heat Treatment Pty Ltd., Turbo Drive, North Bayswater 3153, Australia, for his valued assistance with this chapter. REFERENCES 1. R.W. Reynoldson, Theory and Practice, Heat Treatment in Fluidized Bed Furnaces, ASM International, 1993, p 3–9 2. C. Dawes and D.F. Tranter, Nitrotec Surface Treatment Technology, Heat Treat. Met., Vol 12 (No. 3), 1985, p 70–76

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p125-138 DOI: 10.1361/pnafn2003p125

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

12

Steels For Nitriding MANY STEELS are commercially nitrided: • • • • • • • •

Aluminum-containing low-alloy steels, including the Nitralloy group with 1% Al Medium-carbon, chromium-containing low-alloy steels of the 4100, 4300, 5100, 6100, 8600, 8700, and 9800 series Hot-work die steels containing 5% Cr such as H11, H12, and H13 Air-hardening tool steels such as A-2, A-6, D-2, D-3, and S-7 High-speed tool steels such as M-2 and M-4 Austenitic stainless steels of the 200 and 300 series Martensitic stainless steels of the 400 series such as 422 and 440 Precipitation-hardening stainless steels such as 13-8 PH, 15-5 PH, 17-4 PH, 17-7 PH, A-286, AM350, and AM355

Table 1 lists the compositions of some typical nitridable steels.

Steel Selection Considerations One of the most difficult tasks in nitriding is to select a steel in relation to the operating environment of the part that will not only ensure good nitriding results but will also be cost effective, easy to machine or fabricate, and functional. Several questions must be addressed during the selection process: • •

• • •

What is the product to be manufactured and how complex is the part geometry? Under what type of operating conditions will the workpiece operate? Will there be compressive loads, cyclic loads, impact loads, or tensile loads? Will the workpiece operate under abrasive conditions? Will the workpiece operate under corrosive conditions? What will be the operating temperature of the part? Will it be in a hot or cold environment?

126

/ Practical Nitriding and Ferritic Nitrocarburizing

Table 1

Compositions of selected nitridable steels Composition, %

Alloy steels(a)

SAE 4137 SAE 4142 SAE 4140 SAE 4150 28 Ni Cr Mo V 85 32 Ni Cr Mo 145 30 Cr Ni Mo 8 34 Cr Ni Mo 6 SAE 4337 SAE 4130

C

Cr

Mo

Si

Mn

Ni

V

0.35 0.42 0.40 0.5 0.3 0.32 0.3 0.34 0.38 0.26

1 1 1 1 1.3 1 2 1.5 0.8 1

0.2 0.2 0.2 0.2 0.4 0.3 0.4 0.2 0.4 0.2

0.25 ... 0.25 ... ... ... ... ... ... ...

0.8 ... 0.85 ... ... ... ... ... ... ...

... ... ... ... 2 3.3 2 1.5 1.5 ...

... ... ... ... 0.1 ... ... ... ... ...

Low-alloy steels

C

Si

Mn

P

Cr

Mo

Ni

V

Al

Nitralloy Nitralloy M Nitralloy 135 Nitralloy 135M

0.20–0.30 0.30–0.50 0.25–0.35 0.35–0.45

0.10–0.35 0.10–0.35 0.10–0.35 0.10–0.35

0.40–0.65 0.40–0.80 0.65 max 0.65 max

0.05 max 0.05 max 0.05 max 0.05 max

2.90–3.50 2.50–3.50 1.40–1.80 1.40–1.80

0.40–0.70 0.70–1.20 0.10–0.25 0.10–0.25

0.40 max 0.40 max 0.40 max 0.40 max

... 0.10–0.30 ... ...

... ... 0.90–1.30 0.90–1.30

C

Si

Mn

Cr

Mo

Ni

V

W

1.45 0.55

... ...

... ...

0.3 1.1

... 0.5

... 1.7

0.3 0.1

3.0 ...

C

Si

Mn

Cr

Mo

Ni

V

W

1.55 2.0 1.0 0.95 0.9 2.1 1.65 0.59

... ... ... ... ... ... ... ...

... ... ... 10.5 20.4 ... ... ...

11.5 12.0 5.0 ... ... 11.5 11.5 1.1

0.8 ... 1.0 ... ... ... 0.6 ...

... ... ... 0.1 0.2 ... ... ...

1.0 ... 0.2 0.5 ... 0.2 0.1 0.2

... ... ... ... ... 0.7 0.5 1.9

Special-purpose tool steels

F2 L6 Dimensionally stable tool steels(b)

D2 D3 A2 O1 O2 D6 D2 S1 Hot-work tool steels

H12 H13 H11 H21 H19 H10 High-speed steels

T5 T4 T1 T15 M42 M41 M3 M2 M2 M7 M1

C

Cr

Mo

Ni

V

W

Co

0.36 0.4 0.4 0.3 0.4 0.32

5.2 5 5 2.7 4.3 2.8

1.4 1.3 1.3 ... 0.4 2.8

... ... ... ... ... ...

0.4 1 0.6 0.4 2 0.5

1.3 ... ... 8.5 4.3 0.3

... ... ... ... 4.3 ...

C

Cr

Mo

V

W

Co

0.75 0.8 0.75 1.5 1.08 0.92 1.2 0.87 1.0 1.0 0.83

4 4 4 5 4 4 4 4 4 4 4

0.6 0.7 ... ... 9.5 5 5 5 5 8.7 9

1.6 1.6 1 5 1.2 1.8 3 1.8 1.8 2 1.2

18 18 18 12.5 1.5 6.5 6.5 6.5 6.5 1.8 1.8

9.5 5 ... 5 8 5 ... ... ... ... ...

(a) These are typical alloy steels that will gas or salt bath nitride. (b) The core hardness will diminish in these steels if a low tempering temperature is used during the preharden and temper operation.

• •

Will there be adequate part lubrication? Is further machining after nitriding a consideration? For example, will the part undergo grinding, lapping, or polishing?

After the engineer has gathered the necessary information, the search for the appropriate steel can begin. If several steels are suitable for a particular application, price and availability become additional considerations (Ref 1).

Chapter 12: Steels for Nitriding / 127

Requirements for a Nitriding Steel In the early years of nitriding, Adolph Fry at Krupp Steel (Ref 2) recognized that certain steels responded better from a metallurgical standpoint in terms of surface hardness, core hardness, distortion, cycle time at temperature, and the formation of stable nitrides. Fry discovered that certain elements will respond more readily than others to form stable nitrides during the nitriding process, and this led to the development of the Nitralloy group of steels (see Table 1 for compositions). Of the alloying elements commonly used in commercial steels, aluminum, chromium, vanadium, tungsten, and molybdenum are beneficial in nitriding because they form nitrides that are stable at nitriding temperatures. The effects of specific alloying elements are discussed later in this chapter. Figures 1 and 2 show the influence of alloying elements on hardness after nitriding and depth of nitriding. Aluminum will form very hard nitrides in the nitrided steel surface. Generally the maximum amount of aluminum permitted in the steel is in the region of 1.5%. Above 1% Al will lead to surface cracking under extreme surface load conditions. This is because the core hardness of the material is usually very ductile. If a highly ductile workpiece undergoes severe loading, then there is a strong possibility that the surface of the case will lead to crack propagation.

Fig. 1

Effect of alloying elements on hardness after nitriding. In steels containing several alloying elements, higher hardness values are obtainable than if alloying elements are used separately. Base alloy: 0.35% C, 0.30% Si, 0.70% Mn. Source: Ref 3

128

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 2

Effect of alloying elements on depth of nitriding measured at 400 HV. Nitriding was carried out at 520 °C (970 °F) for 8 h. Source: Ref 3

Molybdenum will form stable nitrides at the nitriding temperature and will reduce the risk of surface embrittlement at the nitriding temperature. Chromium will also form stable nitrides at the nitriding temperature; however, the high chromium content found in some stainless steels makes them more difficult to nitride. Chromium reacts with oxygen to form a chrome oxide barrier on the surface, which must be broken down by depassivation in order for nitriding to be effective. The higher the percentage of available chromium at the steel surface, the more difficult the steel will be to nitride. The positive side of this is usually high surface hardness values. Vanadium in a nitriding steel also is conducive to the formation of stable nitrides. In addition, fine grain toughness will be exhibited within the formed case. Tungsten enables the steel to retain its hardness at high operating temperatures with no loss of surface hardness. Depending on the tungsten content and the general composition, the nitrided steel is able to operate at temperatures up to 590 °C (1100 °F) with enhanced wear characteristics and no appreciable loss of surface hardness. Silicon is considered to be a good nitride former. Though it is usually present as either an oxidizer or a stabilizer, silicon generally is not of sufficient volume to be considered a strong nitride former. Summary. As stated in Chapter 1, all steels will nitride. Steels that contain the above alloying elements will readily form stable nitrides. Steels that do not contain those elements, such as the mild steels and low-carbon

Chapter 12: Steels for Nitriding / 129

steels, will also nitride but will have lower surface hardness because the case formation is limited to pure iron nitride. However, the corrosion resistance of low-carbon steels will be greatly enhanced, successfully withstanding a minimum 100 h salt spray test. Studies have shown that with a deep case (90 h cycle on gas nitriding), low-carbon steels exhibited resistance to 20% salt spray solution up to 150 h.

Can Stainless Steels Be Nitrided? In general, most stainless steels can be nitrided but with some adverse effects on corrosion resistance. Hardness values are generally in the range of 1000 HV or more, depending on the nitriding method. With ion nitriding and control of the nitriding potential, some hardness values greater than 1400 HV have been achieved. Austenitic stainless steels are perhaps the most difficult to nitride. The following types have been successfully nitrided using gas, plasma, and salts: 301, 302, 303, 304, 305, 309, 310, 316, 321, and 347. However, when nitriding these steels at conventional nitriding temperatures, corrosion resistance is seriously impaired, in some cases up to 1000%. The material should be in the annealed condition, which will reduce the risk of blistering or flaking. The oxide film must be removed prior to nitriding, or it will form a barrier that will be difficult to decompose. Removal can be accomplished by wet or vapor blast pickling solution, molten salts, or sputter cleaning. One technique is to use a chloride-based solution as a cleaning agent. Once depassivated, the surface should not be touched by hand; deposition of body oil from fingerprints will inhibit the nitriding effect at the point of contact. Remember, if the core is annealed, the surface can accommodate only abrasion resistance. The core will be too soft to support any kind of constant or cyclical load on the surface. If a load is applied, the case likely will collapse, crack, and flake. Martensitic stainless steels will nitride without exception. This group can be preheat treated to give a supportive core for abrasive and impact applications, as well as torque loads. Once again, remember that the surface must be depassivated before nitriding. Corrosion resistance will be adversely affected but may be protected by using lower-than-normal nitride processing temperatures. However, the lower the process temperature, the slower the diffusion rate, which ultimately means longer cycle time and greater furnace occupancy. Precipitation Hardening Stainless Steels. The same considerations apply to this group of steels as apply to the hardenable martensitic stainless steels. Cycles for Gas Nitriding of Stainless Steels. In general, the singlestage process is used with a process temperature in the region of 490°C (925 °F) (depending on steel composition and preheat treatments), with time at temperature ranging from 24 to 48 h with fairly low dissociation

130

/ Practical Nitriding and Ferritic Nitrocarburizing

rates. The resulting surface metallurgy will be a very shallow or thin compound layer. The cycle times for stainless steels are usually longer than for alloy steels or tool steels.

Example 1: Plasma Nitriding of AISI Type 422 Stainless Steel Surface modification of type 422 stainless steel was accomplished using the pulsed plasma nitriding process. The workpiece design called for an operational service temperature up to 645 °C (1200 °F) and the ability to perform in a highly corrosive and superheated steam environment. The workpiece had previously been nitrided using ammonia gas. Typical composition of type 422 is: Element

Wt%

Carbon Manganese Phosphorus and sulfur Silicon Chromium Nickel Molybdenum Vanadium Tungsten

0.25 1.00 0.025 max 0.7 12 0.8 1.10 0.25 1.1

The same elements that are strong carbide formers (tungsten, molybdenum, and vanadium) are also strong nitride formers. As a result, the material was suitable for nitriding. The preheat treatment condition of the core was tempered martensite that had been hardened and tempered to 340 to 360 HV (~35 to 37 HRC). This was accomplished via the following heat treatment procedure: Process

Preheat 1 Preheat 2 Austenitize Quench Stabilization Temper

Condition

425 °C (800 °F) 760 °C (1400 °F) 1035 °C (1895 °F); 32 mm (1.25 in.) section soaked for 60 min at the process temperature Air cooled Cryogenic treatment using liquid N2 (approximately –70 °C, or –95 °F) 595 °C (1100 °F) for 2 h at the process temperature

The workpiece characteristics that were specified and the actual results of the plasma nitriding process were:

Chapter 12: Steels for Nitriding / 131

Characteristic

Required value

Actual value

Effective case depth, mm (in.) Case hardness, minimum

0.2 (0.008) 960 HV (65 HRC, approximate) 0.005 (0.0002)

0.2 (0.008) 960 HV (68 HRC)

Compound zone maximum γ′, mm (in.)

0.0025 (0.0001)

Plasma nitride process parameters used to achieve these results were: Process

Sputter clean time Sputter clean voltage Sputter retort pressure Process temperature Cycle time Operating pressure Operating voltage Cooling

Parameter

1.5 h 650 V 100 Pa (0.015 psi) 525 °C (975 °F) 16 h at process temperature 300 Pa (0.05 psi) 500 V Partial-pressure N2

Summary Results. The microhardness surveys of four cycles of pulsed plasma nitriding each exhibited similar results with negligible deviations. Previous gas nitriding results had exhibited a surface hardness of 852 HV, which just made the minimum specification requirement. In addition, the surface finish deteriorated from 15 to 35 Ra. With pulsed plasma ion nitriding, the surface hardness increased to 960 HV, a significant improvement (Fig. 3). The surface finish deteriorated approximately 230% using the gas nitriding process, due to a slight roughening of the surface. This necessitated two further grinding operations. The deterioration in surface finish after pulsed plasma ion nitriding was 13.3%. This was within the accepted surface finish requirement, requiring no further machining. Thus, in some instances it is not necessary to grind after plasma nitriding, and the workpiece can be used directly. The part was thermally sensitive to distortion. Precise measurements were taken prior to pulsed plasma nitriding and the growth/distortion was determined to be within the dimensional tolerance levels. The computercontrolled system regulated not only temperature levels, but also current density, process pressure, gas flows, gas ratios, and power levels.

Example 2: Nitriding of AISI Type 440A Stainless Steel A martensitic type 440A stainless steel specimen was nitrided using the pulsed plasma ion process in order to establish the minimum nitriding temperature, bearing in mind that high as-quenched hardness values for

132

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 3

Comparative hardness of plasma nitrided versus gas nitrided type 422 stainless steel. Courtesy of Seco/Warwick Corporation

martensitic stainless steels begin to decrease at temperatures as low as 150 °C (300 °F). Typical composition of type 440A is: Element

Wt%

Carbon Manganese Phosphorus Sulfur Silicon Molybdenum Chromium

0.7 0.85 0.040 max 0.030 max 0.9 0.75 17

The preheat treatment of the steel to achieve the core properties and hardness values was: Preheat 1 Preheat 2 Austenitize Quench Stabilize Temper

370 °C (700 °F) 760 °C (1400 °F) 1015 °C (1860 °F) with a 30 min soak at part temperature Oil Cryogenic treatment using liquid N2 for 20 min 190 °C (375 °F)

Chapter 12: Steels for Nitriding / 133

The nitriding specification was: Characteristic

Case depth Case hardness Compound zone

Required value

0.0375 mm (0.0015 in.) max 800 HV (~64 HRC) 0.0025 mm (0.0001 in.) max γ′

The process parameters were: Process

Sputter time Sputter voltage Sputter retort pressure Process temperature Cycle time Operating pressure Operating voltage Cooling

Parameter

1h 650 V 150 Pa (0.025 psi) 190 °C (375 °F) 16 h 300 Pa (0.05 psi) 450 V Partial pressure with clean, dry N2

Summary Results. The process temperature of 190 °C (375 °F) was too low for effective nitrogen diffusion. Diffusion did occur, but not sufficiently to form a commercially usable case. The tempering curve was further examined for a comparable martensitic grade containing a slightly higher carbon content (AISI type 440C with 1.10% C) (Fig. 4). Based on

Fig. 4

Tempering curve for type 440C stainless steel. Composition: 1.02 C, 0.48 Mn, 0.017 P, 0.011 S, 0.18 Si, 0.54 Ni, 16.90 Cr, 0.64 Mo. Heat treated at 1040 °C (1905 °F), 2 h. Oil quenched from 66 to 94 °C (150 to 200 °F). Double stress relieved at 175 °C (345 °F), 15 min. Water quenched. Tempered 2 h. Heat treated, 9.78 mm (0.385 in.) round. Tested, 9.53 mm (0.375 in.) round. At 260 to 540 °C (500 to 1000 °F). Also, heat treated, 14 mm (0.550 in.) round. Tested, 12.8 mm (0.505 in.) round. At 295 to 760 °C (1100 to 1400 °F). Source: Republic Steel

134

/ Practical Nitriding and Ferritic Nitrocarburizing

the tempering curve, it was concluded that a process temperature of 370 °C (700 °F) could be used. Temperatures above 370 °C (700 °F) cause rapid deterioration of the corrosion resistance of the 440A stainless steel. On that basis, a further cycle was made at a process temperature of 370 °C (700 °F) with the following process parameters: Process

Parameter

Sputter time Sputter voltage Sputter retort pressure Process temperature Cycle time Operating pressure Operating voltage Cooling Final case depth Final case hardness

1h 650 V 150 Pa (0.025 psi) 370 °C (700 °F) 16 h 300 Pa (0.05 psi) 450 V Partial pressure using clean, dry N2 0.05 mm (0.002 in.) 896 HV (~67 HRC)

By using a higher process temperature and extending the cycle time at temperature, nitrogen diffusion will take place and form stable nitrides, but with a very shallow case depth. A higher process temperature causes serious deterioration of the surface corrosion characteristics. The corrosion resistance of another martensitic stainless steel is seen in Fig. 5, with and without nitriding, and compared to a high strength 4140 steel.

Fig. 5

Reduction in corrosion resistance after nitriding of type 422 stainless steel

Chapter 12: Steels for Nitriding / 135

Example 3: Nitriding of AISI Type 630 (17-4 PH) Stainless Steel Type 630 stainless steel is widely used in the aircraft and aerospace industry for gear manufacture and other critical performance items. It is known as a precipitation hardening steel, providing good tensile strength and impact values when heat treated by solutionizing and precipitation treatments. Pulsed plasma ion nitriding technology was considered for this steel because of its ability to provide controlled, repeatable metallurgy without affecting the core. Ion nitrided type 630 also maintains good corrosion resistance. Typical composition of type 630 is: Element

Wt%

Carbon Manganese Silicon Nickel Copper Niobium Tantalum Chromium

0.07 max 0.89 1.00 3.8 3.9 0.3 0.4 16.5

The preheat treatment consisted of: Preheat 1 Preheat 2 Solutionize Quench Precipitation harden Final hardness

315 °C (600 °F) 455 °C (850 °F) 1045 °C (1910 °F) (soak for 20 min at temperature) Oil 480 °C (905 °F) 423 HV (~43 HRC)

The nitride process specification was: Case depth required Core hardness required Compound zone required

0.05 mm (0.002 in.) max 830 HV (65 HRC) min 0.005 mm (0.0002 in.) max

The plasma nitride process parameters were: Process

Parameter

Sputter time Sputter voltage Sputter retort pressure

1h 600 V 150 Pa (0.025 psi) (continued)

136

/ Practical Nitriding and Ferritic Nitrocarburizing

Process

Process temperature Cycle time Operating pressure Operating voltage Cooling

Parameter

480 °C (900 °F) 4 h (at temperature) 300 Pa (0.05 psi) 450 V Partial pressure using clean dry nitrogen gas

Summary Results. The microhardness survey (Fig. 6) showed a surface hardness of 960 HV that diminishes into the material core below the formed case. The transition hardness from effective case to core was noted at 500 HV. The results showed no compound zone. A low nitriding temperature of 480 °C (900 °F) was selected to improve the core hardness value, which it did, with a slight increase in the core hardness of almost 1 HRC. The pulsed plasma nitriding process acted as an additional precipitation treatment (Ref 4). Further cycles were made with similar metallurgical results. It was further observed microscopically that a much denser structure of ferrite stringers emerged in the martensite matrix using Vilella’s reagent (5 mL HCl plus 1 g picric acid plus 100 mL ethanol).

Plasma Nitride Case Depths The following temperature factor values are based on the Harris formula (Ref 5): Case depth = Square root of time × Temperature factor

Fig. 6

Microhardness of AISI 630 (17-4 PH) stainless steel after pulsed plasma ion nitriding. Source: Ref 4

Chapter 12: Steels for Nitriding / 137

Suggested temperature factors for the process cycle time in relation to the required case depth are: Temperature °C

°F

460 470 475 480 500 510 515 525 540

865 875 885 900 930 950 960 975 1000

Temperature Factor

0.00221 0.00233 0.00259 0.00289 0.0030 0.0033 0.0035 0.0037 0.0038

The factors are based on the steel at the selected process time. They do not pertain to nitriding of the higher alloyed steels such as stainless steels. The diffusion rate of nitrogen into the steel surface will reduce dramatically as the alloy content increases. The factors also do not consider the furnace loading and load density, and positioning in relation to potential shielding of the work (e.g., hollow cathode in the process chamber). The factors are based on a simple nitriding steel without the addition of aluminum. The cycle times will be approximate, and serve only as a guide to the cycle time for a particular load. The process technician should keep a record of: • • • • • • • •

Load surface area Load mass Selected process temperature Plasma power conditions Process pressure Process gas flows Grade or grades of steel being processed Case depth achieved

REFERENCES 1. D. Pye, Nitriding Techniques and Methods, Steel Heat Treatment Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker, Inc., 1997, p 721–764 2. A. Fry, The Nitriding Process, ASST Nitriding Symposium, 1929, reprinted in Source Book on Nitriding, American Society for Metals, 1977, p 99–106

138

/ Practical Nitriding and Ferritic Nitrocarburizing

3. K.-E.Thelning, Nitriding, Steel and Its Heat Treatment, 2nd ed., Butterworths, 1984, p 492–544 4. D. Pye, Pulsed Plasma Ion Nitriding and Its Effects on the Surface Modification of Stainless Steels AISI 422, 440A and 630, Surface Modification Technologies VI, T.S. Sudarshan and J.F. Braza, Ed., Minerals, Metals & Materials Society, 1993, p 195–216 5. F.E. Harris, Case Depth, Metals Progress, Vol 44, Aug 1944, p 265

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p139-152 DOI: 10.1361/pnafn2003p139

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

13

Control of the Process Gas in Plasma Conditions PROCESS GAS CONTROL for plasma (ion) nitriding is a matter of estimating the flows necessary to accomplish the required surface metallurgy. Conventional gas nitriding systems measure the dissociation of the ammonia process gas. This is not quite so simple with plasma systems because the process is operating under partial pressure conditions, making it difficult to introduce an effective sampling system into the unit. Compounding the problem is the fact that the process gas is at cathode potential and at the work surface. This chapter reviews several studies aimed at better understanding process gas control in plasma nitriding and its influence on compound zone formation. Emphasis is placed on the effect of sputtering on the kinetics of compound zone formation. Additional information on gas ratios and gas flow can be found in Chapters 8 and 9.

Analysis by Photo Spectrometry A significant development in process gas control analysis can be credited to N. Ryzhov of Moscow State University, who developed what appeared to be a workable control system of the gas species activity of the plasma glow seam (Ref 1). The work, which has not yet been commercialized, is based on the solubility limit of nitrogen in iron and the amount of atomic nitrogen available for diffusion. The gas is evaluated by line-ofsight through an observation port in the furnace process chamber. The system uses photo spectrometry as the principle of operation. The observation unit, sighted onto the plasma glow seam, develops an electrical signal that is then transmitted to a personal computer/programmable logic control (PC/PLC) combination (Fig. 1). This controls the gas-delivery system

140

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Ion nitriding furnace incorporating process gas control analysis system. (a) Furnace layout. (b) Schematic of equipment layout. Source: Ref 1

to the process chamber to manipulate the process gases around the required gas ratios and required surface metallurgy.

Analysis by Mass Spectrometry Szabo and Wihelmi discussed mass spectrometric diagnosis of the surface nitriding mechanism in a direct-current (dc) glow discharge (Ref 2). Their hypothesis was that it makes no difference whether the nitrogen source is ammonia, or nitrogen and hydrogen. Use of hydrogen, they stated, is an important function in the nitriding process. The hydrogen acts

Chapter 13: Control of the Process Gas in Plasma Conditions

as a reducing gas to reduce (with heat) surface oxides on the steel and, perhaps more importantly, to influence and regulate the composition of the compound zone (white layer) as shown in Fig. 2. They also investigated the ability of alloying elements to form nitrides in the steel surface, concluding that the following elements will readily form nitrides (listed in increasing order of ease of nitride formation): • • • • • • • • •

Iron Manganese Silicon Tungsten Molybdenum Chromium Vanadium Titanium Aluminum

To observe the reactions of the process gases in the nitriding chamber, they connected a mass spectrometer. They determined that they could analyze the gas activities and surface reactions taking place at the steel surface. This meant that the solubility limit of nitrogen in iron could be not only observed, but also controlled.

Difficulties Associated with Gas Analysis Because the plasma nitriding process is under vacuum conditions, it is difficult to evaluate accurately the gaseous species activities within the ionized gas glow seam and, more importantly, the amount of nitrogen diffusion into the steel surface in relation to the solubility limits of nitrogen in iron. The Russian method of glow seam observation and analysis of the gaseous activities seemed to hold the most promise as a method of accurately controlling not only the solubility limit of nitrogen in iron, but also the solubility limit of carbon in austenite. Control of the gas activities has eluded scientists and metallurgists in the field of pulsed plasma ion nitriding. Using sensors to measure gas

Fig. 2

Commencement of nitride formation on a steel surface. Note: The hydrogen now acts as a reducing agent. Source: Ref 3

/ 141

142

/ Practical Nitriding and Ferritic Nitrocarburizing

dissociation during gas nitriding or derivatives of that technique remains the most accurate method of process control. Control of salt bath nitriding via titration also allows a good deal of process accuracy.

Kinetic Studies Kinetic studies of compound zone formation in the plasma nitriding of chromium-molybdenum-vanadium steels were carried out by Roli´nski and Sharp (Ref 4). Their work was performed at 540 °C (1000 °F) in a mixture of 30% nitrogen and 70% hydrogen. They found that the process could be described by a half-order polynomial equation and used TableCurve 2D software (SPSS Science, Inc.) to determine the effect of sputtering rate on compound layer growth and composition. A complete account of the work of Roli´nski and Sharp can be found in the Appendix to this chapter.

Conclusions At present, control of the quality of the surface metallurgy and the formation of the nitride diffusion zone during plasma nitriding requires careful process monitoring in terms of: • • • • • • • • •

Gas ratios Gas flows Process vacuum pressure Process time Process temperature Pulse voltage Pulse duration Current density Surface area

Appendix: The Role of Sputtering in Plasma Nitriding E. Roli´nski and G. Sharp The role of sputtering in plasma nitriding has been a subject of many studies in the last 35 years (Ref 5–19). A qualitative approach to analysis of the sputtering rate (SR) in plasma nitriding has been proposed by Keller (Ref 6), who concluded that the parabolic growth of the compound zone (zone containing a mixture of Fe4N and Fe2–3N nitrides) is affected by a linear removal rate of the surface atoms due to sputtering. The first numerical analyses of these processes were done by Marciniak (Ref 13, 14) and Sun and Bell (Ref 19). The calculated and experimentally verified SRs for samples plasma nitrided at 520 °C (970 °F) were about 0.6 g/m2 h (0.1 µm/h) for ion and 0.6 to 0.8 g/m2 h for 0.38C-1.6Cr-Al-Mo steel (Ref 13). The

Chapter 13: Control of the Process Gas in Plasma Conditions

model developed by Sun and Bell (Ref 19) allowed precise calculations of the compound layer growth for a specific steel, the specific nitriding conditions, and the assumed SR. The assumed values of SR were between 0.1 and 0.5 µm/h (Ref 19). If the assumptions of Keller (Ref 6) and Marciniak (Ref 14) are correct, then the kinetic of the compound zone growth y in plasma nitriding can be described by the half-order polynomial equation: y = a + bx + c√x

(Eq 1)

where a is compound zone thickness formed during the ramp-up time, b is SR, c is a coefficient of compound zone growth due to diffusion, and x is nitriding time. The experimental results of Edenhofer (Ref 8) for the kinetics of the compound zone formation on En 9 (AISI 4142) steel plasma nitrided at 450, 530, and 570 °C (840, 985, and 1060 °F) were fitted to Eq 1 using the TableCurve 2D software (Ref 20) and are presented in Fig. 3. The fit of the data for the available 45 h range is very good, since the coefficient of determination, r 2, is high: 0.993, 0.995, and 0.962, respectively. The calculated SR is 0.057 µm/h for 450 °C (840 °F), 0.124 µm/h for 530 °C (985 °F), and 0.427 µm/h for 570 °C (1060 °F). The graph extrapolated to 100 h of nitriding time shows the tendency of the curves: the maximum, respectively, at approximately 65, 50, and 35 h of nitriding and the diminishing values thereafter. At the same time, the growth due to diffusion (the coefficient c) is larger for higher temperatures and smaller for lower nitriding temperatures. Similar graphs based on the experimental data of Marciniak (Ref 14) for 36 H3M (0.36C-3Cr-0.6Mo-0.6Mn-0.27Si) steel nitrided at 530 °C (985 °F)

Fig. 3

Compound zone thickness versus nitriding time for 42Cr Mo4 (AISI 4142 steel) plasma nitrided 3.3 mbar in the atmosphere of 25% nitrogen + 75% hydrogen at 570 °C (1060 °F) (upper curve), 530 °C (985 °F) (middle curve), and 450 °C (840 °F) (bottom curve) based on the experimental data of Edenhofer (Ref 8, 24). The graph is extrapolated over the original limit of 45 h. The fit equations are y = 3,8722 – 0.236x + 2.361√x, y = –0.789 – 0.124x + 1.958 √x, and y = 0.758 – 0.057x + 0.918√x and r 2 = 0.993, 0.995, and 0.962, respectively.

/ 143

144

/ Practical Nitriding and Ferritic Nitrocarburizing

are presented in Fig. 4. The curve with a well-distinguished maximum at about 16 h drawn for nitriding with 50% nitrogen and 50% hydrogen represents a growth of a compound zone consisting of a mix of the epsilon (ε) and gamma prime (γ′) type nitrides (Ref 13). The SR is about 0.425 µm/h. Nitriding with 15% nitrogen and 85% hydrogen produced a γ′-type compound zone and an SR of about 0.036 µm/h. The “diffusion” fraction of the kinetic equation, which represents a nitriding potential, is higher for the samples nitrided with 50% nitrogen and 50% hydrogen than for samples nitrided with 15% nitrogen and 85% hydrogen. However, all of the above predictions for the range of time exceeding the experimental data range may not be accurate since extreme caution is advised in relying on polynomials for extrapolations and foreasts beyond the range of the dataset (Ref 20). The diminishing value of the compound zone thickness during a long nitriding time may have important practical meaning; the compound zone could eventually disappear completely and therefore the nitriding rate would drop down and/or a denitriding of the steel could take place (Ref 9). Wells and Strydom suggested that the compound zone growth might also be affected by redeposition of the sputtered material (Ref 16). This phenomenon could be enhanced by oxygen, which may always be present in a small quantity in industrial systems, and, therefore, the compact portion of the compound zone can be significantly reduced by formation of the oxynitride (Ref 16). Consequently, in our studies, we researched the actual kinetics of compound zone formation in long nitriding with an aim to establish the SR in such a cycle. The 3% Cr-Mo-V steel we investigated is used in the gear industry to achieve an exceptionally deep case. It was then very important

Fig. 4

Compound zone thickness versus nitriding time for 36H3M 3% Cr-Mo steel plasma nitrided at 530 °C (985 °F) in the atmosphere of 50% nitrogen + 50% hydrogen (upper curve) and 15% nitrogen + 85% hydrogen (bottom curve) based on the experimental data of Marciniak (Ref 14). The graph is extrapolated over original limits of 16 and 36 h, respectively. The fit equations are y = 1.275 – 0.425x + 3.295√x, y = 0.5 – 0.036x + 0.479√x, and r 2 = 0.994 and 0.986, respectively.

Chapter 13: Control of the Process Gas in Plasma Conditions

to verify if the thickness of the compound zone would become thinner to the point of disappearing completely with prolonged nitriding.

Experimental Parameters Material and Processing. A quenched and tempered 3% Cr-Mo-V (DIN 39CrMo V13.9) nitriding steel was used for these studies. The Brinell hardness was 321 to 363. The test samples were 25 × 25 × 150 mm (1 × 1 × 6 in.) bars with a ground surface finish of Ra 1.6 µm or better. All samples were blasted with 180 grit aluminum oxide before nitriding. The equipment and processing details are described elsewhere (Ref 21). The nitriding was carried out at 540 °C (1000 °F) nominal temperature, and the nitriding times were from 4 to 400 h. The samples were treated in a direct current (dc) plasma in the atmosphere of 30% nitrogen and 70% hydrogen and a pressure of 3.2 mbar. The ramp-up time was 4 h. Testing Procedure. The samples were cut in half and prepared for metallographic studies. The compound zone thickness was measured at 400× using the digital filar eyepiece of the MICROMET II microhardness tester (Buehler, Lake Bluff, IL). Each sample was tested in four different areas (four sides). The compact portion of the compound zone was measured. The local peaks appearing on the surface were ignored. The accuracy of a single measurement was ±0.1 µm. There was a minimum of five samples used in each run. The x-ray diffraction phase analysis was performed using Cr Kα radiation.

Results and Discussion The compound zone thickness changes with the nitriding time, as shown in Fig. 5. A possible maximum value of approximately 13.5 µm is

Fig. 5

Compound zone thickness versus nitriding time for 3% Cr-Mo-V steel plasma nitrided at 540 °C (1000 °F). The fit equation is y = 6.158 – 0.0294x + 0.933√x and r2 = 0.952. Confidence and prediction intervals represent normal distribution and standard error (small interval) at 95%.

/ 145

146

/ Practical Nitriding and Ferritic Nitrocarburizing

achieved after about 250 h of nitriding, and the final value after 400 h is about 13 µm. Micrographs of the compound zone are presented in Fig. 6. They do not reveal any significant presence of the vapor-deposited, possible oxygen-contaminated layer, as was suggested by Wells and Strydom (Ref 16). The top portion of the compound zone contains some porosity and probably a layer of the nitrides deposited from plasma; however, its main fraction stays very compact. Optical microscopy of the microsections also revealed a presence of frequent conical structures at the surface. Similar features were also observed by others (Ref 22). The intensity of these peaks was higher on samples nitrided longer. The x-ray diffraction showed that initially after 4 h of

Fig. 6

Optical micrographs of surface layers produced on 3% Cr-Mo-V steel by plasma nitriding at 540 °C (1000 °F) for (a) 4 h, (b) 25 h, (c) 144 h, (d) 289 h, and (e) 400 h. Bright field, etched with 2% nital

Chapter 13: Control of the Process Gas in Plasma Conditions

nitriding two iron nitrides, ε (Fe3N) and γ′ (Fe4N), were formed on the surface (Fig. 7). The intensity of the ε patterns is quickly reduced and, after 25 h of nitriding, only two weak diffraction peaks from the (100) and (101) planes could be detected. The 400 h nitriding produced a single-phase γ′ compound zone on the steel surface. It is very likely that the ε-nitride formation was promoted by carbon present in the steel. It is known that carbon stabilizes the ε-carbonitride and that it can diffuse outward from the steel during nitriding (Ref 9, 17, 23, 24). In the short nitriding processes, carbon atoms were diffused toward the surface and, by reacting with iron and nitrogen, helped in the formation of the ε-phase. During the long exposure to the plasma, the surface was decarburized as carbon was sputtered away and replaced by nitrogen. With a lack of carbon atoms, the nitriding potential of the plasma was shifted toward γ′; the lower nitrogen phase and the ε-phase disappeared completely. The results of the compound zone thickness versus nitriding time studies clearly demonstrate the effect of sputtering: a possible maximum value at about 250 h and the diminishing values of the compound zone thereafter. The calculated SR is 0.0295 µm/h, and the coefficient c of compound zone growth due to diffusion of nitrogen is 0.933. The total compound zone thickness a, formed during ramp-up to the final temperature, is about 6 µm. The a value depends on growth due to diffusion of nitrogen and sputtering. Since sputtering was taken into account only from a time when the final temperature was reached, the a value was affected by an error of not counting the material removed. In fact, this value is only about 0.06 µm if we assume that SR was the same during ramping and the final soaking.

Fig. 7

X-ray diffraction patterns of surface layers produced on 3% Cr-Mo-V steel plasma nitrided at 540 °C (1000 °F) for (a) 4 h, (b) 25 h, (c) 144 h, (d) 289 h, and (e) 400 h. Cr Kα radiation

/ 147

148

/ Practical Nitriding and Ferritic Nitrocarburizing

The SR of 0.0295 µm/h from this experiment agrees very well with the value of 0.036 µm/h calculated by us for the literature data of nitriding performed in the atmosphere of 15% nitrogen and 85% hydrogen on similar steel (Ref 14). However, our kinetic studies as well as the studies based on the results of others (Ref 8, 9, 14, 15) could be affected by not taking into account the fact that a phase composition of the steel surface may change during a long nitriding process. In the plasma nitriding of the 3% Cr-Mo-V steel, the equilibrium at the surface was not achieved very rapidly; in fact, it took many hours to produce a “pure” γ′-nitride. It can then be concluded that the sputtering yield depends not only on pressure, temperature, gas composition, and the plasma power density, but mainly on the phase composition of the compound zone that formed on the surface. Some of these parameters were probably different in our experiment than in the experiments carried out by the other researchers (Ref 8, 14, 19, 24, 25); however, as long as we consider γ′ compound zone formation, its SR stayed low. In comparatively short nitriding cycles with a sufficiently high nitrogen content carried out by others (Ref 8, 14), a mixture of the γ ′ and ε was produced and, consequently, the SR was much higher than when the “pure” γ ′-phase was produced. This can be seen in Fig. 8, which represents the nitriding kinetics for shorter (up to 25 h) cycles. The graph is not biased by the phase composition change toward a pure γ ′; the SR is about 0.049 µm/h and the coefficient c about 1.221. If “sputtering-free” plasma nitriding is hypothetically assumed, Eq 1 can be used with the coefficient b equal to zero to see the effect of the remaining (diffusion) fraction of the equation on the kinetic.

Fig. 8

Compound zone thickness vs. nitriding time for 3% Cr-Mo-V steel plasma nitrided at 540 °C (1000 °F). This is a modified form of Fig. 5 from which the data for a “pure” γ′ compound zone were removed (144, 289, and 400 h). The fit equation is y = 5.438 – 0.049x + 1.221 √x and r 2 = 0.958. Confidence and prediction intervals represent normal distribution and standard error (small interval) at 95%.

Chapter 13: Control of the Process Gas in Plasma Conditions

Figure 9 presents two curves drawn for the plasma nitriding processes: one from our studies and the second one based on the work of Marciniak (Ref 14), compared with the curve for gas nitriding with a very low nitriding potential of KN = 0.58 atm–1/2 (Ref 26). The curves for plasma nitriding of 3% Cr-Mo-V steel and the gas nitriding of Armco iron are very similar; they achieve a final value of 25 to 30 µm after 400 h of nitriding. The final value for the plasma nitriding curve of 36H3M steel is about 10 µm. Regardless of the sputtering effect, the plasma nitriding process in the atmosphere of 30% nitrogen and 70% hydrogen can be considered a low nitriding potential process. In a plasma process, activation of the cathode due to ion bombardment from the atmosphere containing sufficiently high nitrogen is very effective, and, therefore, a full coverage of the surface with the compound zone after only a few minutes of nitriding is achieved (Ref 19). This was also evidenced in our experiments by the fact that the curve did not start at the beginning of the coordinate, but earlier (Fig. 5). In a low-potential gas nitriding process, the γ ′-phase nucleates on ferrite extremely slowly and only after substantial time becomes a continuous, compact layer (Ref 26). The experiments showed also that the equilibrium between plasma and the steel surface was not achieved quickly when the atmosphere of 30% nitrogen and 70% hydrogen was used. Instead, an initially formed mixture of γ ′- and ε-phases was slowly converted into a single γ ′-phase structure. The phase composition changes resulted in a reduction of the SR, as well as a reduction in the diffusional growth of the compound zone. Sputtering of the surface in plasma nitriding has then an additional

Fig. 9

Comparison of a hypothetical, “sputtering-free” kinetic of compound zone growth in plasma nitriding of 3% Cr-Mo-V steel at 540 °C (1000 °F) (curve with the middle final value) and 36H3M 3% Cr-Mo steel at 530 °C (985 °F) (Ref 14) (curve with the lowest final value) with γ′ compound zone growth in gas nitriding of Armco iron at 550 °C (1020 °F) with a constant nitriding potential KN = 0.58 atm–1/2 (curve with the highest final value) (Ref 26). The equations are y = 6.158 + 0.933√x, y = 0.5 + 0.479√x, and y = 1.278 + 1.461√x.

/ 149

150

/ Practical Nitriding and Ferritic Nitrocarburizing

effect in the process: It effectively lowers an already low nitriding potential and enhances the ability to reduce the thickness of the compound zone.

Conclusions The experimental data of the compound zone formation taken from the literature were analyzed and presented in graphic forms. It was found by using the TableCurve 2D software that the SR for the γ′ (0.036 µm/h) was smaller than for the ε-type compound zone (up to 0.425 µm/h). At the same time, it was confirmed that the growth of the compound zone due to the diffusion of nitrogen was also slower for the γ′-type, which could be expected (Ref 26). The graphs extrapolated over the experimental data range showed that the kinetic curves may achieve a maximum value and that the sputtering may cause a complete disappearance of the compound zone after long nitriding. These types of kinetic characteristics were more likely for the ε-type compound zone than for γ′. The experiments carried out on 3% Cr-Mo-V steel did not show any disappearance of the compound zone, and its final value after 400 h of nitriding was still about 13 µm. The SR calculated from our experiment was about 0.03 µm/h. This agrees well with the value calculated for kinetic data taken from the literature for a similar steel for presumably the γ′-type nitride (Ref 14). The analysis limited to cycles not exceeding 25 h resulted in a higher rate of sputtering as well as a faster diffusional growth. This fact can be attributed to the presence of the ε-phase in the compound zone. Based on our research, it seems to be likely that the γ′ compound zone can disappear completely because of sputtering after an extremely long nitriding time. Instead, it is more likely that it will become more porous and the specific surface area will be greater. However, this will still need to be proved by additional experimental work. ACKNOWLEDGMENT The Appendix is reprinted (with minor changes) from E. Roli´nski and G. Sharp, The Effect of Sputtering on Kinetics of Compound Zone Formation in the Plasma Nitriding of 3% Cr-Mo-V Steel, Journal of Materials Engineering and Performance, Vol 10 (No. 4), Aug 2001, pages 444 to 448 (reproduced by permission of ASM International). REFERENCES 1. D. Pye, A Review of the Gas Species Activity and Control of Pulsed Plasma Technology during the Nitriding, Carburizing, and Carbonitriding Processes, 1995 Carburizing and Nitriding with Atmospheres, ASM International, 1995, p 347–351 2. A. Szabo, Best Surface GmbH, personal communication, June 2001

Chapter 13: Control of the Process Gas in Plasma Conditions

3. M.A.J. Somers and E.J. Mittemeijer, Oxidschichtbildung und Gleichzeitige Gefugeanderung der Verbindungsschicht (Oxide-Layer Formation and Simultaneous Microstructural Changes in the Compound Layer), Härt. Tech. Mitt., Vol 47 (No. 3), May-June 1992, p 169–174 4. E. Roli´nski and G. Sharp, Effect of Sputtering on Kinetics of Compound Zone Formation in the Plasma Nitriding of 3% Cr-Mo-V Steel, J. Mater. Eng. Perform., Vol 10 (No. 4), Aug 2001, p 444–448 5. H. Knüppel, K. Brotzman, and F. Elserhard, Stahl Eisen, 1958, Vol 75 (No. 26), p 1871 6. K. Keller, Schichtaufbau Glimmnitrierter Eisenwerkstoffe, Härt. Tech. Mitt., Vol 26, 1971, p 120 (in German) 7. M. Hudis, J. Appl. Phys., Vol 44, 1973, p 1489 8. B. Edenhofer, Heat Treatment ’76, Proc. 16th International Heat Treatment Conf., Stratford-upon-Avon, 6–7 May 1976, The Metals Society, 1976, p 7 9. B. Edenhofer, Proc. Heat Treatment ’79, Birmingham, 22–24 May 1979, TMS/ASM, 1979, p 52 10. G.G. Tibbets, J. Appl. Phys., Vol 44, 1974, p 5072 11. E. Roli´nski, Ph.D. dissertation, Warsaw University of Technology, 1978 (in Polish) 12. T. Karpinski and E. Roli´nski, Proc. 8th National Conf. Heat Treatˇ ment, Bratislava, Slovakia, 1978, Dom Techniky CSVTS, p 27 (in German) 13. A. Marciniak, “Processes of the Cathode Heating and Nitriding under a Glow Discharge Condition,” Ph.D. dissertation, Warsaw University of Technology, 1983 (in Polish) 14. A. Marciniak, Surf. Eng., Vol 1, 1985, p 283 15. A. Wells and I. Le R. Strydom, Surf. Eng., Vol 2, 1986, p 283 16. A. Wells and I. Le R. Strydom, Surf. Eng., Vol 4, 1988, p 55 17. T. Lampe, S. Eisenberg, and G. Laudien, Surf. Eng., Vol 9, 1993, p 69 18. H. Michel, T. Czerwiec, M. Gantois, D. Ablitzer, and A. Ricard, Surf. Coating Technol., Vol 72, 1995, p 103 19. Y. Sun and T. Bell, Mater. Sci. Eng., Vol A224, 1997, p 33 20. TableCurve®2D, Version 4, SPSS Inc., 1998 21. E. Roli´nski, F. LeClaire, D. Clubine, G. Sharp, D. Boyer, and R. Notman, J. Mater. Eng. Performance, Vol 9, 2000, p 457 22. Y. Sun, N. Lou, and T. Bell, Surf. Eng., Vol 10, 1994, p 279 23. J. Zysk, Metaloznastwo I Obrobka Cieplna, No. 6, 1973 (in Polish) 24. B. Edenhofer, Ibsen Industries International GmbH, private communication, Aug 2000 25. J.G. Conybear and B. Edenhofer, Proc. 6th Int. Conf. Heat Treatment of Materials, Chicago, IL, 28–30 Sept 1988 26. L. Maldzinski, W. Liliental, G. Tymowski, and J. Tacikowski, “New Possibilities of Controlling the Gas Nitriding Process by Utilizing Simulation of Growth Kinetics of Nitride Layers,” presented at the 18th ASM Conf., Rosemont, IL, Oct 1998

/ 151

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p153-162 DOI: 10.1361/pnafn2003p153

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

14

Processing with Nitriding THE NITRIDING PROCESS can be applied to various materials and part geometries. This chapter focuses on tool steels, pure irons, low-alloy steels, and maraging steels.

Hot-Work Tool Steels The hot-work group of tool steels is usually considered to be the AISI H-series and includes chromium-base, tungsten-base, and molybdenumbase steels. While all of the hot-work tool steels contain chromium ranging from 2 to 12%, they are distinguished by their principal alloying element. All can be readily processed via gas nitriding, salt bath nitriding, or ion nitriding.

Forging Dies Selection of a hot-work steel grade depends on the forge die application. For many forging steel applications, the steel of choice is H13, which is classified as a deep-hardening chromium hot-work steel containing 5% Cr and 0.40% C. This steel can be readily water cooled while in service and has a good toughness factor after nitriding—provided that the preheat treatment has been done correctly in terms of core hardness for case support. For the diffused case to perform within its operating environment, the core must be able to support the case when a compressive load is placed on the steel component (Fig. 1). Core hardness generally is determined by the hardness at which the steel can be cut, rather than the best hardness for supporting the case. The appropriate hardness for case support is around 44 to 47 HRC, which produces a tough, springlike condition. This core hardness will allow some flexibility in the die without taking a permanent set for deformation. Another advantage of the nitrided case is that it will withstand hightemperature operating conditions with no significant loss of surface hardness. Nitrided hot-work tool steels are unlike carburized steels that rely on the diffusion of carbon and then a phase transformation to martensite,

154

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Core support of the nitrided case on a forging die. Source: Ref 1

which necessitates a quench (and results in distortion and the possibility of mixed phase conditions). If a mixed phase is present, the nitriding process will decompose it. It must be expected that with the decomposition of the mixed phase, some size change will occur due to transformation of the retained austenite to martensite. Other considerations are the surface metallurgy requirements of the die, including case depth, compound layer formation, and temperature. Each factor is discussed later in this chapter. Case Depth. A deep case is unnecessary. With a deep case formation, the surface will begin to lose its flexibility, no matter how well the preheat treatment has been conducted. Inflexibility leads to surface cracking, and thus to press downtime for die repair. Compound Layer Formation. The thickness and phase construction of the compound layer significantly influence die performance. If a thick case is produced, there is a strong likelihood that a thick, inflexible compound layer will be produced on the die surface. Therefore, the cycle must be run with a thin diffused nitrided case with a thin compound surface layer. Recommended case depth is a maximum of approximately 0.25 mm (0.010 in.). The thinner the diffused case, the thinner the compound layer. With gas and salt bath nitriding methods, the thickness of the compound layer can be expected to be roughly 10% of the total measured case depth. However, these values will change with both controlled nitriding (dilution) and particularly with ion nitriding. The latter two methods offer a greater degree of surface metallurgy control. Temperature. Process temperature selection plays a significant role in the thickness of the diffused case and formed compound layer. If higher process temperatures are selected for gas or salt bath nitriding, there is an inherent danger of nitride networking on corners. This is a very brittle phase condition, and care should be taken to minimize its potential. Process temperature traditionally has been selected based on the temper-

Chapter 14: Processing with Nitriding / 155

ing temperature of the steel, without considering the potential for nitride networking at higher temperatures. If the surface metallurgy and the nitride potential can be controlled, the possibility of nitride networking can be greatly reduced.

Aluminum Extrusion Dies Aluminum extrusion dies typically are manufactured using the hotwork steel H13 described earlier. The main area of concern with extrusion dies is the bearing face area (Fig. 2). The mechanics of operation are based on the press load factor on the aluminum billet to be extruded. The aluminum billet is usually preheated to around 425 °C (800 °F). At room temperatures, the surface of the aluminum billet will oxidize to form aluminum oxide. Aluminum oxide is extremely abrasive and will begin to abrade and wear the bearing surface of the extrusion die. The die aperture will become larger and out of tolerance, resulting in a costly shutdown of the press. Typical extrusion press configurations are shown in Fig. 3. Nitriding—gas, salt bath, or fluidized bed—enhances the hardness of the die bearing surface and reduces wear. Again, a deep case is not required; a shallow case will suffice up to a maximum of 0.25 mm (0.010 in.), with formation of a compound layer in the region of 10% of the total case depth. The compound layer will wear off as the extruded aluminum is pushed over the bearing surface. Aluminum oxide formation becomes much more aggressive due to the frictional forces now being developed as the hot

Fig. 2

Schematic cross section of an aluminum extrusion die made from H13 steel showing the bearing (wear) surface and a core with hardness of 38 to 44 HRC. Source: Ref 1

156

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 3

Extrusion presses. (a) Schematic of a horizontal extrusion press showing a hydraulically powered ram forcing the heated aluminum billet through the die. (b) Typical direct-drive hydraulic extrusion press. 1, hydraulic power unit; 2, tie rods; 3, butt shear; 4, extrusion platen; 5, container shifting cylinders; 6, swiveling operator’s console; 7, die slide; 8, container; 9, container housing; 10, billet loader; 11, press base; 12, billet loader cylinders; 13, pressing stem; 14, crosshead; 15, side cylinders; 16, cylinder platen; 17, main cylinder

aluminum is pushed through the die (Fig. 4). Thus, the wear factor becomes even greater. Ideal die surface metallurgy will reduce formation of the compound layer on the surface. This can be accomplished by controlling the process gas using the controlled nitride method (dilution) or ion (plasma) nitriding. The two-stage process, which uses a higher process temperature of 565 °C (1050 °F), is not recommended because of the possible risk of nitride networking.

High-Speed Steel Cutters Enhancing the surface hardness of high-speed steel cutters via the nitriding process offers many advantages. However, once again, a deep case is not required. A short cycle is necessary only to diffuse a case of no greater than 0.038 mm (0.0015 in.). The formed case will be in the region

Chapter 14: Processing with Nitriding / 157

Fig. 4

Bearing surface of an aluminum extrusion die, demonstrating the wear process due to hot aluminum extrusion. (a) Untreated die. (b) Die with nitrided surface

of 1100 HV or harder and will be supported by a substrate material hardness in the region of 850 HV. Remember, the nitride processing will give the high-speed steel a further temper, producing a dimensional stabilizing effect. There will be a very slight, but insignificant, increase in size due to the diffusion. Another procedure is to follow the nitriding with a thin-film deposition using the plasma-assisted deposition technique. The cutter is nitrided, and then the nitrided surface is immediately coated with titanium nitride. Two process techniques are accomplished during the same furnace operation.

158

/ Practical Nitriding and Ferritic Nitrocarburizing

Gears Nitriding is finding greater acceptance as a surface hardening method for precision manufactured gears, although it requires careful steel selection. The steel selected must not be a high-carbon grade; otherwise, there is a risk of forming ε as the dominant phase in a dual-phase compound zone. This could lead to brittle fracture of the compound layer, depositing fine pieces of fractured steel from the gear surface that will usually lodge between the meshing teeth, thus causing further deterioration of the gear pressure face. Furthermore, the steel must not have a high aluminum content. Otherwise, high hardness will result, leading to premature chipping on tooth corners and possibly on the gear pressure face. The aluminum-bearing steels (Nitralloy steels) are not suitable for gears that require nitriding treatment. Therefore, use a steel with a low carbon content and no aluminum, which will still give the appropriate core hardness value. Control formation of the compound zone on the immediate steel surface by using: • •

Controlled nitriding (dilution) Ion nitriding

When using the ion nitriding process for gear heat treatment, pressure must be controlled to ensure that the plasma glow seam is uniformly positioned over the entire gear tooth surface. This is necessary for uniform case depth on the pressure face and tooth root (Fig. 5). As described below, surface metallurgy—in terms of growth, compound layer formation, and case uniformity—is critical to gear performance. Growth. By selecting a process temperature around 485 to 500 °C (900 to 925 °F) and then controlling it, the compound zone will not form as thickly

Fig. 5

Illustration of ion nitriding pressure that is too low (toward high vacuum), resulting in no nitriding at the tooth root. Source: Ref 1

Chapter 14: Processing with Nitriding / 159

as at higher temperatures and the risk of nitride networking will be reduced, along with the risk of edge or corner chipping. Remember, the solubility limit of nitrogen in iron increases as temperature increases. The solubility limit using a process temperature of approximately 485 °C (900 °F) is approximately 6 to 7%. If compound layer thickness is reduced, growth also is reduced. If surface growth is reduced, less stock will have to be removed by lapping or grinding operations. In fact, the gear pressure face could be machined slightly undersize and “grown” into size, thus reducing expensive machining time. Selection of a lower process temperature thus offers important advantages to the gear manufacturer. Gear teeth must be fully deburred before ion nitriding. Otherwise, localized hot spots will likely occur at burrs. When this happens, the gear tooth will appear dark around the area of the burr. Compound Layer Formation. Process temperature affects formation of the γ′- and ε-phases within the compound zone. A lower process temperature tends to reduce formation of the ε-phase, whereas higher process temperatures encourage its formation—particularly if the steel has a higher carbon level. A greater presence of ε-phase means that the immediate gear surface will have superb wear resistance properties but no impact value. The pressure face will begin to crack and chip with high impact loading. If the gear has been previously tempered at 510 to 540 °C (950 to 1000 °F) and the core hardness is temperature sensitive, the core hardness may be reduced by up to 2 to 4 Rockwell points due to the process time at the nitriding temperature. This could take the gear core hardness out of specification. Pretreatment of the gear to establish the case support core hardness is extremely important. Emphasis should be placed on producing a core hardness that will ensure a good case support and improve fatigue bending performance of the gear tooth rather than a core hardness that will ease machining. Case Uniformity. If a uniform case depth is not maintained, then the gear tooth could fail under service loads. Nonuniform case formation usually results in a shallow case in the gear tooth root. With salt bath nitriding, case uniformity usually is not a problem as the molten salt contacts all gear surfaces. With gas nitriding, it can be a serious problem due to gas stagnation in the root of the tooth. This can be caused by inadequate gas circulation. With ion nitriding, using too low a process pressure or failing to adequately control the process pressure can cause a nonuniform glow seam that does not reach the root of the gear tooth (Fig. 6).

Pure Irons Pure iron can be successfully nitrided, even with gas and salt bath methods. (Please note that this discussion is restricted to nitriding and not ferritic nitrocarburizing, which will be discussed in Chapters 18 to 23.)

160

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 6

Effect of process pressure in ion nitriding of gear teeth. Courtesy of Plateg GmbH

The case that forms is strictly of iron nitride. However, the resulting hardness value is low—in the region of 35 HRC (345 HV). Resistance to surface corrosion and torque improves significantly. Ion nitriding of pure iron produces higher hardness values in the region of 60 HRC (700 HV). This is accomplished by manipulating the process gas ratios so that the ratio of nitrogen to hydrogen is approximately 5:1, higher than would be expected from the decomposition of ammonia gas, which encourages iron nitride formation at the surface. The hardness value will be no greater than 700 HV, but corrosion resistance will significantly improve. This technique often is used when corrosion resistance is more important than wear resistance.

Low-Alloy Steels Low-alloy steels that are nitrided using the gas or salt techniques have low surface hardness values but improved torque and corrosion resistance. The same applies to the ion nitriding technique. Increasing the nitrogento-hydrogen ratio significantly raises surface hardness values (though they will not be high) and further improves corrosion and torque resistance. Remember, low-alloy steels (and even the cast irons) contain no alloying elements to form significant stable nitrides in the steel surface. Therefore, the nitriding potential of the process gas must be raised (via increased nitrogen content) to encourage the formation of iron nitrides. This means high nitrogen flows in relation to hydrogen during ion nitriding. Once again, it is important to consider the solubility of nitrogen in the Fe-N phase diagram (Ref 2) (see Chapters 1 and 3 for examples of the Fe-N binary phase diagram).

Maraging Steels Maraging steels can be nitrided, but the surface hardening mechanism is very different from that of conventional nitriding. As the process tem-

Chapter 14: Processing with Nitriding / 161

perature passes through 450 °C (850 °F) on heat up, a phase change from ferrite to austenite occurs—along with a subsequent size change. The nitrogen diffuses into the maraging steel at that point, and the solubility limit of nitrogen in iron is exceeded. After the cycle time is completed, the maraging steel must be cooled quickly in the process chamber under nitrogen. This will produce a shallow layer on the steel surface due to the rapid cooling of fine nitrogen-austenite products. This product is detrimental in terms of surface mechanical properties. Therefore, the nitriding process temperature for maraging steels should be lower than for conventional steels. The process temperature must be lower than the ferrite-to-austenite phase transformation temperature of approximately 450 °C (850 °F). A process temperature of between 425 and 450 °C (800 and 850 °F) will produce a surface hardness of approximately 67 HRC (900 HV). This lower-than-normal process temperature results in an extended case formation time (Ref 3). The maraging steels can be successfully ion nitrided at lower temperatures to avoid the phase change and subsequent growth, with better control of the hydrogen-to-nitrogen process gas ratios. The ratio generally would be approximately 5 parts hydrogen to 1 part nitrogen or even 6 parts hydrogen to 1 part nitrogen. This means that the solubility limit of nitrogen in iron is not reached, and thus there can be no risk of possible corner networking.

Higher Alloyed Steels Once again, the greater the level of alloying elements, the more difficult it becomes for nitrogen to diffuse into the steel surface and form stable nitrides. This is evident for the whole range of stainless steels and the higher alloyed tool steels. The net result of nitriding the stainless steels is a very high surface hardness. REFERENCES 1. D. Pye, “Practical Nitriding” course notes, Pye Metallurgical Consulting, 1997 2. D. Hawkins, Fe-N (Iron-Nitrogen) Phase Diagram, Metallography, Structures and Phase Diagrams, Vol 8, Metals Handbook, 8th ed., American Society for Metals, 1973, p 303 3. J.B. Seabrook, Working with Maraging Steels for Nitriding, Met. Prog., July 1963, p 78–80

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p163-166 DOI: 10.1361/pnafn2003p163

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

15

Stop-Off Procedures for Selective Nitriding STOP-OFF COATINGS prevent nitriding of selected areas on components. The success of a coating depends on such variables as its density and thickness, its adhesion to steel, and part surface finish.

Methods for Selective Gas Nitriding There are essentially two stop-off techniques for gas nitriding: electroplating and paint-on methods. Electroplating. One method of selective gas nitriding is to plate the areas to be stopped-off using copper or bronze. Nickel and silver are also effective but costly, restricting their use to special applications. Copper plating is perhaps most widely used, because it is least expensive in relation to other types of plated deposition. However, it is still an expensive method—not so much because of the copper plate cost, but because of the labor-intensive part preparation required. All of the areas to be nitrided must be masked prior to copper plating. Thickness and density of the deposit material will determine the effectiveness of the plating. The following is a guide to electroplated deposition thickness: • • •

Bronze and copper: Up to 25 µm (0.001 in.) for ground finishes; up to 50 µm (0.002 in.) for rough finishes Nickel: More effective, allowing thinner coatings, up to 25 µm (0.001 in.); also exhibits excellent resistance to nitrogen penetration Silver: Fairly thick deposits, up to 100 µm (0.004 in.)

Copper and silver are relatively easy to strip after completion of nitriding. However, copper stripping can be expensive due to effluent problems caused by the cyanide-base plating solution, which must be neutralized. Nickel is difficult to strip, and the stripping process could affect part surface quality.

164

/ Practical Nitriding and Ferritic Nitrocarburizing

Paint-On Methods. Proprietary brands of stop-off paints generally contain a liquid carrier and a metallic compound such as copper or even tin and zinc. For copper paint, the carrier liquid is usually an alcohol-base material. When applying copper-base paint, follow the paint manufacturer’s application instructions and do not take shortcuts. Use a high-quality paintbrush that will not leave bristles on the painted steel surface. When the steel is heated to the process temperature, loose bristles on the surface will burn off and expose the steel immediately beneath. Nitriding will take place in the area of the burnt-out paint bristles, which can wreak havoc with cutting tools. Some stop-off paints are water-based and must be correctly cured. Any water remaining in the stop-off paint will manifest itself during nitriding. The water will form a steam pocket under the paint, producing a bubble. When the bubble bursts, a localized hard spot will form on the exposed steel. Stop-off paint residues can be reduced by brushing or washing, or removed by light blasting with fine abrasives.

Methods for Selective Salt Bath Nitriding Copper Plating and Paint. Selective nitriding can be accomplished in salt baths by stopping-off nitrogen penetration with either copper plate or copper-base paint. Because cyanide-base salts can dissolve copper, salt baths with relatively low cyanide contents must be used. One successful salt formulation contains 8 to 10% sodium cyanide (NaCN) with approximately 45% barium chloride (BaCl) energizer. Noncyanide nitriding salts will not dissolve copper. Following the stop-off procedure, the deposited copper should be inspected to ensure it does not contain pinholes. In addition, if too much plated work is processed for short cycles through the cyanide salt bath, the copper plate will begin to strip off and go into solution with the cyanide salt. When fresh work is introduced into the salt bath, the copper in solution in the cyanide will deposit onto the exposed areas, reducing the effectiveness of the nitriding salt bath. Partial Immersion. Another method for selective nitriding entails partial immersion into the salt bath so that only the immersed areas are nitrided. This method depends wholly on part geometry.

Methods for Selective Ion Nitriding Masking an area for plasma ion nitriding follows this simple rule: “What plasma can see, it will nitride; what plasma cannot see, it cannot nitride.” The area to be masked should be covered with a piece of shim stock steel attached to the component. The shim stock serves as an effective mechanical barrier to nitrogen diffusion. Some paints are effective stop-offs in the ion nitriding process; however, extreme caution must be exercised during the sputter cleaning stage. Sputter cleaning should not be aggressive; otherwise, the painted surface will begin to transfer the paint onto the furnace wall or the anode. The

Chapter 15: Stop-Off Procedures for Selective Nitriding

sputtered paint will migrate and adhere to the process wall, leading to a change in the electrical response characteristics of the wall. Once a cycle has been completed, the inner wall should be cleaned of deposits. Another stop-off consideration is that when a component is in contact with the cathodic hearth, it will not nitride. Therefore, a support system is needed. As shown in Fig. 1, the part can be mounted onto steel points. The points will maintain the cathodic contact with the furnace hearth and nitriding can take place. The case depth will be slightly more shallow at the point of contact but not to a significant degree. Threaded holes must be protected from nitriding. This is accomplished simply by partially inserting a threaded stud into the hole. Screwing three or four threads deep is sufficient (Fig. 2).

Fig. 1

Extrusion dies lifted from the furnace hearth to allow the plasma glow seam to cover the die completely. Note the die supports (steel tubes) used to maintain cathode potentials.

Fig. 2

Masking of blind tapped holes on a component for plasma nitriding.

/ 165

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p167-184 DOI: 10.1361/pnafn2003p167

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

16

Examination of the Nitrided Case EXAMINING AND EVALUATING the nitrided case is generally accomplished by hardness testing and microscopic examination. This chapter discusses both characterization methods, as well as sample preparation.

Hardness Testing Hardness testing is perhaps the most widely used method for evaluating a nitrided case. Hardness of a steel is used as a cross-reference to many other properties, such as tensile strength and impact strength. Generally, hardness testing of a nitrided case is accomplished with a low load or microhardness load below 1 kg. Do not use the Rockwell machine with a 150 kg load, which will punch through the nitrided case and provide an incorrect reading. Hardness testing can be categorized as either macrohardness or microhardness. Macrohardness testing includes: • • • •

Wilson hardness using low load Vickers hardness testing Rockwell superficial testing Knoop hardness testing

A microtest is usually conducted as a hardness traverse test with load applications below 1 kg. It can be carried out by: • • •

Rockwell superficial testing Vickers hardness testing Knoop hardness testing

168 / Practical Nitriding and Ferritic Nitrocarburizing

Testing Hardness Profiles through the Case The first method is to step-grind through the case and measure the hardness as the “steps” go deeper into the case. This requires a complex grinding method and is somewhat time consuming (Fig. 1). The second method is to cut a sample at right angles to the case and pregrind and polish the cut surface. Once this has been accomplished, a hardness profile can be constructed through the case using a microhardness test method (Fig. 2, 3). This method of hardness testing requires either (a) a test coupon of the same material that is in the same metallurgical condition or (b) a sacrificial component.

Sample Preparation The methods discussed in this section apply to preparation of samples for both microhardness testing and microscopic examination. Cutting, or sectioning, of a sample must be carried out under the coolest conditions possible. This is accomplished by “flood cutting,” where the

Fig. 1

Step-ground specimen for hardness traverse method of measuring depth of medium and heavy cases. Arrows show locations of hardnessindenter impressions.

Fig. 2

Cross section of a round test coupon cut through to expose the nitrided case in preparation for a microhardness traverse test. The exposed surface must be polished before testing.

Chapter 16: Examination of the Nitrided Case / 169

Fig. 3

Results of a microhardness traverse test on a nitrided test coupon

sample is sectioned using an abrasive cutoff wheel while submerged in the cutting fluid. Do not use excessive pressure or localized overheating likely will occur, adversely affecting the cut surface metallurgical condition and subsequent test results. Mixing a rust inhibitor into the coolant will reduce the risk of surface oxide formation on both the sectioned sample and the remaining sample portion. Diamond sawing is another method of cold cutting. This gives a very cool cut and reduces the amount of surface scratching usually caused by abrasive wheel cutting. This improved surface finish considerably reduces the pregrind time (Fig. 4, 5). When using an abrasive wheel cutoff unit, the wheel residuals and the metallic fines that result from sectioning must be removed from the recirculating cooling system. The fines will build up quickly in the drain and delivery hoses, significantly reducing the pipe diameter and thus restricting

170 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 4

Fig. 5

Laboratory abrasive cutoff device

Diamond saw used for sectioning of very hard specimens. Very thin cuts can be made with this tool.

the coolant drainage and delivery system. This applies to the machine sump and the recirculating reservoir, both of which must be cleaned frequently. In addition, the bearings on which the drive motor is connected to the wheel load handle must remain well lubricated. If the bearings become tight, then the force required to depress the abrasive wheel could lead to an excessive load being placed on the sample being sectioned, creating the potential for localized overheating of the sample surface. Vapor degreasing is often used to clean the specimen after it has been machined prior to mounting. Hot vapors of a chlorinated or fluorinated solvent remove oils, greases, and waxes that may be on the specimen.

Chapter 16: Examination of the Nitrided Case / 171

Mounting of the sample prior to examination is necessary for good metallurgical evaluation. Mounting can reduce the risk of “edge rounding,” allowing the formed and diffused case to be accurately observed by both hardness traverse testing and microexamination after etching. Tables 1 and 2 list the properties of commonly used thermosetting and thermoplastic mounting materials. Tables 3 and 4 show typical problems with compression mountings and castable materials. There are essentially two types of mounting systems: the cold-mount system and the hot-mount system. Each has its own advantages and disadvantages. The two-component cold-mount system comprises a polymer and a catalyst that must be accurately measured, mixed, and poured over the specimen in a mold. An identification piece or small plastic clip should also be placed with the sample to identify the surface of the nitrided sample to be examined (Fig. 6).

Table 1 Typical properties of thermosetting molding resins Molding conditions Temperature Resin

°C

°F

Pressure MPa

Time, min

psi

Heat Coefficient distortion of thermal temperature(a) expansion °C °F in./in.°C

Abrasion Polishing rate, rate, µm/min(b) µm/min(c) –5

Transparency

Chemical resistance

Bakelite (wood-filled)

135–170 275–340 17–29

2500–4200

5–12

140

285

3.0–4.5 × 10

100

2.9

Opaque

Attacked by strong acids and alkalies

Diallyl phthalate (asbestos-filled)

140–160 285–320 17–21

2500–3000

6–12

150

300

3.5 × 10–5

190

0.8

Opaque

Attacked by strong acids and alkalies

(a) Determined by method ASTM D 648. (b) Specimen 100 mm2 (0.15 in.2) in area abraded on slightly worn 600-grit silicon carbide under load of 100 g at rubbing speed of 105 mm/min (4 × 103 in./min). (c) 25 mm (1 in.) diam mount on a wheel rotating at 250 rpm covered with synthetic suede cloth and charged with 4 to 8 µm diamond paste

Table 2 Typical properties of thermoplastic molding resins Molding conditions Heating Temperature Resin

°C

°F

Cooling

Pressure MPa

psi

Time (min)

Temperature °C

°F

Pressure MPa

psi

Time (min)

Transparency

Coefficient Heat distortion of thermal Abrasion Polishing temperature(a) expansion, rate, rate, °C °F in./in.°C µm/min(b) µm/min(c) –5

Chemical resistance

Methyl methacrylate

140– 285– 165 330

17– 29

2500– 4200

6

75– 85

165– 185

max

max

6–7

Water, white to clear

65

150

5–9 × 10

...

7.5

Polystyrene

140– 285– 165 330

17

2500

5

85

185– 212

max

...

6

...

65

150

...

...

...

Polyvinyl formal

220

27

4000

...

...

...

...

...

...

Light brown, clear

75

165

6–8 × 18–5

20

1.1

Not resistant to strong acids

45

1.3

Resistant to most acids and alkalies

Polyvinyl chloride

430

120– 250– 160 320

0.7

100

nil

60

140

27

4000

...

Opaque

60

140

–5

5–18 × 10

Not resistant to strong acids and some solvents, especially ethanol ...

(a) Determined by method ASTM D 648. (b) Specimen 100 mm2 (0.15 in.) in area abraded on a slightly worn 600-grit silicon carbide paper under load of 100 g at rubbing speed of 105 mm/min. (c) 25 mm (1 in.) diam mount on a wheel rotating at 250 rpm covered with a synthetic suede cloth and charged with 4–8 µm diamond paste

172 / Practical Nitriding and Ferritic Nitrocarburizing

Table 3 Typical problems of compression mounting materials Problem

Cause

Solution

Too large a section in the given mold area; sharp cornered specimens

Increase mold size; reduce specimen size.

Excessive shrinkage of plastic away from sample

Decrease molding temperature; cool mold slightly prior to ejection.

Absorbed moisture; entrapped gases during molding

Preheat powder or premold; momentarily release pressure during fluid state.

Too short a cure period; insufficient pressure

Lengthen cure period; apply sufficient pressure during transition from fluid state to solid state.

Insufficient molding pressure; insufficient time at cure temperature; increased surface area of powdered materials

Use proper molding pressure; increase cure time. With powders, quickly seal mold closure and apply pressure to eliminate localized curing.

Powdered media did not reach maximum temperature; insufficient time at maximum temperature

Increase holding time at maximum temperature.

Inherent stresses relieved upon or after ejection

Allow cooling to a lower temperature prior to ejection; temper mounts in boiling water.

Thermosetting resins

Radial split

Edge shrinkage

Circumferential splits

Burst

Unfused Thermoplastic resins

Cottonball

Crazing

Fig. 6

Epoxy mounts with binder clips to hold the specimen perpendicular to the polished surface

Chapter 16: Examination of the Nitrided Case / 173

Table 4 Typical problems of castable mounting materials Problem

Cause

Solution

Too violent agitation while blending resin and hardener

Blend mixture gently to avoid air entrapment.

Insufficient air cure prior to oven cure; oven cure temperature too high; resin-to-hardener ratio incorrect

Increase air cure time; decrease oven cure temperature; correct resin-to-hardener ratio.

Resin-to-hardener ratio incorrect; resin has oxidized

Correct resin-to-hardener ratio; keep containers tightly sealed.

Resin-to-hardener ratio incorrect; incomplete blending of resin-hardener mixture

Correct resin-to-hardener ratio; blend mixture completely.

Resin-to-hardener ratio incorrect; incomplete blending of resin-hardener mixture

Correct resin-to-hardener ratio; blend mixture completely.

Insufficient air cure prior to oven cure; oven cure temperature too high; resin-to-hardener ratio incorrect

Increase air cure time; decrease oven cure temperature; correct resin-to-hardener ratio.

Too violent agitation while blending resin and hardener mixture

Blend mixture gently to avoid air entrapment.

Resin-to-hardener ratio incorrect; oxidized hardener

Correct resin-to-hardener ratio; keep containers tightly sealed.

Resin-to-hardener ratio incorrect; incorrect blending of resin-hardener mixture

Correct resin-to-hardener ratio; blend mixture completely.

Acrylics

Bubbles Polyesters

Cracking

Discoloration

Soft mounts

Tacky tops Epoxies

Cracking

Bubbles

Discoloration

Soft mounts

The primary disadvantage of the cold-mount system is the difficulty of obtaining two parallel faces after curing and retrieval from the mold. Its primary advantage is its low cost. The second method is the hot-mount system, also known as compression molding. The mold is usually made of a thermosetting material: • • •

Bakelite (phenolics) Epoxy Diallyl phthalate

174 / Practical Nitriding and Ferritic Nitrocarburizing

This system requires a pressure mold equipped with a heater system to ensure that full curing takes place while the sample and mold material are under compression. The curing time can be from 8 to 15 min, depending on the type of thermosetting system. The rise in temperature from the curing heat will not be sufficient to disturb the surface metallurgy of the nitrided sample. The temperature rise is usually approximately 165 °C (300 °F). Pregrinding. The sample preparation techniques in this section are based on single-mounted samples and not multisampling. Hardness testing requires a polished, unscratched surface. This necessitates pregrinding. The pregrind involves an initial rough grind using 180-grit silicon carbide paper followed by intermediate pregrinding steps using 320-, 400-, and 500-grit papers. Do not spend too much time on coarse-grit grinding: instead, concentrate on careful sample surface preparation using finer grit sizes of 500 and higher. For rough polishing, use an 800- to 1200-grit (maximum) paper. The surface finish quality will be determined by the pregrind wheel rotational speed. Generally, this is accomplished on a rotary wheel running at approximately 350 rpm (Fig. 7). The abrasive silicon carbide paper must be kept well flushed with water. The action of pregrinding will load up the surface of the paper with metal particles as well as mount material from the sample face; this can cause “rescratching.” Ideally, the used pregrind sample paper should be discarded. If the paper is reused, a burnishing/glazing effect can occur on the steel sample surface. This will prevent a true image from being examined after etching. Discarding each paper can become expensive, so the decision rests with the technician or metallurgist.

Fig. 7

Fine grinder for wet or dry grinding of metallographic specimens

Chapter 16: Examination of the Nitrided Case / 175

Some silicon carbide papers and polishing cloths are self-adhesive. The platen wheel must be thoroughly clean and free of adhesive before placing a new paper or cloth onto the platen. If any adhesive contaminant remains, the platen will not present a completely flat surface for the pregrind/rough polish step. The sample can be rough ground—wet or dry—using a rugged belt sander with an initial silicon carbide paper of 180 grit (Fig. 8, 9). If dry grinding is performed, be extremely careful not to cause localized overheating of the sample surface.

Fig. 8

Fig. 9

Coarse grinder for wet or dry grinding of metallographic specimens

Hand grinders using 76 × 280 mm (3 × 11 in.) strips, 230 × 355 mm (9 × 14 in.) sheets, and 100 mm × 137 m (4 in. × 150 yard) rolls of abrasive paper

176 / Practical Nitriding and Ferritic Nitrocarburizing

Another pregrinding method is to use silicon carbide strips of a precut length, laid out on a smooth, flat surface. Once again, sufficient water flow is important to ensure that the silicon carbide strip is well flushed. The secret to reducing the polishing time is careful pregrinding. Polishing follows the pregrinding operation. Do not let the sample sit after the pregrind for any length of time. If the exposed sample surface is wet, it will begin to oxidize, which could cause very slight pitting. The polishing operation should start immediately after pregrinding to reduce the risk of oxide formation. One of two polishing mediums can be used: •

•

Diamond paste. While diamond paste (Fig. 10) is perhaps the hardest material and will produce very highly polished surfaces, additional lubrication is necessary to ensure good distribution of the paste over the polishing cloth to improve polishing efficiency. Recommended particle sizes for the final polish are 1 µm or 0.3 µm. Diamond pastes are very expensive. Aluminum oxide slurry in suspension. Aluminum oxide slurries for final polishing can have particle sizes of 1 µm down to 0.5 µm. The particles are suspended in water or in paste or powder form. Selection depends on personnel and availability. Such slurries are self-lubricating and economical.

The flat wheel of the polishing machine should be tightly fitted with a well-lubricated soft-nap polishing cloth. The rotational speed of the wheel should be approximately 400 to 450 rpm. Once the polishing is complete, rinse the polished sample under lukewarm running water, then

Fig. 10

Application of diamond paste to a nylon cloth for preliminary polishing

Chapter 16: Examination of the Nitrided Case / 177

rinse with alcohol. Examine the sample for possible surface staining and/or residual surface grease. Do not touch the surface with your fingers; any body oil present will inhibit etching. Great care must be taken during the entire pregrind and polish procedure to prevent edge rounding, a condition that causes difficulties when trying to align the diamond indentor at the sample edge and when trying to focus on the edge during microscopic examination.

Microhardness Testing Case Depth. After polishing, hardness testing must be performed before microscopic examination of the surface. To establish the effective case depth accomplished by nitriding, the transition hardness between case and core as accepted by the International Organization for Standardization (ISO) is 532 DPN (diamond pyramid numeral) (approximately 53 HRC). There is still much discussion regarding case depth measurement. Many claims are made regarding the achievable case depth of some derivative nitriding processes. Be cautious when being told of deep case depths. The ISO and the European DIN specification writers have invested a great deal of time in defining and understanding case depth. However, many processes do not define the case depth measurement, and others state that “the case depth achieved is x.” The conclusion drawn by the uninitiated is that the total case depth is hard, but this is not true. In this writer’s opinion, the real case depth is the effective case depth and not the total case depth. Quoting the total case depth (which while relevant, is not always well defined) is misleading. The microhardness test is based on the resistance to indentation principle. As with any hardness testing procedure, the unit must be tested with the standard test block and calibrated monthly. The test can be carried out using a load of 10 g minimum up to 1000 g. The indentor must be positioned at precisely 90° to the sample being tested. If the immediate sample face is not at an exact right angle to the indentor, a uniform indentation will not be obtained and accurate readings cannot be made in the diffused case. When setting the x-y traverse, carefully note the starting point from the surface transversely across the nitrided case. When starting the traverse through the diffused case, it is important to start as close to the edge (surface) as possible to accomplish an almost true surface reading. If the indentation is made too close to the edge, the diamond indentor can “jump” off the edge and may likely be damaged.

Etching of the Sample The sample should be etched only after microhardness tesing has been completed. If etched beforehand, the hardness indentations may not be clear enough to read accurately.

178 / Practical Nitriding and Ferritic Nitrocarburizing

Etchants Nital is perhaps the most commonly used etchant for a visual case metallurgical examination. Simple-to-prepare, nital consists of 4 to 6 mL HNO3 (nitric acid) and 94 to 96 mL ethyl alcohol (or denatured alcohol). The common error when using nital involves etching time. The sample is usually oversoaked in the etchant, resulting in a surface that is too dark for accurate examination. The specimen should be etched for approximately 3 to 8 s, taking care not to overetch. The exact etching time, of course, depends on the type of steel being examined. Nital will reveal the compound zone on the immediate surface and stable nitrides in the case directly below. It will not, however, show the two phases of the compound zone. The compound zone will show as the surface white layer. In the case of a low-alloy steel or a plain carbon steel, the iron nitride precipitates will be seen as well-defined straight lines, below which the steel core will exhibit iron nitride precipitates. Picral is another common etching medium used to examine a nitrided surface. It is not complex, consisting of 4 g picric acid and 100 mL ethanol. The addition of 0.5 to 1% zephiran chloride improves etch rate and uniformity. Specimens should be immersed for 3 to 15 s, taking care not to overetch. Care should be exercised when using picric acid (refer to the Material Safety Data Sheet, MSDS). Proper storage of picric acid powder is critical to operator and technician safety, and the powder should be kept very moist or even under water. Tips for Using Etchants. Here are a few helpful guidelines: • • •

• • • • •

•

When mixing etchants, always add reagents to the solvent unless specific instructions indicate otherwise. Where water is given as the solvent, distilled water is preferred because the purity of tap water varies. Methanol is usually available only as absolute methanol. When using this alcohol, it is imperative that approximately 5 vol% of water be added whenever an etchant composition calls for 95% methanol. Generally the alcohol used is denatured alcohol. For conversion of small liquid measurements, there are approximately 20 drops per 1 mL. Etching should be carried out on a freshly polished specimen. Gentle agitation of the specimen or solution during etching will result in a more uniform etch. The etching times given are only suggested starting ranges and not absolute limits. In electrolytic etching, a direct current (dc) is implied unless otherwise indicated. An economical source of dc current for small-scale electrolytic etching is a standard 6 V lantern battery. Microscope objectives can be ruined by exposure to hydrofluoric acid fumes from etchant residue inadvertently left on the specimen. This

Chapter 16: Examination of the Nitrided Case / 179

•

problem commonly occurs when the specimen or mounting media contain porosity and when the mounting material (such as Bakelite) does not bond tightly to the specimen, resulting in seepage along the edges. In all cases, extreme care should be taken to remove all traces of the etchant by thorough washing and complete drying before placing the specimen on a microscope stage.

Etchant Removal. Once etching has been completed, quickly wash the sample under clean, cold running water. Otherwise, the etchant will continue to etch the specimen surface. After thorough water rinsing, rinse the surface with clean denatured alcohol. Then take a household hair dryer set on very low heat and blow-dry the alcohol across the face of the specimen in one direction. Ensure that the sample is dry on all sides of the mount and then examine. If surface stains have occurred, lightly repolish the specimen on the rotating polishing cloth wheel and re-etch, completing the procedure once again.

Safety Precautions Here are some practices that will contribute to the safe preparation of specimens. The list is not exhaustive. Obviously, care must be taken when machining samples. Preparing Etchant. When preparing an etchant, the operator or technician is using corrosive and violently reactive acids. It is of utmost importance to follow these safety precautions (Ref 1): • • • • •

•

Wear the proper safety clothing, including goggles, face mask, rubber gloves, rubber apron, long sleeves, and shoe protection. Mix acid to water, not water to acid (Fig. 11). Use proper solution mixing, storage, and handling equipment and facilities (e.g., acid-resistant glass). Wipe up all spills and leaks, no matter how small, using the “spill kit” and “spill treated wipers.” Dispose of all solution not correctly identified by composition and concentration. When in doubt, throw it out—but do so in a responsible, approved manner (not down the drain). Read and follow the MSDS for storage and handling of all chemicals. Observe all printed cautions issued by the reagent, acid, or chemical manufacturer.

Vapor degreasing is a cleaning process that uses the hot vapors of a chlorinated or fluorinated solvent to remove soils, particularly oil, greases, and waxes. Extreme care should be taken in handling the vapor. Follow these precautions (Ref 2): • •

Provide adequate ventilation. Do not breathe the vapor fumes.

180 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 11 • • •

Add acid to water, not vice versa. Wear a full apron, a full face mask, rubber gloves, long sleeves, and shoe protection.

Wear protective clothing, including gloves, face mask, and apron. Do not allow any of the liquid to come in contact with the skin. Ensure that the vapor does not overheat or come in contact with highintensity light, which could cause formation of gases such as phosgene gas, dichloroacetyl chloride gas, and carbon dioxide gas.

Once the surface of the component has been degreased, care must be taken in its handling. Use cotton gloves or rags when moving the component into the nitriding retort. Otherwise, contamination from fingerprints will cause resistance to nitrogen penetration, leading to soft spots.

Optical Light Microscopy Optical light microscopy is the metallurgist’s most important tool for observing the structure of the nitrided case. Generally, a sample must be etched before the nitrided case can be seen. The choice of microscope should take into account a number of factors: • • • •

How many samples are examined daily? Is it necessary to observe only the nitrided case? Is a permanent record (i.e., a micrograph) needed? Is visual imaging analysis required?

Chapter 16: Examination of the Nitrided Case / 181

Microscopes range widely in cost and performance. The availability of spare units and service from the supplier is another consideration. For quality control and troubleshooting, light microscopy up to 1000 to 1200× will cover most metallurgical requirements. When research of the development of the case and surface metallurgy is involved, the user can usually obtain scanning electron microscopy services through a local metallurgical or materials science school. Examples of case microstructures are shown in Fig. 12 to 18.

Fig. 12

AMS 6470 steel with 0.15 to 0.35% Pb added, oil quenched from 900 °C (1650 °F), tempered 2 h at 605 °C (1125 °F), surface activated in manganese phosphate, and gas nitrided 30 h at 525 °C (975 °F). Structure is a white layer of Fe2N and a matrix of tempered martensite. 2% nital. 400×

Fig. 13

Same material and heat treating conditions as described in Fig. 12, except nitrided 36 h. The depth of the nitride layer has increased, and platelets of iron nitride can be seen in the case. 2% nital. 400×

Fig. 14

Same steel and processing as Fig. 12, except slack quenched and ground heavily before nitriding. Because the surface was not chemically activated before nitriding, nitrogen diffusion was retarded. 2% nital. 400×

Fig. 15

4140 steel, oil quenched from 845 °C (1550 °F), tempered 2 h at 620 °C (1150 °F), surface activated in manganese phosphate, and gas nitrided 24 h at 525 °C (975 °F). Structure is white layer of Fe2N, Fe3N, and Fe4N, and tempered martensite. 2% nital. 400×

Fig. 16

H13 steel, heated to 1030 °C (1890 °F) in a vacuum, quenched in nitrogen gas, triple tempered at 510 °C (950 °F), surface activated in manganese phosphate, and gas nitrided 24 h at 525 °C (975 °F). White surface layer is iron nitride. Grain-boundary networks of nitride are present throughout the martensitic case. 2% nital. 300×

Chapter 16: Examination of the Nitrided Case / 183

Fig. 17

18% Ni maraging steel (300 CVM), solution treated 1 h at 815 °C (1500 °F), surface activated, and gas nitrided 24 h at 440 °C (825 °F). Etching has made the nitride surface layer and grain-boundary nitrides appear black. Modified Fry’s reagent. 1000×

Fig. 18

4140 steel, quenched and tempered to 30 HRC, then ion nitrided 24 h at 510 °C (950 °F). Monophase surface layer of Fe4N, plus a diffusion zone of nitride containing tempered martensite. Nital. 750×

REFERENCES 1. N.I. Sax, Handbook of Dangerous Materials, Reinhold Publishing, 1951 2. H.B. Elkins, Chemistry of Industrial Toxicology, John Wiley & Sons, 1959 SELECTED REFERENCES • B.L. Bramfitt and A.O. Benscoter, Metallographer’s Guide: Practices and Procedures for Irons and Steels, ASM International, 2002 • C. Johnson, “Metallography, Principles & Procedures,” Leco Corporation, St. Joseph, MI

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p185-192 DOI: 10.1361/pnafn2003p185

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

17 Troubleshooting

PROBLEMS often occur during nitriding, just as with any other heat treatment process. They can take the form of: • • •

Process problems Steel problems Machining problems

Troubleshooting is a process of elimination and plain old detective work. One must be both observant and systematic during the troubleshooting procedure.

Gas Nitriding Surface Cleanliness. A typical problem with gas nitriding originates from part surface cleanliness. If the work surface is contaminated with a hydrocarbon-based substance such as cutting oils, machining oils, lapping compounds, or fingerprints, the affected area will not successfully nitride, and the area below it will be soft. The remedy is to ensure absolute surface cleanliness prior to gas nitriding. This can be accomplished by degreasing using either chemical or ultrasonic methods. If the part surface is contaminated with a cutting fluid that is chlorideor sulfide-based, then serious surface pitting and loss of hardness may result. It can be safely said that any grease or acidic compound on the surface of a steel is very likely to cause nitriding problems. Loss of Gas Dissociation. When gas dissociation cannot be achieved during the nitriding process, something within the process hardware is taking the gas flow away from the components being nitrided. Check that there is no restriction in the feedline from the ammonia gas source to the process chamber. If the line is internally oxidized, it will restrict the ammonia flow. Internal oxidation of the process gas-delivery system

186

/ Practical Nitriding and Ferritic Nitrocarburizing

results from the incorrect pipe material selection. Check also that the inlet pipe to the process chamber is clean and clear. If the support fixture for the work is made from a low-alloy material such as mild steel, then it will pull the dissociated gas away from the workpiece. This also applies to the process chamber fabrication material. Even if the process chamber is made from the correct material, it can begin to pull the ammonia necessary for the nitriding process away from the workpiece if it is reaching the end of its useful life, or if it has become saturated with nitrogen. This can be remedied by shot blasting the inside surface of the process chamber or by regenerating the chamber by heating up to a temperature of 785 °C (1450 °F), followed by light shot glass bead blasting or sandblasting. Surface Discoloration. When the process chamber is opened after nitriding, the parts occasionally may be discolored with an almost rainbow effect. The discoloration is an oxide formation on the steel surface and is in no way detrimental to the part. In fact, it will enhance the surface corrosion resistance. This is the principle behind the oxynitride procedure, where the steel surface is deliberately oxidized in a controlled manner to improve its corrosion resistance. Many engineers have been unnecessarily concerned regarding surface discoloration after the gas nitriding process. Case Exfoliation. When the case exfoliates, or peels off, this is usually indicative of surface decarburization present on the immediate surface of the steel. Surface decarburization can occur as a result of: •

•

Insufficient stock removal from the steel surface, leaving a decarburized layer that will result in defective surface metallurgy. The remedy is to ensure removal of more than 10% of the surface stock thickness. A decarburizing condition during the preheat treatment operation that has left the surface seriously decarburized

Components that exhibit these surface conditions cannot be salvaged and are scrap. Should the surface exhibit a failure of the case due to chipping (e.g., on the pressure face of a gear), it is usually indicative of an aluminum-bearing (Nitralloy) steel. The aluminum-bearing steels have very high surface hardness values and cannot be subjected to an impact load condition. If a gear tooth tip begins to chip off, it generally indicates a supersaturated case formation of nitrogen, where the solubility limit of nitrogen in iron has been exceeded, resulting in nitride networking and a brittle case. This usually takes place on sharp corners and edges. The only course of action is to investigate the gas dissociation control. If the surface of the case exhibits light, flaky peeling, some type of surface contamination may be present. Investigate the cleaning treatments

Chapter 17: Troubleshooting

prior to nitriding. Surface contamination is the most probable cause of any flaking or exfoliation of the nitrided surface. Contamination may be due to: • • • • •

Grease Oil Residual cutting fluids Residual heat treatment salts (see discussion below on salt bath nitriding) Insufficient precleaning

If the flaking is persistently seen on most loads processed through the furnace, the source of contamination may likely be within the furnace process retort. Examine the internal faces of the retort for signs of contamination. The interior walls may simply require a cleaning operation or wipe down. If necessary, the wipe down can be accomplished by using an alcohol-based solution and hand wipes. (Appropriate ventilation should be used to ensure that the alcohol fumes do not overcome the operator.) It may be advisable to consider either a burnout of the process retort or complete replacement. Orange Peel Effect. If the workpiece surface shows what can only be described as an orange peel effect, with the appearance of dimples, then once again it can be traced back to surface decarburization. The remedy is to investigate the surface machining and the preheat treatment conditions. The part is not salvageable unless there is a deep case and the surface dimpling can be ground off. This, however, is not a practical solution. Case crushing will occur when the case is too thin and the core hardness too low in relation to the designed surface load conditions. This can be seen on the surface of nitrided gears. It can be avoided by: • • •

Increasing the case depth Improving the core hardness by reducing the previous tempering temperature, or by reducing the nitride process temperature Changing the steel chemistry if the steel will not give the required core hardness on preheat treatment

Salt Bath Nitriding Two criteria are necessary to maintain a good nitriding salt bath and ensure reasonably repeatable nitriding conditions: process salt chemistry and bath cleanliness. Process Chemistry. The subject of discussion here is salt bath nitriding and not ferritic nitrocarburizing. There is a great distinction between the two processes. It is important to understand the chemical makeup of the nitriding salt. Once the salt chemistry is understood, along with the basics of salt titration for measuring nitriding activity, the bath should be titrated at least once per day in order to maintain a consistent bath activation. The usual practice is to titrate for cyanide only, not carbonate.

/ 187

188

/ Practical Nitriding and Ferritic Nitrocarburizing

Bath Cleanliness. Because there are many sources of bath contamination, the bath must be regularly desludged. Desludging should take place at least once per day, or even once per shift if the bath is operating on a shift basis. This is extremely important. The main contaminant in the bath comes from iron oxide and the precipitation of products of the carbonate content. The sources of iron oxide are fixtures, soft iron support wire for suspending parts into the bath, and, if the bath is made from mild steel, from the inside of the bath itself. Nitriding baths usually are fabricated from mild steel or a low-carbon or low-alloy steel but in some instances can be made from a heat-resisting stainless steel. Postwashing Treatments. If the workpieces have been previously heat treated through salt baths such as neutral salts followed by a quench into a marquenching salt, thorough postwashing is necessary to prevent exfoliation or flaking of the case. Salt Bath Nitriding of Alloy and Tool Steels. When nitriding the higher alloyed steels and tool steels, the nitrogen diffusion rate will not be as great as for the lower alloy steels. Therefore, do not confuse a shallow case as being caused by the cyanide level of the bath; it is simply a matter of a slower rate of diffusion. Cycle times are usually short, in the region of 5 min up to 180 min, without quench.

Ion Nitriding With the ion nitriding process, process problems as well as metallurgical problems can occur. Process problems generally will affect the process metallurgy. Care of the furnace is mandatory, even in terms of initial installation of the equipment. Considering the cost of water, the water-cooling system usually is a closed-loop system that goes through a water chiller or cooler. The system must be piped from the inlet of the cooler/chiller to the return of the chiller/cooler. If the pipework is made from galvanized steel tube, then very aggressive corrosion will begin to occur that over time will restrict water flow. Use either copper or stainless steel tubing. O-Ring Seal. Another common problem involves the overuse of vacuum grease on the O-ring seals before commencing the process cycle. The grease can be your friend, but it can also be your enemy in terms of vacuum accomplishment. Like sticky flypaper, too much grease on the seal will attract dust. The amount of vacuum grease required is as much as it takes to make the tip of your finger look wet. It is not necessary to squeeze the grease from the tube into the palm of your hand. Wipe the seal clean before closing the process chamber and then apply the vacuum grease sparingly. Do not use any other type of grease. Process Vessel Cleanliness. Bell-type plasma nitriding furnaces require regular cleaning. Each week, wipe the inside of the process chamber walls to remove any carbon particles and sputtered particles, taking care to keep particle matter from falling onto the O-ring seal. It is important that the vacuum sealing arrangement be protected.

Chapter 17: Troubleshooting

Backstreaming. Each week, check the vacuum pump oil in terms of level and condition. Ensure that the vacuum pipe connecting the vessel to the pump is fitted with a valve or trap to prevent the possibility of backstreaming of vacuum pump oil when a pressure differential occurs between the process chamber and the vacuum pump. Overheating at a workpiece corner or edge can be visually seen through the nitride furnace sight port. This likely is due to a problem with high voltage, high pressure, or a combination of both. Check the vacuum pressure level and decrease the pressure. Alternatively, reduce the process voltage. These adjustments, separately or together, will reduce the overheating. The same applies when a hollow cathode is observed. If the localized overheating is within 55 °C (100 °F), then nonuniform case metallurgy will occur. Temperature uniformity within the process retort, be it gas or plasma nitriding, is mandatory. Loss of Nitriding. If the plasma glow seam provides only partial coverage, the portion of the workpiece that is not covered by the glow will not be nitrided. Partial coverage is caused by a process chamber pressure that is too high toward atmospheric pressure. The only way to correct this is to reduce the pressure until the glow seam uniformly covers all the surfaces to be nitrided (Fig. 1).

Fig. 1

Complete glow seam coverage, ensuring complete nitriding of all surfaces. Courtesy of Plateg GmbH

/ 189

190

/ Practical Nitriding and Ferritic Nitrocarburizing

Another condition that can cause the loss of glow seam coverage is when the surface area available for nitriding exceeds the original design characteristics of the furnace. This is particularly true for continuous direct current (dc) systems. The design basis of the continuous dc unit is the power ratio of current density in relation to surface area to be treated. Therefore, if the surface area exceeds the original design capabilities of the system, then there will be a loss of plasma coverage and subsequent partial nitriding of the component. Arc discharge is likely to occur if the plasma pulse duration is too long. The solution is to shorten the pulse duration time to the point where no discharge occurs. Too high a pulse voltage can cause the discharge, especially if the pulse duration time is long. All of this is based on the assumption that the operator can see the arc discharge through the furnace viewing port. If the operator is not in attendance, the arc discharge will not be seen. It will, however, be evidenced by the appearance of the workpiece after the cycle is complete. The furnace technician must understand the possible causes of the arc discharge phenomenon and be able to program the unit so that arc discharging will not occur. For continuous dc plasma furnaces, the technician can only adjust the process voltage and/or the process pressure (Fig. 2, 3). Chipping on corners and edges is usually caused by the oversaturation of nitrogen in iron. Remember, the solubility of nitrogen in iron at traditional nitriding temperatures is between 5 and 7 wt% (changing, of course, with changes in alloying elements in the steel). Oversaturation means that excess nitrogen is not in solution, but has migrated to grain boundaries at corners and edges, forming nitrides and thus weakening the area. It is only possible to observe this phenomenon by light microscopy. The excessive nitrides at the grain-boundary areas will be seen as excessive

Fig. 2

Workpiece during plasma nitriding with continuous dc glow discharge. Numerous micro-arcs are visible on the workpiece surface and may produce microscopic damage. A large concentration of micro-arcs can result in an avalanche-like increase in power. A big arc will form, destroying the surface by melting due to overheating. Courtesy of Plateg GmbH

Chapter 17: Troubleshooting

Fig. 3

The same workpiece as in Fig. 2, but during plasma nitriding with pulsed dc glow discharge. Conditions such as vacuum pressure, gas mix, and power input remain the same. By using pulsed dc with a repetition frequency of about 10 kHz, the formation of micro-arcs is suppressed. Courtesy of Plateg GmbH

white lines in the corner or edge area. The corrective action is to reduce the nitrogen-to-hydrogen ratio or reduce the process temperature. However, once the nitride networking has occurred, it cannot be removed, but only adjusted for on the next process cycle. It is thus important to place a test coupon in the load (preferably of the same material and condition as the parts being processed) for metallurgical examination after processing. Low surface hardness after ion nitriding can be attributable to a number of possible occurrences, even if the case has formed. It usually indicates a low nitriding potential, or low nitrogen-to-hydrogen ratio. This can be determined either optically or by microhardness testing. If the unit uses pulsed dc, low surface hardness could indicate a problem with pulse duration and repetition. Power that is on for too short a time period and then off too long means a loss of nitriding effect on both the surface and the case depth. Surface contamination is another possible cause of low surface hardness. Some cutting fluids are silicone based; any residual deposits will act as a barrier to nitrogen diffusion. Surface flaking is usually caused by some form of surface contamination. The contamination can be due to decarburization from previous heat treatment operations, machining operations using coolants that contain silicones, or inadequate cleaning after salt bath treatment. Another problem source is contaminated grit on glass bead blast or shot blast. Precleaning is mandatory for both gas nitriding and plasma nitriding, regardless of what system is used.

/ 191

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p193-200 DOI: 10.1361/pnafn2003p193

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

18

What Is Meant by Ferritic Nitrocarburizing? FERRITIC NITROCARBURIZING accomplishes surface treatment of a part in the ferrite region of the iron-carbon equilibrium diagram (Fig. 1). As the process takes place in the ferrite region, both nitrogen and carbon diffuse into the steel surface. The process is categorized as a thermochemical treatment and is carried out at temperatures between 525 and 650 °C (975 and 1200 °F); the typical process temperature is approximately 565 °C (1050 °F). The purpose of the process is to diffuse nitrogen and carbon atoms into a solid solution of iron, thus entrapping the diffused atoms in the interstitial lattice spaces in the steel structure (Ref 1). As with the nitriding procedure, there are many methods and derivatives of ferritic nitrocarburizing. These are discussed in the chapters that follow.

Process Benefits Ferritic nitrocarburizing improves the surface characteristics of plain carbon steels, low-alloy steels, cast irons, and sintered ferrous alloys. As described in later sections of this chapter, resistance to wear, fatigue, and corrosion are improved with the introduction of nitrogen and carbon. Scuffing resistance means the resistance to wear on the metal surface. This is accomplished by changing the nature of the surface compound layer, which is also known as the white layer. The completed compound layer will form with both epsilon (ε) and gamma prime (γ ′) phases. The dominant ε-phase resists abrasive wear. Fatigue properties of steel are greatly improved by altering the composition of the compound layer. This means that treated steel has greater resistance to fatigue failure than an untreated steel (Ref 1). Corrosion Resistance. After ferritic nitrocarburizing, steel parts can withstand many hours in a salt spray environment, whereas an untreated plain carbon steel will fail the corrosion test very rapidly.

194

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

The iron-carbon equilibrium diagram. The nitrocarburizing process is carried out in the ferrite region (alpha iron) of the diagram.

Low Distortion. Another major advantage of the ferritic nitrocarburizing process is that the procedure is carried out at a low temperature that prevents phase changes in the steel (from ferrite to austenite), thus reducing the risk of distortion. Distortion is the result of the release of induced stresses, the thermal shock of quenching, and the risk of incomplete transformation to martensite. No phase change occurs during the ferritic nitrocarburizing treatment.

Chapter 18: What Is Meant by Ferritic Nitrocarburizing? / 195

Early History of Ferritic Nitrocarburizing Ferritic nitrocarburizing has been a proven process for many years and is now gaining much acceptance by engineers. This increased interest in the process, the author believes, is due to engineers gaining a better understanding of materials selection and metallurgists gaining a greater understanding of process capabilities and restrictions. In addition, many furnace manufacturers want to serve their clients by developing new and more efficient process methods and equipment. The early methods of ferritic nitrocarburizing were accomplished in low-temperature (550 °C, or 1020 °F) salt baths working on the principle of the decomposition of cyanide to cyanate (in the ferrite region). Imperial Chemical Industries in England pioneered the salt bath process, which was called the “Sulfinuz” treatment (Ref 2). The salt also contained a sulfur compound in its chemistry. The process was based on the formation of: •

• •

Nitrides: The nitrides were formed as a result of the nitrogen component contained in the cyanide salt. The nitrogen diffused into the steel to form iron nitrides in low-alloy steels and stable nitrides in higher alloyed steels. Carbon: The carbon was supplied from the salt in limited quantities and formed carbides, interspersed with the formed nitrides. Sulfides: The sulfur addition to the salt formed sulfides in the case, providing a self-lubricating property.

The action of the molten salt at the process temperature also caused slight surface porosity on the treated steel. This allowed the surface pores to become minute reservoirs, retaining lubricant on the immediate surface. The net result was that the treated component resisted scuffing and exhibited excellent resistance to frictional wear problems. The process was a great success with high-speed spindles and high-speed cutting tools. It did, however, require careful salt bath analysis on a daily basis (Ref 1). Another challenge of the process was that the salt was not very water soluble. The treated component required extensive hot water cleaning after treatment. Cleaning became a major issue. Problems associated with salt bath processing led to experimentation with gaseous methods of ferritic nitrocarburizing. Experiments were conducted in the late 1950s with gaseous methods by Cyril Dawes of Joseph Lucas Ltd. in England. The company successfully applied for a patent on the process in 1961 (Ref 3). The gaseous procedure produced a porous layer very similar to the layer produced with the Sulfinuz process (with the exception of forming surface sulfides), which claimed to provide good antifrictional properties. The process patent stated that the gaseous atmosphere consisted of ammonia,

196

/ Practical Nitriding and Ferritic Nitrocarburizing

with a hydrocarbon gas and other small amounts of carbon-containing gases (Ref 3). An important study that contributed greatly to the scientific understanding of gas nitrocarburizing treatments and compound layer structure was published by Prenosil in 1965 (Ref 4). As a result of the study, many companies developed variations of the original patented process and the procedure was accepted by engineers and metallurgists alike. Advances in gaseous nitrocarburizing did not stop or hinder the process technique of using salt baths for the ferritic nitrocarburizing process. If anything, it spurred on the salt manufacturers to develop more environmentally friendly salts and cleaner procedures. Degussa of Germany developed the salt bath process of “Tufftride,” a two-component process that formed both nitrides and carbides in the immediate surface of the steel (Ref 1). The process will produce only very shallow case depths, approximately 0.05 mm (0.002 in.) deep, but with high surface hardness values, good fatigue properties, and excellent corrosion resistance. The process cycle times are relatively short (in the region of 1.5 h), followed by a quench (Fig. 2). Once again, the process relies on the decomposition of cyanide to cyanate, which is accelerated by the introduction of a titanium aeration tube. The aeration tube passes air through the molten salt from the bottom of the salt pot. The system requires good operational maintenance in terms of regular bath desludging, salt analysis, and periodic regeneration. This requires raising the bath temperature to 575 °C (1070 °F) and holding for approximately 2 h, followed by another desludging operation. The purpose is to precipitate out of the molten salt any free iron originating from work support baskets and fastening wire used to wire the components in place in the work basket (Fig. 3).

Fig. 2

Typical time-temperature process cycle for a ferritic nitrocarburizing procedure using salt baths

Chapter 18: What Is Meant by Ferritic Nitrocarburizing? / 197

Fig. 3

Work-holding fixtures and wiring techniques used in liquid nitrocarburizing. (a) Typical holding basket for small parts, equipped with a funnel for loading parts into the basket without splashing. Funnel, which is made of sheet metal, also insures that parts are coated with salt before nesting together. Basket may be made of carbon or alloy steel rod and steel wire mesh. Work must be free from oil, or the parts will stick together. Parts must be dry. (b) Inconel basket of simple design. Upper loop of the handle is for lifting; lower loop accommodates a rod which supports the basket over the furnace. (c) Simple basket with trays, intended for small parts. Trays provide a maximum of loading space without adversely affecting circulation. Entire fixture is made of Inconel. (d) Netted fixture, of Inconel, for holding small parts with a head or shoulder. (e) Methods of wiring small parts. Black annealed steel wire is used for parts weighing less than 10 lb; annealed stainless wire is used for heavier parts. (f) Hooks, made of nickel alloy rod, for holding circular parts. (g) Method for holding large parts in which tapped handling holes are available or can be provided. Nickel alloys are used for such fixtures because of the need for high-temperature strength. Resistance to oxidation is not a factor, as liquid carburizing salts are reducing. (h) Rack for holding six small crankshafts; exploded view shows a crankshaft in position. (i) Special rack for carburizing the outside diameters of bearing races. Holding plates are made of mild steel; rods, of Inconel.

With the advent of pulsed plasma technology in the early 1980s for ion nitriding, it did not take long to realize that another method of ferritic nitrocarburizing had been discovered. This procedure was soon commercialized. Advantages include faster process cycle times, less surface cleaning and preparation, deeper case formation, and better control of surface metallurgy formation. Equipment is now being built that is capable of performing

198

/ Practical Nitriding and Ferritic Nitrocarburizing

both pulsed plasma ion nitriding and ferritic nitrocarburizing in the same process chamber and with the same pulsed power pack. The procedure offers a more controllable, repeatable surface metallurgy.

Why Ferritic Nitrocarburize? The physical benefits of ferritic nitrocarburizing have been listed. The choice to nitrocarburize is an economic one, when compared to other methods of achieving the same benefits. Figure 4 presents an approximate cost comparison of various surface treatments (Ref 5). Besides the direct cost of the equipment, the process selection procedure should consider the total investment costs. Cost of floor space involves direct purchase or rental of space. Remember, floor space also includes storage area for fixtures and fittings and workload preparation area. Installation costs are sometimes overlooked. The cost of installation means the cost of unloading equipment from the delivery vehicle and positioning the equipment in place. Will riggers need to be hired? It also means the cost of a new facility if one is built to accommodate the new equipment, including all plumbing, electrical wiring, gas delivery systems, water delivery system, and effluent exhaust system. Thermochemical: Carburizing Nitriding Nitrocarburizing Electrochemical: Chromium Cobalt + Cr3C2 Electroless: Nickel Plasma sprayed: WC-Co Al2O3 Combustion gun sprayed: 13% Cr wire Ni-Cr-B and fuse Ni-Cr-B + WC and fuse Surface weld: Iron-base Cobalt-base Vapor deposited: CVD TiC PVD TiN Cost

Fig. 4

Approximate relative costs of various surface treatments. Source: Ref 5

Chapter 18: What Is Meant by Ferritic Nitrocarburizing? / 199

Cost of Insurance and Freight. What type of crating or container will be used? After loading is complete, a visual inspection should be carried out; if possible, digital photographs should be taken of the load in its position in the event of possible insurance claims when the contents are later uncrated. Other related costs and concerns include: • • •

•

• •

Loading and delivery from the manufacture site to the point of departure Type of shipping line (conference or nonconference) Paperwork delays. If the equipment is shipped internationally, incomplete or improper paperwork can cause serious delays at either the port of departure or the port of arrival. Such delays can be very costly. Check on daily demurrage rates and duties payable. Insurance. The equipment must leave the manufacturing facility fully insured. Before installation, check the suitability of the intended site with the insurance carrier. Does the room or building have the necessary fire protection? Would the existing fire protection system damage the new equipment? Road transport from the port of arrival. Road transportation permits may be necessary if the vehicle load is considered a wide load. Access. Be sure that before the equipment arrives, doors and wall apertures are large enough to allow easy access of the equipment into the facility. It can be embarrassing if the furnace will not fit through the door. Preplanning models can sometimes be used to navigate large equipment through plants.

Operating costs include materials, energy, disposal of spent chemicals, labor (including training), rejected materials, and time. All these costs must be evaluated on a per item or other basis before making a final decision.

Training To ensure that the furnace goes together the first time (and hopefully starts the first time), at least two primary discipline people—the operating person and the maintenance person—should visit the manufacturing site when the furnace is being assembled. They also should be present after the hot trials to see how the furnace is dismantled. Photographically document the critical assembly areas using a camera or video recorder. Project training can then be broken down into: •

•

An understanding of both the process and its results: This means understanding the process principles, the method of nitrogen diffusion, and the expected results in relation to the steel being treated. An understanding of the equipment performance: This means understanding the operation, functions, and capabilities of the equipment, as well as reactions of the process in relation to part geometry.

200

/ Practical Nitriding and Ferritic Nitrocarburizing

While many more considerations may arise when preparing to ship a furnace from one place to another, the previously mentioned ones will serve to stimulate thinking between the team responsible for delivering the equipment and the client. In order for a furnace project to be successful for both the purchaser and the seller, there must be clear lines of communication regarding each party’s responsibilities. This must include expectations of performance from both the furnace manufacturer and the client. REFERENCES 1. T. Bell, Ferritic Nitrocarburizing, Met. Eng. Q., May 1976, reprinted in Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 266–278 2. The Cassel Manual of Heat-Treatment and Case Hardening, 7th ed., Imperial Chemical Industries Ltd., United Kingdom, 1964 3. Joseph Lucas Ltd., United Kingdom, British Patent 1,011,580 4. B. Prenosil, Structures of Layers Produced by Bath Nitriding and by Nitriding in Ammonia Atmospheres with Hydrocarbon Additions, Härt.-Tech. Mitt., Vol 20 (No. 1), April 1965, p 41–49 (BISI translation 4720) 5. J.R. Davis, Ed., Surface Engineering for Corrosion and Wear Resistance, ASM International, 2001, p 191

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p201-218 DOI: 10.1361/pnafn2003p201

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

19 Salt Bath Ferritic Nitrocarburizing

FERRITIC NITROCARBURIZING is a diffusion process that is a modified form of nitriding, not a form of carburizing. In the process, nitrogen and carbon are simultaneously introduced into the steel while it is in the ferritic condition, that is, at a temperature below which austenite begins to form during cooling. As shown in Fig. 1, methods of accomplishing the process include salt bath procedures, gaseous methods, and ion (plasma) procedures, as well as the process trade names. There are slight variations in the processes identified by the different trade names, as well as slight variations in the resulting surface metallurgy, but overall they all constitute ferritic nitrocarburizing. The process is carried out at a higher temperature than the nitriding procedure, generally in the region of 540 to 625 °C (1000 to 1155 °F). Case depth, once again, depends on the residence time at the selected process temperature. Early work was reported by Professor Tom Bell on the Sulfinuz salt bath process, which was based on the diffusion of cyanide-based salts with sulfur diffusion (Ref 1). He further reported on the Degussa Tufftride process, which involves the decomposition of cyanide to cyanate (Ref 2). A gaseous process that is based on a mixed-gas composition of ammonia and propane in equal proportions further demonstrated that the compound layer would be predominantly ε-phase in the immediate surface of the treated steel (Ref 3). The same investigator, Prenosil, found that the introduction of oxygen at the end of the process cycle oxidized the immediate surface of the steel due to the higher dew point levels in the furnace process chamber. The oxide layer was found to have a higher resistance to surface corrosion than if the steel was left untreated. The key elements of the process are: • • •

Temperature Time Suitable nitrogen source

• •

Suitable carbon source Sealed and controlled environment

202 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Various trade names for gaseous, salt bath, and ion (plasma) ferritic nitrocarburizing processes. Fluidized-bed procedures also are available.

It is now more than 40 years since the development of the low-temperature cyanide-based salt bath nitriding process. Salt bath nitriding will form not only the compound layer at the steel surface, but will allow nitrogen as well as carbon to diffuse into the surface to improve fatigue strength, torsional strength, and tensile strength. In addition, the process will further improve wear resistance, galling resistance, and corrosion resistance. Another advantage of both ferritic nitrocarburizing and nitriding is that they are low-temperature processes. This means that distortion will be kept to an absolute minimum. However, to say these processes are distortion free is misleading.

Low-Cyanide Salt Bath Ferritic Nitrocarburizing Salt bath nitrocarburizing was first established in the late 1940s when high-cyanide nitrocarburizing salt baths were introduced. Environmental considerations and the increased cost of detoxification of cyanide-containing effluents have led to the development of low-cyanide nontoxic salt bath nitrocarburizing treatments. This section summarizes the development of one commercially successful low-cyanide treatment known as the Melonite process developed by Houghton Durferrit (Ref 4).

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 203

Melonite Process The Melonite process, also known as Meli 1, uses a molten salt bath of a special composition. It provides a wear- and scuff-resistant surface on steels, sintered irons, cast irons, and similar materials. Treated parts exhibit excellent wear and corrosion resistance and good sliding characteristics. The process salt bath is mainly an alkali cyanate plus an alkali carbonate melted in a steel pot and fitted with an aeration system (as in the original Tufftride process). The cyanate component thermally reacts with the component surface to form alkali carbonate. Either continuously or intermittently, the bath is treated to regenerate the carbonate product back to a cyanate. The process forms a complete multilayer surface case that comprises a compound layer and a diffusion layer. The surface compound layer, consisting of different compounds of iron, nitrogen, and oxygen, resists abrasion corrosion and scuffing and is fairly stable at elevated operating temperatures. Surface hardness depends on the steel that is being treated, and researchers at Durferrit claim surface hardness values ranging from 800 to 1500 HV. The diffusion layer consists of nitrides of the appropriate alloying elements and formed carbides. Like the surface hardness, the case depth will vary according to the type of steel being treated. Simply stated: The lower the alloy content of the steel, the lower is the resulting hardness value. The case depth, however, is deeper. The higher the alloy content of the steel, the higher is the resulting hardness value. The case depth, however, is shallower (Fig. 2). Compound Layer. During salt bath nitrocarburizing by the Melonite process, a nitrocarburized layer is formed consisting of the outer compound layer (ε-iron nitride) and the underlying diffusion layer. The base material determines the formation, microstructure, and properties of the compound layer. The compound layer consists of compounds of iron, nitrogen, carbon, and oxygen. Due to its microstructure, the compound layer does not possess

Fig. 2

Influence of chromium on diffusion layer hardness and total case depth in various 0.40 to 0.45% C steels. Source: Ref 5

204 / Practical Nitriding and Ferritic Nitrocarburizing

metallic properties. Particularly resistant to wear, seizure, and corrosion, it also is stable almost to the temperature at which it was formed. Compared with plasma or gas nitrocarburizing, compound layers with the highest nitrogen content can be obtained by the Melonite process. Layers with high nitrogen content give better protection against wear, particularly corrosion, than those with a low content. Depending on the material used, the compound layer will have a hardness of about 800 to 1500 HV. Figure 3 compares surface layer hardnesses produced by various processes. In the metallographic analysis of salt bath nitrocarburized components, the compound layer is defined clearly from the underlying diffusion layer as a slightly etched zone. During the diffusion of atomic nitrogen, the compound layer forms. The growing level of nitrogen results in the solubility limit in the surface zone being exceeded, which causes the nitrides to precipitate and form a closed compound layer. In addition to the treatment parameters (temperature, duration, and bath composition), the levels of carbon and alloying elements in the materials to be treated influence the obtainable layer thickness. As stated previously, the higher the alloy content of the steel, the higher the resulting surface hardness value. The case depth, however, will be shallower. The profile shown in Fig. 4 is for Meli 1 bath at 580 °C (1075 °F) with the usual treating duration of 60 to 120 min. The compound layer obtained was 0.1 to 0.2 mm (0.0004 to 0.0008 in.) thick on most materials. Depending on the application, the process may consist of only the first quench (Q) cycle in Fig. 4, or it will include one or both of the post-treatments. Here we are considering only the first Q cycle. Diffusion Layer. The material largely determines the depth and hardness of the diffusion layer. The higher the alloying content of the steel, the lower the nitrogen penetration depth at equal treating duration. On the other hand, hardness increases with higher alloying content. 1400 480°C(895°F) 1.5 h Vickers hardess HV 0.1

1200

480°C(895°F) 3.0 h 480°C(895°F) 6.0 h

1000

580°C(1075°F) 6.0 h

800 600 400 200 4

Fig. 3

8

12

16 20 24 28 Distance from surface, in. × 10–4

32

36

40

Variation in hardness with distance from the surface of AISI HNV 3 (X45CrSi9.3) steel treated by Melonite process with varying conditions. Source: Ref 4

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 205

In the case of unalloyed steels, the crystalline structure of the diffusion layer is influenced by the cooling rate after nitrocarburizing. After rapid cooling in water, the diffused nitrogen remains in solution. If cooling is done slowly, or if a subsequent tempering is carried out, some of the nitrogen could precipitate into iron nitride needles in the outer region of the diffusion layer. This precipitation improves the ductility of nitrocarburized components. Unlike unalloyed steels, part of the diffusion layer of highly alloyed materials can be better identified metallographically from the core structure due to improved etchability. However, the actual nitrogen penetration is also considerably deeper than the darker etched areas visible metallographically. Cooling does not influence diffusion layer formation to any noteworthy extent. Figure 5 shows the compound layer thickness for various materials in relation to treatment time. Q

P

Q

Air 350–400 °C (660–750 °F)

350–400 °C (660–750 °F)

Postoxidizing

Polishing

Oxidizing + cooling

Nitrocarburizing

Preheating

350–400 °C (660–750 °F)

Time

Fig. 4

Time versus temperature profile for the QPQ nitrocarburizing treatment cycle. Source: Ref. 6

10

18 Compound layer in ten thousands

Temperature

580 °C (1075 °F)

Carbon steels e.g., 1015, 1045 Low-alloy steels e.g., 4140

Compound layer thickness Melonite MELI 1 at 580°C (1075°F)

High-alloy steels e.g., D3, H11, 304 Cast iron

6

4

2

0 0.5

1

2

3

4

Treating time in the MELI 1 bath, h

Fig. 5

Compound layer thickness in relation to treating time for various materials. Note that the time scale is logrithmic. Source: Ref 4

206 / Practical Nitriding and Ferritic Nitrocarburizing

Surface Hardness and Tensile Strength. The surface hardness obtainable by the Melonite treatment is influenced primarily by the composition of the material. The higher the content of nitride-forming alloying elements (Cr, Mo, Al, V, Mn, Ti, and W), the greater the surface hardness. Table 1 gives average tensile strengths and surface hardnesses for salt bath nitrocarburized steels. Wear Resistance and Running-In Properties. Due to the intermetallic composition of the compound layer, friction and the tendency to weld with a metallic counterpart are reduced. Melonite treated components exhibit excellent sliding and running-in properties, as well as greater wear resistance. Wear tests and practical applications repeatedly confirm the superior wear resistance of salt bath nitrocarburized parts over traditional casehardened, induction-hardened, or hard-chrome-plated surfaces. In many cases the wear resistance of the compound layer is further improved by an oxidative post-treatment. For example, components such as transmission shafts, plug gages, and hydraulic aggregates have a longer service life after Melonite treatment than after hard chrome plating. What about the wear resistance of the diffusion layer? Figure 6 compares the wear behavior of rocker arms treated by two different heat treatment processes. It shows the wear on the surface face of the rocker arm that ran against a salt bath nitrocarburized camshaft made from chilled cast iron. Although the surface hardness of the case-hardened rocker arm was slightly reduced by nitrocarburizing, the much improved wear resistance due to the presence of the compound layer to approximately 80 h running time is clearly visible. After 70 to 80 h, the wear profile runs parallel to that of the case-hardened rocker arm, which is attributable to the protection Table 1 Tensile strength and surface hardness for various nitrocarburized steels Tensile strength after hardening and tempering at 600 °C (1110 °F), MPa

Steel designation

Approximate surface hardness after 90 min at 580 °C (1075 °F) using the Melonite process

Germany

U.S.

2h

6h

HV 1

HV 10

HV 30

Ck15 C45W3 Ck60 20MnCr5 53MnSi4 90MnV8 42CrMo4 X19NiCrMo4 55NiCrMoV6 50NiCr13 X20Cr13 X35CrMo17 X210Cr12 X210CrW12 X165CrMoV12 45CrMoW58 X32CrMoV33 X38CrMoV51 X37CrMoW51 X30WCrV93

1015 1045 1060 5120 ... O2 4140 ... L6 ... 420 ... D3 ... ... ... H10 H11 H12 H21

600 750–850 750–900 800–950 850–950 1000–1200 900–1200 900–1100 1200–1400 1200–1350 1000–1200 1000–1200 1500–1700 1500–1800 1400–1900 1500–1800 1700–1800 1700–1900 1700–1900 1500–1800

550 700–800 700–800 800–900 800–900 900–1100 900–1100 900–1000 1150–1300 1100–1200 1000–1200 1000–1200 1400–1600 1400–1650 1400–1700 1400–1700 1600–1750 1500–1700 1600–1800 1500–1700

350 450 450 600 450 550 650 600 650 600 >900 >900 >800 >800 >800 800 >900 >900 >900 >900

300 350 350 450 400 450 500 500 550 500 600 700 600 600 650 700 850 850 800 850

200 250 250 400 350 400 450 450 500 450 450 550 450 500 500 600 700 700 700 800

Source: Ref 7

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 207

afforded by the diffusion layer. A spontaneous increase in wear after the loss of the compound layer was not observed (Ref 7). This test again showed that a high surface hardness does not automatically mean that the protection against wear is also very high. Assessment of a material or mating materials depends on the wear mechanism involved. Nitrocarburized running partners have proved themselves to be very good, particularly under adhesive wear conditions. Their tendency to seize is much lower than that of other surface layers. Figure 7 shows the results.

Salt Bath Nitrocarburizing plus Post Treatment (Ref 6)

Loss of weight of the running surface, g

As an adjunct to Melonite salt bath ferritic nitrocarburizing (as well as other nitrocarburizing methods such as the Kolene Nu-Tride process discussed later in this chapter), a mechanical polish and postbath oxidative 0.25

Case hardened

0.20

Case hardened and Melonite treated

0.15

Rocker arm: CrMo alloyed case hardening steel Camshaft: chilled cast iron, Melonite treated

0.10

Test conditions: 1000 rpm Load: 745-845 MPa (110-120 ksi) Oil: SAE 10W30 (80°C, or 175°F)

0.05

0

20

40 60 Running time, h

80

100

Fig. 6

Influence of surface treatments on rocker arm wear. Curves on left are for case-hardened components. Curves at right are for case-hardened and Melonite-treated components. Source: Ref 4

Fig. 7

Wear behavior of drawing dies after different surface treatments. SNC, salt bath nitrocarburizing. Source: Ref 5

208 / Practical Nitriding and Ferritic Nitrocarburizing

10

Weight loss, g /in.2

0.0112 (7.2)

0.011 (7.1)

5

0.008 0.004 (2.3)

Weight loss, g /mm2

0.016

0.0005 (0.3) 0

0 Chromium Nickel 20 µm 20 µm

SBN

QPQ

Fig. 8

Corrosion resistance of various surface treatments on steel based on field immersion tests. Test conditions: Full immersion for 24 h in 3% sodium chloride plus 3 g/L hydrogen peroxide. Salt bath nitrocarburized with no post-treatment. Source: Ref 6

treatment are carried out on the nitrocarburized surface. The quench-polishquench (QPQ) process is based on a sequence of process steps that occurs directly following the nitrocarburizing cycle. The process follows the same time-temperature profile shown in Fig. 4. The process begins with the nitrocarburizing segment—that is, preheat—salt bath nitrocarburize, and salt bath quench, which produces a compound layer of ε iron nitride. The next step is a mechanical polish of the nitride layer by vibratory polishing, lapping, centerless grinding, or similar means. Finally, to optimize corrosion resistance, the component is reimmersed in the salt quench bath for 20 to 30 min, rinsed, and oil dipped. The level of corrosion protection provided by the QPQ variant is shown in Fig. 8. The results demonstrate that the QPQ process provides maximum corrosion resistance compared with chromium plating, nickel plating, and conventional salt bath nitrocarburizing. Another comparative evaluation of corrosion resistance based on the ASTM B 117 salt spray test is shown in Fig. 9. These results also show the superior protection provided by the QPQ treatment, even after 336 h exposure to the salt spray testing environment. The QPQ treatment also improves antifrictional characteristics and fatigue properties of steel parts.

Kolene Nu-Tride Process (Ref 8) The Kolene Corporation offers a proprietary salt known as Nu-Tride for surface treatment. The procedure, which will be referred to generically and subsequently as salt bath nitrocarburizing (SBN), is controlled by the use of cyanates and carbonate salts in a molten condition as the source for nitrogen and carbon. A typical process reaction is: 8CNO → 2CO3 + 4CN + CO2 + [C]Fe + 4 [N]Fe

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 209

100

Surface area corroded, %

88 h 75

88 h 72 h

72 h

50

25

88 h

336 h

0 Hard chrome

Fig. 9

Electroless nickel

QPQ

Corrosion resistance of surface-treated steel spool shafts used in automotive steering columns based on ASTM B 117 salt spray test. Source:

Ref 6

As the ferrous component is immersed into the molten salt, a catalytic surface reaction takes place, causing the cyanate to decompose and release the nitrogen and carbon for the process. The decomposition of the salt to supply the nitrogen component of the surface metallurgy satisfies the solubility limit of nitrogen in iron of 6 to 9%. The surface carbon content will be in the region of 1% in a solid solution in the ferrous component surface. In addition (as seen in the traditional nitriding process), a surface compound zone begins to nucleate in which the ε-nitride is the dominant phase in partnership with the γ′-phase.

Dimensional Stability The SBN process is conducted at subcritical temperatures of 580 °C (1075 °F); therefore, thermally induced microstructural phase changes and accompanying volume changes do not occur. This permits the treated component to be cooled from the SBN bath without drastic quenching, thus minimizing dimensional changes and distortion. Accordingly, components may be finish machined and/or ground prior to SBN. Any dimensional growth resulting from the treatment is predictable and reproducible, given that the part has been thermally stabilized prior to final machining (i.e., sufficiently tempered or stress relieved), typically at a minimum of 595 °C (1100 °F). Stampings and other parts with thin cross sections are well suited for SBN because of the capability for dimensional stability and no distortion.

The SBN Processing Sequence Referring to Fig. 10, the process begins with a prewash and preheat cycle at 400 °C (750 °F) to ensure that the parts are clean and dry. A load

210 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 10

Kolene QPQ process cycle. Source: Ref 9

of components that has been uniformly preheated will reduce thermal shock and permit more efficient recovery of the SBN bath temperature. The load is then transferred to the salt bath and held at 580 °C (1075 °F) for a predetermined dwell time, depending on the required depth of compound layer. From the salt bath, the parts are quenched into an oxidizing salt bath (KQ-500) at a lower temperature, typically 400 °C (750 °F), and held from 5 to 20 min. This intermediate quench cycle is a key ecological feature of the SBN process since during this exposure to the oxidants, cyanates and any cyanides generated within the nitriding bath and contained as part of the dragged-out salt are effectively destroyed. Intermediate quenching also retards the cooling rate, thus minimizing thermally induced distortion. After the oxidizing quench, parts are cooled to room temperature, rinsed, and, if required, subjected to post treatment. This may include mechanical polishing, if surface finish is of concern, or the Kolene QPQ treatment to develop maximum corrosion protection and/or enhance the cosmetic appearance. The QPQ treatment provides a lustrous dark finish.

Process Control An important factor in producing the desired monophase ε iron nitride when nitrocarburizing is control of the nitrogen and carbon activities of the processing environment. For SBN, this is accomplished by monitoring and regulating the cyanate (CNO–) concentration within the operating salt bath to 34 to 38%. As more parts, and thus more iron, are put through the bath, cyanate content decreases until regeneration becomes necessary. Based on results of standard titration analyses, an appropriate amount of an organic polymer regenerant is added. This reacts with the carbonate in the bath and converts it to cyanate.

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 211

Metallurgical Results The SBN process can be performed on a number of ferrous-based materials, including carbon steels, low-alloy steels, tool steels, and stainless steels. Special nitriding grades respond to SBN, but the presence of nitride-forming alloying elements is not required to produce a compound zone. Formation of nitride in low-carbon steel is seen in Fig. 11. Cast gray iron and ductile (nodular) iron are also treatable by this diffusion process. The response to SBN as measured by depths of compound zones and diffusion zones formed and by the associated hardnesses of both zones is directly related to the material composition. In general, with an increase in carbon content and/or in the amount of alloying elements, the rate of diffusion decreases, thus resulting in shallower depths of nitrogen penetration (Fig. 12a and b). Diffusion rates are particularly sensitive to materials containing nitride-forming elements such as chromium and molybdenum. Carbon and Low-Alloy Steels. The SBN process time cycles are most often limited to 2 h since diffusion rates decrease with time and long cycles become increasingly nonproductive. Typically, a compound zone depth of 7.5 to 20 µm (0.0003 in. to 0.0008 in.) would be produced in 90 min for carbon or low-alloy steels with diffusion zone depths ranging from 0.4 to 0.75 mm (0.015 to 0.030 in.), depending on the material (Fig. 13a and b). Tool Steels. Compound zone depth requirements and associated time cycles for tool steels depend on the application. High-speed cutting tools require a very thin compound layer and thus are treated for only 10 to 20 min at a reduced temperature (540 °C, or 1000 °F) to maintain base-material hardness. This provides resistance to chip welding at the cutting edge. The SBN treatment also benefits die casting and forging dies by reducing heat checking, soldering, diffusion, and wear from molten metals (Fig. 14). Cycle times for hot-work die steels may require 2 h to achieve the desired properties. The performance of stamping dies is also increased by reducing galling and seizing (Fig. 15).

Fig. 11

Microstructure of low-carbon steel treated by SBN plus Kolene KQ-500 process. 500×. Source: Ref 9

Fig. 12

Thickness as a function of nitrocarburizing time for various alloys. (a) Rate of diffusion decreases with increasing carbon and alloying content, resulting in shallower penetration. (b) Total nitriding depth of specific alloys. Source: Ref 5

Fig. 13

Low-carbon steel SBN treated and then aged to precipitate nitride needles. Source: Ref 10

Fig. 14

Gross heat checking in a low-alloy tool steel forging die due to excessive temperature. Heat checking occurred after an undetermined number of 225 kg (500 lb) nickel-base alloy preforms had been forged from an average temperature of 1095 °C (2000 °F). Source: Ref 11

8

Taber abraser test AISI 1117 Steel

Wear, in. × 10–4

7

Untreated

6 5

Case hardened to HRC 54

4 3

SBN

2 1 0 0

10

20

30

40

50

60

70

Cycles × 103 8

Taber abraser test AISI 1045 Steel

Wear, in. × 10–4

7

Hardened and tempered to HRC 28

6 5

SBN

4 3 2 1 0 0

10

20

30

40

Cycles ×

Fig. 15

50

60

70

103

Wear resistance comparison between SBN treated specimens and alternate treatments for two steels. Source: Ref 10

214 / Practical Nitriding and Ferritic Nitrocarburizing

Stainless steels are SBN treated primarily to introduce wear resistance and antigalling properties. While developing these properties, there could be significant degradation of the stainless (corrosion resistance) qualities of the material, which should be considered when specifying any nitrocarburizing process. Stainless steels respond quickly to diffused nitrogen, developing an extremely hard compound zone of complex nitrides (Fig. 16). Zone depths of 25 µm (0.001 in.) may be developed within 1 h with a diffusion layer depth of only 75 µm (0.003 in.) (Fig. 17). Cast irons also respond to SBN but at a somewhat reduced rate. Cycle times of 90 to 120 min are normally specified to generate compound zone depths of 75 to 200 µm (0.003 to 0.008 in.) and diffusion zones to 0.4 mm (0.015 in.). Both gray iron and nodular iron (Fig. 18a and b) are generally treated to improve wear resistance and increase fatigue strength.

Fig. 16

Compound layer depth, in. × 10–4

Compound layer in type 316 stainless steel consisting entirely of Sphase. SBN, 455 °C (850 °F) for 5 h. Marble’s reagent, 1000×. Source: Ref 12

Diffusion Curves 304 & 316 SS, Annealed SBN in Nu-Tride *Modified Nu-Tride

35 30 25

630°C (1165°F) +

580°C (1075°F)

20 510°C (950°F)*

15 10

455°C (850°F)*

5

400°C (750°F)*

0 0

2

4

6

8

10

SBN time, h

Fig. 17

Diffusion response of annealed types 304 and 316 austenitic stainless steel to SBN in Nu-Tride at 400 to 625 °C (750 to 1160 °F). Source: Ref 13

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 215

Fig. 18

Cast irons treated in SBN. (a) Gray iron. SBN, 580 °C (1075 °F) for 2 h, followed by salt bath quenching. Nital. Original magnification 500×. (b) Ferritic nodular iron. SBN, 580 °C (1075 °F) for 90 min, followed by salt bath quenching. Nital. Original magnification 500×. Source: Ref 10

A comparison of the microstructures of various ferritic metals after SBN is given in Fig. 19.

Engineering Properties The end result of SBN is the presence of two distinct layers of a nitrogenbearing microstructure, the outermost identified as the compound zone and the subjacent layer called the diffusion zone. Each of the zones contributes to improving performance by enhancing specific engineering properties such as wear resistance, lubricity, corrosion resistance, and fatigue strength. From these, other benefits in performance are realized, including excellent running-in properties, antigalling, and antiseizing characteristics, and reduced tendency for fretting corrosion. Wear resistance is perhaps the most significant property resulting from SBN. The ability of the compound zone to resist wear depends on whether the wear is adhesive or abrasive. Adhesive wear occurs when two components are in relative motion in an essentially abrasive-free environment. Under these conditions, the intrinsic physical characteristics of the compound zone (i.e., hardness and lubricity) notably improve the sliding and running-in behavior, and consequently increase the resistance to adhesive wear (Fig. 20). The phase composition of the compound zone that demonstrates the best wear resistance consists predominantly of ε-nitride phase (monophase preferred) with a very small amount of γ′-phase. Resistance to abrasive wear depends on the relative hardnesses of the abrading substance and of the compound zone. For unalloyed carbon steels, the compound zone hardness is equivalent to approximately HRC 55, thus providing only short-time resistance to abradants of higher hardness. One method of increasing hardness, and thus abrasive wear resistance, is to increase the content of nitride-forming elements of the base material.

216 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 19

Microstructures of various ferritic materials that have undergone salt bath nitrocarburizing. All etched in 3% nital. All 500×. Courtesy of Kolene Corp. (a) Ferritic nodular iron; 90 min at 580 °C (1075 °F), oxidizing molten salt quench. (b) Low-carbon steel; 90 min at 580 °C (1075 °F), oxidizing molten salt quench. (c) Type 304 stainless steel; 90 min at 580 °C (1075 °F), oxidizing molten salt quench. (d) AISI D2 tool steel; 90 min at 580 °C (1075 °F), oxidizing molten salt quench. (e) H13 medium-carbon hot-work tool steel; 120 min at 580 °C (1075 °F), oxidizing molten salt quench

Progressive increases in chromium from 0% (carbon steel) to 17% (stainless steel) result in a corresponding increase in hardness up to HRC 70 (minimum). Lubricity is the other engineering property that influences resistance to adhesive wear. The compound zone has an inherent low coefficient of friction (see Fig. 20) and functions as a solid-film lubricant by providing a nonmetallic interface between mating surfaces. Applications that are thus benefited include forming tools, extrusion dies, and sliding and rotating systems that rely on good running-in properties.

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 217

Mating disks same treatment 0.4 Coefficient of friction, µ

No lubrication Lubricated with SAE 30 oil

0.3

0.2

0.1

0.0 Chrome plated, 30 µm (0.0012 in.)

Fig. 20

Case hardened

SBN 90 min 580°C (1075°F) plus SBQ

Frictional properties for various surface treatments. SBQ, salt bath quenching. Source: Ref 10

Corrosion resistance increases via SBN, most notably with the Kolene QPQ process. Figures 8 and 9 show how QPQ treatments improve corrosion properties. Fatigue strength under bending and torsional loading develops as a result of nitrogen present in the diffusion zone and particularly the amount in solid solution. Improvement in fatigue strength produced through SBN occurs in a range of materials and in varying degrees (Table 2). Lowcarbon steels with relatively low strength display the greatest increase in fatigue strength (from 80 to 100%), whereas the higher alloy steels and cast irons exhibit from 20 to 80% improvement in strength.

Other Methods for Salt Bath Nitrocarburizing Sursulf, a ferritic nitrocarburizing procedure developed in France, also is based on the principle of the decomposition of a low-percentage cyanide to cyanate that is mixed with a low-temperature carbonate salt for the activated carbon in the salt. The salt also has a sulfur compound in the salt analysis to assist with the formation of surface sulfides as well as some surface porosity. The porosity will form small reservoirs to hold surface lubricants. The lubricants will also work with the sulfur to form an almost self-lubricating surface with high wear resistance and improved corrosion resistance. Tenoplus is a two-stage high-temperature nitrocarburizing process. The first stage is conducted at 625 °C (1160 °F) and held at that temperature, followed by a cooldown to a lower temperature of 580 °C (1075 °F). Subsequent cooldown takes place in an oxidizing bath to a lower temperature of 350 to 400 °C (660 to 750 °F). After polishing, the workpiece undergoes a controlled oxidizing process in a special salt bath (Ref 6).

218 / Practical Nitriding and Ferritic Nitrocarburizing

Table 2 Improvement in fatigue strength Alloy treated

Low-carbon steels Medium-carbon steels Stainless steels Chrome/manganese (low-carbon) steels Chrome alloy (medium-carbon) steels Cast irons

Fatigue strength improvement, %

80–100 60–80 25–35 25–35 20–30 20–80

Source: Ref 10

REFERENCES 1. T. Bell, Ferritic Nitrocarburizing, Met. Eng. Q., May 1976, reprinted in Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 266–278 2. Cassel Manual of Heat Treatment and Case Hardening, 7th ed., Imperial Industries Ltd., United Kingdom, 1964 3. B. Prenosil, Hutn. Listy, 1962, Vol 17 (No. 6), p 414–424 (available as translation BLL RTS 9520) 4. R. Willing-Lepenies and C. Faulkner, “New Developments in Salt Bath Nitrocarburizing,” product literature, Houghton Durferrit, Valley Forge, PA 5. G. Wahl and S. Alwart, Improvement of Tribological Properties through Nitrocarburizing 6. Nitrocarburizing, Surface Hardening of Steels: Understanding the Basics, J.R. Davis, Ed., ASM International, 2002, p 195–212 7. R. Willing-Lepenies and C. Faulkner, “Melonite-QPQ-Process,” product literature, Houghton-Durferrit, Valley Forge, PA, p 9 8. “The Theory and Practice of Molten Salt Bath Nitriding,” product literature, Kolene Corporation, Detroit 9. J.R. Easterday, “The Kolene QPQ Process,” product literature, Kolene Corporation, Detroit 10. J.R. Easterday, “Salt Bath Ferritic Nitrocarburizing,” product literature, Kolene Corporation, Detroit 11. Tool Materials, ASM Specialty Handbook, ASM International, 1995, p 228 12. J.R. Easterday, Expanding the Temperature Range for Salt Bath Nitrocarburizing, Ind. Heat., Vol 70 (No. 1), Jan 2003, p 34–38 13. J.R. Easterday, “Influence of SBN on Corrosion Resistance of Stainless Steels,” product literature, Kolene Corporation, Detroit

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p219-230 DOI: 10.1361/pnafn2003p219

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

20 Gaseous Ferritic Nitrocarburizing

GASEOUS FERRITIC NITROCARBURIZING, like salt bath nitrocarburizing, involves the introduction of carbon and nitrogen into a steel in order to produce a thin layer of iron carbonitride and nitrides, the “white layer” or compound layer, with an underlying diffusion zone containing dissolved nitrogen and iron (or alloy) nitrides. The white layer enhances surface resistance to galling and wear. The diffusion zone significantly increases the fatigue endurance limit, especially in carbon and low-alloy steel. The compound-diffusion layer may contain varying amounts of gamma prime (γ′) and epsilon (ε) phase, cementite, and various alloy carbides and nitrides. The exact composition is a function of the nitride-forming elements in the steel and the composition of the atmosphere. Following thorough cleaning (vapor degreasing is adequate for most applications), parts are nitrocarburized near 570 °C (1060 °F), a temperature just below the austenite range for the iron-nitrogen system. Treatment times generally range from 1 to 4 h. Although there are a number of proprietary gas mixtures, most contain ammonia (NH3) and an endothermic gas. Batch furnaces with integral oil quenches are ideally suited for gaseous nitrocarburizing.

Development of the Process The gaseous nitrocarburizing process was first developed in 1961 at Joseph Lucas (Industries) Ltd. in England. This treatment produced on mild steel a porous layer that was claimed to have good antifrictional properties. The published patent revealed that the gaseous atmosphere consisted of ammonia and hydrocarbon or other carbon-containing gases of unspecified proportions and that the treatment was undertaken in the temperature range of 450 to 590 °C (840 to 1095 °F). During the 1960s, further research led to the development of a large range of gas nitrocarburizing processes

220 / Practical Nitriding and Ferritic Nitrocarburizing

throughout the world. Some of these processes, most of which are proprietary, go by the names of: • • • • • • • •

Nitemper NitroTec Deganit Nitroc Lindure Controlled nitrocarburizing Soft nitriding Vacuum nitrocarburizing (using a conventional cold-wall furnace)

References 1 and 2 provide detailed information on these various processes.

Process Principles The atmosphere method relies (as does the salt bath procedure) on a constant supply of nitrogen (from ammonia), carbon from a suitable gaseous source, and oxygen. Ammonia is considered to be the most readily and actively available source of nitrogen for the process and is blended with the other process gas supplies of carbon (from a hydrocarbon gas) and oxygen from other sources. As in gas nitriding, the cracked nascent ammonia gas will dissociate at the steel surface and react with the hydrocarbon gas to form both nascent nitrogen and free carbon. The gases will allow carbon dioxide to be generated in relation to the classical water gas reaction: CO2 + H2 ↔ H2O + CO

If we assume that ammonia is supplied at a constant pressure to the process, a drop in the partial pressure of hydrogen will occur. This in turn will increase the nitriding potential of the process (Ref 3) and lead to: NH3 + CO ↔ HCN + H2O

This means that some hydrogen cyanide (HCN) gas will form as a byproduct of the process. The hydrogen cyanide will contribute both nitrogen and carbon to the process reaction, thus improving the nitriding potential of the total process gas. The metallurgical results of the process are very similar to the classical nitriding process, with the exception that now there is carbon in the layer. The thickness of the compound layer is determined by: • • • •

Time Temperature Process gases Pretreatment of the steel

Chapter 20: Gaseous Ferritic Nitrocarburizing / 221

The rate of growth of the compound layer is approximately the square root of the process time at the chosen process temperature. The primary objective of the ferritic nitrocarburizing process is to form nitrides and, most importantly, surface carbon, which will encourage the ε-nitride phase in the surface of the steel. The surface carbon comes from three sources: • • •

Decomposition of the process gas Steel chemistry Hydrocarbon additive gas

Gaseous Supply As stated earlier, gas generally is supplied by using an endothermically generated gas with an addition of ammonia. The percentages are approximately 50% for each gas, depending, of course, on the surface metallurgy required. Some proprietary procedures use an exothermically generated gas as the primary process gas, followed by the addition of ammonia. The exothermic gas decomposes to provide a small percentage of carbon monoxide, along with hydrogen and nitrogen from the ammonia, which will decompose in a catalytic reaction at the steel surface. In some proprietary procedures, methane is added to promote ε-nitride formation in the steel surface. The procedure can then be followed with a deliberate and controlled surface oxidation by adding a source of oxygen to the process. The oxygen level must be kept to below combustible levels (less than 2% oxygen). The result will be a surface layer of free oxides with a low degree of surface porosity.

Properties of Gaseous Ferritic Nitrocarburized Components The properties obtained as a result of the gaseous ferritic nitrocarburizing process are as varied as those obtained with other process techniques. Surface properties, however, can be modified simply by modifying the process parameters. The compound zone properties formed by the ferritic nitrocarburizing process are generally very similar. There will be a significant difference if the immediate surface has been oxidized after completion of the ferritic nitrocarburizing treatment. The nonoxidized surface will show a significant increase in: • • • • •

Wear resistance Torsional strength Corrosion resistance (except on stainless steels) Surface hardness Fatigue strength

222 / Practical Nitriding and Ferritic Nitrocarburizing

These properties are determined by traditional methods of property measurement.

Industrial Applications Applications for the ferritic nitrocarburizing process are widely varied. The typical type of steel used for the process is a low-carbon, low-alloy steel. However, the process can also be applied to medium- and high-carbon steels with various metallurgical results. Typical components that undergo gaseous nitrocarburizing include: • • • • • • • • • • • • • • • • • • •

Machine spindles Cams Timing gears Powder metal components (see discussion below) Die casting dies (see discussion below) Shot sleeves Exhaust valves Cylinder liners Camshafts Crankshafts Steel water valves Ductile iron pump housings Automotive components, including suspension struts, stickshift levers, and window winding mechanisms Gas spring pistons Door lock mechanisms Gears (see discussion below) Machine slides Cylinder liners Water pump components

Many other components can be successfully ferritic nitrocarburized. Reduced material and processing costs can result in significant savings. When an engineer is considering the use of gaseous ferritic nitrocarburizing, he or she should consult with the heat treater or metallurgist to discuss process advantages and limitations as well as the metallurgical behavior of the part after the process treatment. Some important considerations for three important application areas are discussed below. Die Casting. The die casting industry needs surface treatments that will improve wear resistance, improve release properties, and resist the buildup or welding of aluminum, magnesium, and zinc on mold and core surfaces. This has led tool and die makers and mold makers to review the properties and performance of ferritic nitrocarburized treatments in terms of applications and performance on tool steels.

Chapter 20: Gaseous Ferritic Nitrocarburizing / 223

Gear Heat Treatment. Most gear manufacturing depends on control tolerance and precision treating in terms of distortion. This means that a process that can offer a significant reduction in distortion and an improvement in gear performance will be closely scrutinized. Ferritic nitrocarburizing is performed at a low temperature (in the ferrite region) and introduces both carbon and nitrogen into the surface of the steel, followed by a quench to improve fatigue resistance with a minimal amount of distortion. Gear manufacturers are finding that the process can be applied to gear cutting hobs, cutting tools, and broaching tools. The process also improves antiscuffing properties and compression stress at the gear surface. Sintered Steels. Extensive research work has been done on sintered powder metallurgy (P/M) ferrous alloys, and ferritic nitrocarburizing is now a fully commercial process for the P/M industry. The degree of success depends on the level of compaction of the sintered part. Both gaseous and plasma nitrocarburizing processes have found commercial use.

Safety Considerations Because the gaseous ferritic nitrocarburizing process involves a combustible atmosphere that is explosive when operated below the self-ignition temperature, safety precautions must be rigorously enforced. It is mandatory that all safety purge systems are in place and fully operable and that the correct purge sequencing has occurred before combustible gases are admitted to the furnace. The process operating temperature is in the region of 570 °C (1060 °F). The normal combustion gas ignition temperature is 750 °C (1350 °F), so any temperature below this will cause an explosion if the process gas is not admitted according to the furnace manufacturer’s written procedures.

Appendix: Gaseous Nitrocarburizing— A Suitable Alternative for the Heat Treatment of Automotive Crankshafts K. Bennett, BOC Ltd. and Q. Weir and J. Williamson, Leyland Vehicles Ltd. The remarkable properties conferred on low-alloy steels by nitrocarburizing have been highlighted comprehensively in the literature in the past few years (Ref 4–6). In addition to enhanced fatigue resistance, the increase in wear resistance, attributed to the formation of a thin compound layer (Fig. 1) composed of the single-phase ε-nitride or carbonitride, has been shown to be dramatic. Typically, a compound layer of some 12.5 to 20 µm has been reported as a result of a conventional 3 h nitrocarburizing treatment at a temperature of 570 °C (1060 °F) (Ref 4).

224 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Typical compound layer on nitrocarburized mild steel

It has been suggested recently (Ref 7) that the properties obtained by the 60/80 h gaseous nitriding of classical nitriding steels can be matched by a 12/20 h nitrocarburizing treatment (Fig. 2). In this context, an added advantage of nitrocarburizing would be the replacement of the brittle duplex white layer, characteristic of gaseous nitriding, by the more ductile single-phase ε compound layer. This would eliminate the frequent necessity for postheat treatment machining. Six-cylinder engine crankshafts, manufactured in B.S. 970: 1970 708A42 steel (En 19C), are normally nitrided for a process period of 60 h. The service conditions in respect of the crankshaft journals are such that while good wear resistance is required, a high indentation resistance is not essential. A prime requirement exists for a good fatigue resistance. Following consultation with BOC Ltd., Leyland Vehicles Ltd. decided to initiate a program of work aimed at evaluating the properties to be obtained by short-cycle nitrocarburizing (3 h duration in this instance) of such automotive crankshafts. If it were possible to produce crank-shafts with acceptable resistance to both wear and fatigue by such a treatment, then the cost savings would be considerable relative to gaseous nitriding. It was decided that the evaluation should be based upon surface hardness tests, microhardness traverses, metallographic examination, and fatigue tests. Such tests would be reinforced by both engine testing and road trials on the actual components.

Process Technique The geometry of the components is such that it was considered essential to treat vertically in order to control distortion (Fig. 3). The only suitable equipment available at Leyland for this purpose was a Wild Barfield pit fur-

Chapter 20: Gaseous Ferritic Nitrocarburizing / 225

Fig. 2

Comparative hardness profiles produced by 80 h nitriding and 16 h nitrocarburizing

Fig. 3

Furnace load of crankshafts

nace which was normally used for gas carburizing. This is shown schematically in Fig. 4. Some reservations were expressed as the normal technique of venting ammonia-bearing exhaust gases to atmosphere could not be accommodated. The location of the furnace and the height of the heat treatment shop presented additional problems in this respect. However, it was

226 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 4

Schematic diagram of the Wild Barfield pit carburizing furnace used in the nitrocarburizing trials

found in practice that the exhaust burnt quite readily during processing at the nitrocarburizing temperature. The furnace was modified by the inclusion of a pilot flame at the exhaust pipe. It was decided that the furnace atmosphere would be the standard nitrogen/ammonia/carbon dioxide nitrocarburizing system (Ref 5–7). A further furnace modification was carried out to facilitate the separate delivery of carbon dioxide to the furnace chamber. (It is an accepted fact that, should ammonia and carbon dioxide be delivered through the same manifold, then a reaction will occur between the two gases. The manifold will quickly become blocked with heavy deposits of ammonium carbamate.) With a furnace volume of 2 m3 (70 ft3), the standard purge atmosphere flow employed equated to six volume changes per hour. During processing, the standard atmosphere flow was four volume changes per hour. The full treatment details were as follows: • • •

Clean and degrease workpieces and jigs thoroughly. Load furnace standing at ambient temperature. Seal and purge with nitrogen at purge flowrate for 1 h. Switch on heat after 1 h and reduce nitrogen flowrate to four volume changes per hour.

Chapter 20: Gaseous Ferritic Nitrocarburizing / 227

• •

• •

• •

Raise furnace temperature to 570 °C (1060 °F) under nitrogen atmosphere only. When furnace attains a temperature of 570 °C (1060 °F), adjust atmosphere makeup to effect nitrocarburizing, while maintaining a process flowrate of four volume changes per hour. A nominal ingoing atmosphere composition would be 70% ammonia, 25% nitrogen, and 5% carbon dioxide (Ref 5). Nitrocarburize for 3 h at 570 °C (1060 °F). Turn off ammonia and carbon dioxide at the end of the process period. Increase nitrogen flow to compensate and proceed to reduce furnace temperature. Reduce furnace temperature to below 400 °C (750 °F). Unload furnace and cool components either by oil quenching or slow cooling.

It was found necessary in practice to reduce the furnace temperature below 400 °C (750 °F) in order to control dimensional movement and to avoid possible cracking in thin-wall areas. The cooling rate below 400 °C (750 °F) would not appear to affect the metallurgical properties of the treated components. It was evident, however, that the slower-cooled components exhibited a lower distortion factor.

Typical Results On hardened and tempered steel containing 0.40% C, 0.83% Mn, 0.24% Si, 0.17% Ni, 1.06% Cr, and 0.17% Mo, typical results after 3 h nitrocarburizing were as discussed below. Hardness. Surface hardness of the order of 580 to 650 HV5 were achieved. Hardness traverses on sectioned samples indicated microhardnesses in the compound layer in excess of 700 HV0.01, with the highest values being recorded close to the interface of the compound layer and the substrate. Core hardness was 320 to 330 HV5. Case Depth. Hardness profiles (Fig. 5) showed effective case depths (to 500 HV) of 0.125 to 0.15 mm (0.005 to 0.006 in.). Metallographic examination revealed a typical continuous unbroken compound layer (Fig. 6). The depth of the compound layer was of the order of 25 to 37.5 µm, which is considerably deeper than that expected of a conventional nitrocarburizing treatment of the same duration on the same material. Fatigue Resistance. Fatigue testing of treated crankshafts was carried out at the Motor Industry Research Association (MIRA). The results obtained (Fig. 7) point to a fatigue resistance twice that of untreated components. Wear Resistance. Crankshafts nitrocarburized in the 3 h treatment were run in a six-cylinder turbocharged engine on an experimental test bed in a 1884 h endurance trial. No wear was observed on the crank pins; a maximum of 6 µm (0.00025 in.) wear occurred on the main journals.

Fig. 5

Typical hardness profile on a crankshaft after 3 h nitrocarburizing

Fig. 6

Compound layer on a crankshaft after 3 h nitrocarburizing

Fig. 7

Fatigue tests results

Chapter 20: Gaseous Ferritic Nitrocarburizing / 229

Conclusions The results of the trials demonstrated that: •

• •

The wear resistance conferred on a 708A42 crankshaft by a 3 h nitrocarburizing treatment in a nitrogen-based atmosphere is substantial and comparable with that produced by conventional 60 h gas nitriding of the same material. In view of the fact that the effective case depth of the nitrocarburized component is approximately 50% of that of a nitrided component, the wear performance can only be attributed to the properties of the compound layer. The fatigue resistance obtained by 3 h nitrocarburizing is within specification for the automotive crankshaft. The adoption of a short-cycle nitrocarburizing treatment in this instance would yield a saving in heat treatment costs of some 85%.

In addition it has been shown that the nitrocarburized surface facilitates the use of a cheaper bearing than that required for the more friable surface of the nitrided component. ACKNOWLEDGMENT This appendix was reprinted with minor changes from K. Bennett, Q. Weir, and J. Williamson, Gaseous Nitrocarburising—A Suitable Alternative for the Heat Treatment of Automotive Crankshafts, Heat Treatment of Metals, Vol 8 (No. 4), 1981, p 79–81. Reproduced by permission of Wolfson Heat Treatment Centre. REFERENCES 1. T. Bell, Gaseous and Plasma Nitrocarburizing, Heat Treating, Vol 4, ASM Handbook, ASM International, 1991, p 425–436 2. T. Bell, Ferritic Nitrocarburizing, Met. Eng. Q., May 1976, reprinted in Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for Metals, 1977, p 266–278 3. J. Grosch, Heat Treatment with Gaseous Atmospheres, Steel Heat Treatment Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker, 1997, p 663–719 4. T. Bell, Ferritic Nitrocarburising, Heat Treatment of Metals, Vol 2, 1975, p 39–49 5. C. Dawes, D.F. Tranter, and C.G. Smith, Reappraisal of Nitrocarburising and Nitriding When Applied to Design and Manufacture of NonAlloy Steel Automobile Components, Heat Treatment ’79, Book No. 261, The Metals Society, 1980, p 60–68; also in Metals Technology, Vol 6 (Part 9), Sept. 1979, p 345–353 and Journal of Heat Treating, Vol 1 (No. 2), Dec. 1979, p 30–42

230 / Practical Nitriding and Ferritic Nitrocarburizing

6. W.I. James, Practical Experience with Nitrogen-Based Nitrocarburising, Heat Treatment of Metals, Vol 6, 1979, p 13–15 7. K. Bennett, Advances in Nitrogen-Based Nitrocarburising, Proceedings of the 18th International Conference on Heat Treatment of Materials (Detroit, 6–8 May 1980), p 146–160

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p231-240 DOI: 10.1361/pnafn2003p231

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

21

Equipment for Ferritic Nitrocarburizing EQUIPMENT for the ferritic nitrocarburizing process is quite diverse, covering salt bath furnaces, atmosphere furnaces, and plasma furnaces. Each of these is examined in this chapter.

Salt Bath Furnace Equipment Early on, ferritic nitrocarburizing was accomplished in salt baths. Advantages of the salt bath furnace include: • • • • •

Simple design Fueled by either natural gas or electricity Easy to operate Excellent heat transfer and temperature uniformity Not capital intensive

Major disadvantages of the salt bath procedure have been cleanliness and effluent disposal. Good housekeeping of the salt bath work area is essential not only in terms of cleanliness but also personal safety. Effluent disposal is a subject addressed by all of the process salt manufacturers, who now produce process salts with very low cyanide contents, to less than 4 wt%. Although the salt still contains cyanide, decomposition and subsequent neutralization of the cyanide effluent are much easier. Salt bath equipment used for nitrocarburizing is essentially similar in design to salt bath furnaces used for other processes. Although batch installations are most common, semicontinuous and continuous operations are possible. Two typical salt bath lines are shown in Fig. 1 and 2. The equipment relies on a salt bath pot constructed of a material that is nonreactive with the process salt. The pot can be made entirely of titanium or can be titanium lined to reduce the capital cost of the process bath equipment.

232 / Practical Nitriding and Ferritic Nitrocarburizing

Fig. 1

Internally heated salt bath furnace with immersed alloy electrodes and ceramic tile lining

Fig. 2

Internally heated salt bath furnace with submerged graphite electrodes and a modified brick lining for use with carburizing salts

Chapter 21: Equipment for Ferritic Nitrocarburizing / 233

Atmosphere Furnace Equipment Atmosphere nitrocarburizing requires a furnace that is both gastight and airtight to ensure safety of the operator and the equipment. Furnace design is of paramount importance to prevent an ingress of unwanted oxygen. The furnace can be either the cold-wall vacuum design for partial pressure processing or the integral quench atmosphere design. It should be noted that because ferritic nitrocarburizing takes place below the atmosphere design limits of an integral quench furnace, the furnace manufacturer should be consulted. Integral quench furnace doors are typically safety interlocked. On a normal integral quench furnace, the external door cannot be opened at a chamber temperature below 760 °C (1400 °F). This feature is to eliminate the risk of an explosion. Consult the furnace manufacturer regarding low-temperature operations such as ferritic nitrocarburizing.

Plasma-Assisted Furnace Equipment Plasma-assisted ferritic nitrocarburizing systems (Fig. 3) often differ from plasma nitriding systems in terms of process vessel materials of construction. This is because the equipment operates at process temperatures up to 110 °C (200 °F) higher than those used in plasma-assisted nitriding. Process temperatures can reach a maximum of 650 °C (1200 °F). The principle of operation remains the same, however, with the process conducted in a partial-pressure condition. A major difference is that the plasma-assisted ferritic nitrocarburizing furnace usually will have an

Fig. 3

Typical plasma nitrocarburizing furnace and associated control system. Courtesy of Plateg GmbH

234 / Practical Nitriding and Ferritic Nitrocarburizing

additional source of oxygen for the subsequent oxynitride procedure, which provides surface enhancement by deliberate surface oxidation. Improvement in corrosion resistance is shown in Fig. 4 and 5.

Process Control of Surface Metallurgy Because most steels can be treated by plasma-assisted ferritic nitrocarburizing, it is necessary to thoroughly understand the steel being processed in relation to process capabilities. In the case of low-alloy steels, there are insufficient alloying elements to accomplish any real surface strength. Because all steels will readily absorb nitrogen at temperatures above 205 °C (400 °F) but do not always form stable nitrides (other than iron nitrides), iron nitride formation can be encouraged by supplying

Fig. 4

Electric fan motor treated by the Nitrotec process (right) and zincplated (left). Both were subjected to a 250 h neutral salt spray test.

Fig. 5

Towing ball hitch treated by the oxynitride process (right) and untreated (left). Both subjected to salt spray test. Courtesy of Plateg

GmbH

Chapter 21: Equipment for Ferritic Nitrocarburizing / 235

additional nitrogen. This is accomplished by increasing the nitrogen-tohydrogen ratio at the flowmeters (or mass flowmeters). The classical gas nitriding ratio of nitrogen to hydrogen is: 2NH3 ↔ 2N + 3H2

or a ratio of 1 part nitrogen to 3 parts hydrogen. A plain low-carbon steel will nitride at a conventional nitriding temperature of around 500 °C (925 °F) but with a resulting hardness of only about 35 HRC. Corrosion resistance will improve dramatically but not surface hardness. A higher process temperature of approximately 580 °C (1075 °F) should be considered, and temperatures can reach as high as 620 °C (1150 °F). Determination of the process temperature depends on the steel analysis and the surface metallurgical requirements. A plain low-carbon steel also requires a higher nitriding potential. This is accomplished by simply increasing the process gas nitrogen-to-hydrogen ratio by volume to a ratio of approximately 5:1 and up to 7:1. This will result in surface hardness values of up to 700 HV (59 HRC). As previously stated, the carbon source can come from the steel itself or by the addition of a hydrocarbon gas to the process gas flow. In the case of a plain low-carbon steel, there is insufficient carbon in the steel to promote formation of the dominant ε-nitride phase; therefore, hydrocarbon gas (methane or propane) must be added. The usual recommendation is approximately 2% by volume of the total gas flow of nitrogen and hydrogen. This will be sufficient to initiate and complete formation of the ε-phase in the steel surface.

Fig. 6

Typical plasma nitriding PLC screen display. Courtesy of Plateg GmbH

236 / Practical Nitriding and Ferritic Nitrocarburizing

How Is the Process Controlled? The process is controlled by the use of a personal computer/programmable logic controller (PC/PLC), which will receive appropriate signals from process monitoring points on the furnace. The computer will then respond to any set point signal deviation and take whatever corrective action is necessary (Fig. 6). To accurately control the gas flow, mass flow controllers are used. A mass flow controller can control flow rates ranging from centiliters per hour to liters per hour far more accurately than a conventional flowmeter.

How Much Gas Is Used during the Process? Gas consumption depends on: • • • • •

Temperature Time at temperature Surface metallurgy required Volume of the process vessel Surface area to be treated

All of these factors influence the volume of process gas required during the process time. A general estimate is approximately 1.5 L/min/m3 of the volume of the process vessel with a total work surface area of 22 ft2 (2 m2). Based on 1.5 L/min and given that there are approximately 4 L per U.S. gallon: 1.5 × 60 min =

90 L/h = 22.5 gal by volume of process gas 4

This makes plasma-assisted ferritic nitrocarburizing cost competitive when compared with the conventional gaseous procedure. The reason for the lower process gas consumption is that gas is being consumed strictly for plasma generation and is not being used as a “sweep gas” to prevent gas stagnation within the process chamber. Gaseous nitriding and ferritic nitrocarburizing processes often require an overpressure condition to ensure gas uniformity throughout the process chamber, particularly when considering blind holes. In addition, shorter process cycles greatly improve equipment productivity. If there is a continual flow of workload for the plasma nitride furnace, turnaround time is faster than for gaseous nitrocarburizing. Faster floor-tofloor cycle times, the elimination of postcleaning operations, and the potential elimination of postgrinding add up to a cost-effective procedure.

How Deep Can the Case Go? Do not be misled by claims of case depths of 0.75 mm (0.030 in.) in an hour. Such case depths could not possibly be achieved via carburizing other than by using high temperatures of around 1040 °C (1900 °F).

Chapter 21: Equipment for Ferritic Nitrocarburizing / 237

The rate of solid-state diffusion of any element into the surface of any steel is governed by the laws of the physics of diffusion. In other words, the diffused element cannot go into the steel any faster than the laws of physics will allow it to go. Many claims are made of very deep case depths being accomplished with very short cycle times at process temperatures in the region of 580 °C (1075 °F). The problem here is that “case depth” usually is not defined. Is the claim being made for total case depth or effective case depth? What is meant by core hardness? Is it the core plus 5 HRC points, or is it the actual core hardness itself? This is significant in terms of actual case depth. If the reported case depths are truly achievable in the times specified, then the process has a great deal more to offer both the engineer and the metallurgist than has been previously thought. It would make great sense to dispense with the carburize process and go with the ferritic nitrocarburize process in terms of: • • • •

Reduction of distortion Improved part cleanliness Improved productivity and efficiency Elimination of post-operation cleaning

A formula developed by F.E. Harris for case depth as a function of time and temperature for carburizing can serve as an approximate guide for plasma-assisted ferritic nitrocarburizing (Ref 1). The guide is based on a plain low-carbon steel, using the formula: Case depth = K √t

where the case depth is in inches (for case depth in millimeters divide by 25.4), t is the time in hours, and K is the temperature factor given in Table 1. The rate of nitrogen and carbon diffusion will begin to slow as the alloying content of a steel is increased by the addition of chromium, Table 1 Temperature factor for estimating case depth Temperature °C

°F

Temperature factor, K

495 510 525 540 550 565 580 585 595 605 620 635 650 660 675

925 950 975 1000 1025 1050 1075 1085 1100 1125 1150 1175 1200 1225 1250

0.00046 0.00056 0.00068 0.00081 0.00097 0.00116 0.00136 0.00146 0.00160 0.00187 0.00218 0.00252 0.00291 0.00334 0.00382

238 / Practical Nitriding and Ferritic Nitrocarburizing

molybdenum, aluminum, tungsten, vanadium, manganese, and especially nickel. It is difficult to say by how much the diffusion will be retarded because of the many available alloying element variations. For a steel containing all of these elements (not accounting for percentage variations), the diffusion rate can be retarded by as much as 17 to 20%.

Ferritic Oxynitrocarburizing This process is simply an addendum at the end of the ferritic nitrocarburizing procedure. Oxynitrocarburizing involves introducing oxygen— in the form of steam, oxygen, or nitrous oxide—in a controlled manner into the process chamber. The process gas must be carefully selected. Steam may cause problems with electrical equipment such as power feedthroughs and control systems. Nitrous oxide or oxygen is preferred; nitrous oxide tends to be more user friendly to valves and control systems than oxygen. The thickness of the oxygen-bearing compound zone after treatment will be determined by the time at temperature and the cooldown time. A typical part that has been ferritic oxynitrocarburized is shown in Fig. 7. The primary reason for oxygen treatment after ferritic nitrocarburizing is to enhance surface corrosion resistance. The procedure is comparable to

Fig. 7

Milling cutter after ferritic oxynitrocarburizing. Courtesy of Plateg GmbH

Chapter 21: Equipment for Ferritic Nitrocarburizing / 239

Fig. 8

Shafts treated by the oxynitride process (right) and untreated (left). Both subjected to salt spray test. Courtesy of Plateg GmbH

the black oxide type of treatment and will enhance the cosmetic surface appearance of the steel component (Fig. 8). Just as with salt bath treatments, the surface finish of the component will depend on the surface finish of the component prior to treatment. The higher the polish of the component prior to ferritic nitrocarburizing, the better the finish after the oxidizing procedure. The surfaces of oxynitrocarburized parts have been subjected to salt spray corrosion tests, exceeding 200 h by a considerable margin. Resistance to the salt spray test will be determined by the resulting oxide layer thickness. The oxidization treatment following ferritic nitrocarburizing has almost no cost attached to it, other than a portion of the amortization of the equipment. Generally, there is little or no power consumption. The only other associated expenses would be the cost of the oxidation process gas, followed by furnace occupancy. REFERENCES 1. Heat-Treating, Cleaning, and Finishing, Vol 2, Metals Handbook, 8th ed., p 98

Practical Nitriding and Ferritic Nitrocarburizing David Pye, p245-250 DOI: 10.1361/pnafn2003p245

CHAPTER

Copyright © 2003 ASM International ® All rights reserved. www.asminternational.org

23

Evaluating the Process PROCESS EVALUATION begins just as for the traditional nitriding methods. The same precautions must be taken so as to not disturb the sample face being investigated for case depth, case hardness profile, and visual microscopic evaluation. Sample cutoff, surface pregrind, polish, wash and rinse, and etch are conducted exactly as described in Chapter 16, “Examination of the Nitrided Case.”

Case Depth Evaluation Case depth can be measured by one, or all, of three methods: • • •

Total case depth to core hardness Total case depth to core hardness plus 5 HRC scale points Effective case depth to 513 HV hardness value

Each measurement method should be stated on the part drawing or by the customer, and whoever processes the work must be very clear as to what is required in terms of case depth. The most effective and accurate method of measuring case depth is by a microhardness traverse from the surface through the case of a sectioned, preground, and polished sample. Optical examination via light microscopy will show the case metallurgy, from which an assessment of the case depth can be made, but this method is not completely accurate. The load mass for the traverse will be determined by the accomplished case depth to be measured. Shallow case depths should be measured with a light load, up to 200 g. The load selection can go as low as 10 g; however, as the impression becomes smaller it will be become difficult to see and measure with an older microscope eyepiece.

Case Hardness It is strongly recommended that the case hardness not be measured directly on the steel surface with a heavy load (for example, at 150 kg as

246

/ Practical Nitriding and Ferritic Nitrocarburizing

used with the Rockwell C scale). The indenter is likely to penetrate through the formed case and into the core, thus giving a false hardness reading. For a nitrided case depth from 0.5 to 1.0 mm (0.020 to 0.040 in.), it is acceptable to use the Vickers hardness test system with a light load up to 10 kg (Fig. 1). The sample surface should be clean and free from decarburization and must be presented squarely 90° to the central axis of the penetrator. Any deviation from this will result in a false reading. The microhardness tester is usually used for a case traverse at right angles to the case (Fig. 2, 3). Microhardness testing requires a highly polished specimen surface in order to clearly see the diamond indenter impression in the case (Fig. 4). The sample must be held firmly in an appropriate holding device (Fig. 5).

What If the Formed Case Has Low Hardness Values? There are two reasons for low case hardness; each is a result of the process control. First, poor precleaning of the material surface can reduce nitrogen diffusion into the steel. The steel surface must be clean and free from contaminants such as chlorides, sulfides, silicon products, cutting oils, and lapping compounds.

Fig. 1

Vickers hardness testing machine

Chapter 23: Evaluating the Process / 247

Fig. 2

Fig. 3

Microhardness testing machine

Microhardness testing system with features including cameras and computer imaging. Courtesy of NewAge Industries

248

/ Practical Nitriding and Ferritic Nitrocarburizing

Fig. 4

Fig. 5

Comparison of identations made by Knoop and Vickers indenters

Typical fixtures used for holding and clamping workpieces for microhardness testing

The low hardness generally indicates that the case has not been processed to the correct depth specification. In other words, the case depth is too shallow and the indenter punches through it. The remedy is to either increase the case depth requirement or, if the case depth is correct, select a lighter load for the hardness test. The second reason is an imbalance of the furnace atmosphere. If the gaseous method of ferritic nitrocarburizing is being used, check the gas flow rates and relative volumes. Next, check the gas decomposition within the furnace. This can be accomplished by a gas analyzer that measures the carbon monoxide, carbon dioxide, and hydrogen contents and adjusts the gas composition and flow rates accordingly.

Chapter 23: Evaluating the Process / 249

If using the salt bath method of ferritic nitrocarburizing, analyze the salt composition for cyanide and cyanate composition. If the resulting analysis shows an excess of carbonate or cyanate, regenerate the bath. This is accomplished simply by raising the bath temperature to 595 °C (1100 °F) and vigorously aerating the bath for approximately 90 min, followed by desledging. If the plasma-assisted technique of ferritic nitrocarburizing produces a shallow case, then: • •

The process cycle time is too short to accomplish the required case depth (assuming that the gas flow rates are correct). The gas flow rates could be incorrect. Increase the nitrogen flow rate to increase the nitriding potential of the process.

In the fluidized-bed method of ferritic nitrocarburizing, a shallow case can occur due to low nitrogen volumes. Set the process gas flows accurately at the outset.

Corrosion If surface corrosion develops after ferritic nitrocarburizing, the initiating source must be established. For example, is the corrosion originating from an external environmental source? If so, check the operating atmosphere or liquid for acidic content and adjust the process parameters accordingly. The corrosion may also be generated by an uncontrolled salt bath, gas mixture, or plasma condition. If this is the case, check that the salt bath composition is within specified operating limits and adjust accordingly. Check also that the gas dissociation and gas ratios are within the required operating limits and that the process temperature is correct. Lastly, check that the plasma conditions are correctly set in terms of power density, voltage, amperage, pressure, gas ratios, and process temperature. Material selection may also play a role in initiating corrosion. Does the selected material have the necessary corrosion resistance for the given environment? Even with improved corrosion resistance imparted by the nitrocarburizing treatment, a more corrosion-resistant base material may have to be selected.

Distortion Distortion is always a contentious subject. Despite the low nitrocarburizing process temperature, there is no absolute guarantee that distortion will not occur. Distortion can be classified as either shape distortion (warpage) or size distortion (Ref 1). Shape distortion originates from induced residual stresses from such operations as forging, rolling, and machining and will result in twisting,

250

/ Practical Nitriding and Ferritic Nitrocarburizing

Melting chemistry variations

Annealing temperature

Rough machine

Stress relieve Distortion

Rolling or forging temperature

Normalize temperature

Fig. 6

Preheat treatment

Final machine

Ferritic nitrocarburize

Flow diagram showing factors contributing to distortion

bending, and ovality. The degree of distortion that occurs will be determined by the amount of residual stress present in the steel. The residual stress can be in the steel as a result of mixed phases due to incomplete transformation at the preharden and temper operation. The only way to reduce the induced mechanical stress is to stress relieve prior to final machining and prior to nitriding. To deal with the possible mixed phase problem, resulting from pre-heat-treatment, it may be necessary to cryogenically treat the steel, followed by tempering to deal with any transformed retained austenite to martensite. This will stabilize the steel prior to the ferritic nitrocarburizing procedure if retained austenite was present. Size distortion results from thermochemical treatments such as ferritic nitrocarburizing, and from variations in the steel chemistry from heat to heat. If the steel has been prehardened and tempered, the prior austenitizing temperature, as well as time at the austenitizing temperature, will influence grain size (Ref 2). The flow diagram in Fig. 6 presents factors that influence distortion during final heat treatment. Remember, both ferritic nitrocarburizing and nitriding will act as a stress-relieving process. If residual induced stress has not been dealt with beforehand, then the procedure will act as a stress relieve operation and distortion will occur. Steel will respond to heat from any source. REFERENCES 1. G.E. Totten, C.E. Bates, and N.A. Clinton, Handbook of Quenchants and Quenching Technology, ASM International, 1993, p 443 2. G.E. Totten and M.A.H. Howes, Chapter 5, in Steel Heat Treatment Handbook, Marcel Dekker, 1997, p 251–292

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

Index A Alloy steels AISI 4140, microstructure 182(F), 183(F) compositions 126(T) compound zone thickness 143(F), 144(F), 145(F), 146(F), 148(F), 149(F), 182(F), 183(F) hardness 35, 160, 161 salt bath ferritic nitrocarburizing 211, 212(F) salt bath nitriding 188 Alloying elements, effect of 4–5, 7–8, 20–21, 35, 127–129, 127(F), 128(F), 161, 203(F), 206 American Gas Co. 3 Ammonia. See also Gas nitriding decomposition of 3, 23–27, 50–51, 77–78, 77(F), 117–118, 185–186, 220 leaks in retorts 44–46 safety precautions 46–47 Applications 153–161 Aubert, Pierre 6 Automotive crankshafts, gaseous ferritic nitrocarburizing of 223–229

B Bell, Thomas 67, 201 Berghaus, Bernhard 9, 72 British standard nitriding steels 5, 5(T)

C Carbon, and compound zone 32 Carbon steels AISI 1015, nitrocarburized microstructure 35(F) AMS 6470, nitride networking 40(F) AMS 6470, nitrided case 181(F), 182(F) hardness 35, 203(F)

nitriding 35–36 for retorts 43 SAE 5115, salt bath nitrided 57(F) salt bath ferritic nitrocarburizing 211, 211(F), 212(F), 216(F) Case depth. See also Nitrided case; Surface hardness automotive crankshafts 227 carburizing vs. nitriding 13(F), 14(F), 15–16, 15(F), 16(F) determination of 68–69, 69(T), 136–137 extrusion dies 155 ferritic nitrocarburizing 245 fluidized–bed nitriding 117(F), 118(T) forging dies 154 gears 159, 160(F) ion ferritic nitrocarburizing 236–237, 237(T) ion nitriding 136–137 microhardness testing 177, 245–248, 246(F), 247(F), 248(F) temperature factor 69(T), 137(T), 237(T) Case formation 31–37. See also Compound zone (layer) Central Alloy Steel Corp. 7 Compound zone (layer) in automotive crankshafts 223, 224(F) carbon, influence of 32 control of 65–69, 82–83 dual phase formation 4(F), 31(F), 32(F), 33(F), 65(F), 66(F), 141(F) early studies 7 in extrusion dies 83(F), 155–156 and Floe process 8–9, 34, 66, 67(F) in forging dies 154 in gears 159 kinetic studies 142–150 and Melonite process 203–204, 204(F) and sputtering 142–150 tests 66 thickness 32–34, 35, 36, 118(T), 142–150, 143(F), 144(F), 145(F), 146(F), 147(F), 148(F), 149(F), 205(F), 220–221 and two stage process 66, 67(F)

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

252 / Practical Nitriding and Ferritic Nitrocarburizing

Contaminants. See Surface contaminants Copper deoxidized, for retorts 44 for selective nitriding 163–164 Core hardness automotive crankshafts 227 extrusion dies 155(F) forging dies 153, 154(F) gears 159 high-speed steel cutters 157 and preheat treatment 28–29 Corner effect. See Nitride networking Costs, various surface treatments 198(F) Crankshafts, gaseous ferritic nitrocarburizing of 223–229 Cutters, high-speed steel 156–157 Cutting fluids. See Surface contaminants

D Dawes, Cyril 195 Decarburization. See Surface contaminants Degussa 54, 55, 196, 201 Deposition, by ion nitriding 76–77, 157 Die casting, and gaseous ferritic nitrocarburizing 222 Dies drawing, wear behavior 207(F) extrusion 81(F), 83(F), 155–156, 155(F), 156(F), 157(F), 165(F) forging 153–155, 154(F), 213(F) Diffusion zone (layer) 4(F), 31–32, 31(F), 32(F), 35, 65(F), 203(F), 204–205 Dilution process 3, 5, 67–68 Distortion control of 18–20, 120 and ferritic nitrocarburizing 194, 249, 250(F) in gears 158–159 growth 19–20, 19(F), 27–28, 27(F), 120, 121–122, 158–159 postmachining 122–123 shape 18, 120, 121, 249 size 18, 119–120, 249 stock removal 122 Double-stage process. See Floe process; Two-stage process

E Electroplating 163, 164 Enameling, for retorts 44 Equilibrium diagrams iron-carbon 17(F), 194(F) iron-nitrogen 2(F) Equipment. See Furnaces; Oxygen probes; Programmable logic controllers; Sensors; Thermocouples; Trays and fixtures

Etching 177–179 Examination methods 167–183

F Ferritic nitrocarburizing. See also Ferritic oxynitrocarburizing; Gaseous ferritic nitrocarburizing; Ion ferritic nitrocarburizing; Salt bath ferritic nitrocarburizing advantages 193–194, 198–199 corrosion 248–249 costs 198–199, 198(F) distortion 249, 250(F) early history 195–198 equipment 231–239 hardness testing 245–248 process evaluation 245–250 surface preparation 241–243 training 199–200 Ferritic oxynitrocarburizing 234, 234(F), 238–239, 238(F), 239(F) Fingerprints. See Surface contaminants Firth Brown Steels 5 Fixtures. See Trays and fixtures Floe, Carl F. 9, 34, 66 Floe process 8–9, 26. See also Two-stage process Fluidized-bed nitriding furnaces 111–118, 112(F), 113(F), 114(F), 115(F) gas dissociation 117–118 oxynitriding 116 Fry, Adolph 4–5, 6, 7, 8, 14, 20, 127 Furnaces fluidized-bed nitriding 111–118, 112(F), 113(F), 114(F), 115(F) gas nitriding 8(F), 23(F), 39–51, 39(F), 41(F), 45(F) gaseous ferritic nitrocarburizing 224–227, 226(F), 233 ion ferritic nitrocarburizing 233–237, 233(F) ion nitriding 10(F), 84, 85(F), 89–109, 89(F), 95(F), 95(T), 99(F), 107(F), 109(F) salt bath ferritic nitrocarburizing 231, 232(F) salt bath nitriding 55–63, 56(F)

G Gas ionization. See Ion nitriding Gas nitriding. See also Nitriding case crushing 187 discoloration 186 exfoliation 186–187 furnace design 8(F), 23(F), 39–51, 39(F), 41(F), 45(F) furnace heating 47–49

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org Index / 253

gas circulation 40–41, 41(F) gas dissociation 50–51, 185–186 insulation 42–43 leak prevention 44–46 orange peel effect 187 oxygen probe control 51 process control and instrumentation 49–51 retort construction and maintenance 43–44 safety precautions 46–47 sensors 51 stop-off procedures 163–164 surface contaminants 185, 186–187 temperature control 40, 49–50 troubleshooting 185–187 Gaseous ferritic nitrocarburizing applications 222–223 of automotive crankshafts 223–229 early history 195–196 furnaces 224–227, 226(F), 233 gaseous supply 221 process development 219–220 process principles 220–221 safety considerations 223 surface cleanliness 241–242 surface properties 221–222, 227, 228(F) trade names 202(T), 220 Gears 21, 81(F), 158–159, 158(F), 160(F), 187, 222 General Electric Co. 10, 72 Glass, for retorts 43–44 Glow discharge process. See Ion nitriding Growth. See Distortion

H Hardness. See Core hardness; Hardness testing; Surface hardness Hardness testing of automotive crankshafts 227, 228(F) hardness profiles 168, 168(F), 169(T), 228(F) macrohardness 167 microhardness 167, 177, 245–248, 246(F), 247(F), 248(F) sample preparation 168–177 Harris, F.E. 69 Heat input requirement 48–49 High-speed steels compositions 126(T) cutters 156–157 Homerberg, V.O. 8 Houghton Durferrit 202, 203

I Imperial Chemical Industries 54, 195 Insulation, for gas furnaces 42–43 Interstitial diffusion 24, 25(F)

Ion ferritic nitrocarburizing case depth 236–237, 237(T) corrosion resistance 234, 234(F), 238–239, 239(F) early history 197–198 furnaces 233–237, 233(F) gas consumption 236 oxynitrocarburizing 234, 234(F), 238–239, 238(F), 239(F) process control 234–235 surface cleanliness 243 trade names 202(T) Ion nitriding advantages 85–86, 107–108 of blind holes and cavities 75, 102–103, 103(F), 165(F) case depths 136–137 cathode and anode 93–94 cold-wall continuous dc system 84, 85(F), 89–94, 89(F), 95(T), 100(F), 190, 190(F) compound zone 34, 68, 82–83, 142–150, 143(F), 144(F), 145(F), 146(F), 148(F), 149(F) deposition techniques 76–77 diffusion techniques 76 early developments 9–11, 71–72 equipment 89–109 and ferritic nitrocarburizing 197–198 furnace loading 105 furnaces 10(F), 84, 85(F), 89–109, 89(F), 95(F), 95(T), 99(F), 107(F), 109(F) of gears 158–159, 160(F) glow discharge characteristics 73(F), 74–75, 74(F), 79(F), 96(F), 106(F), 189(F) heating elements 91, 97(F), 103–104 hollow cathode effect 102–103, 103(F) hot-wall pulsed dc system 84, 85(F), 94–101, 95(F), 95(T), 97(F), 99(F), 100(F), 191, 191(F) of maraging steels 161 masking 105–106, 164–165, 165(F) nitride networking 80–81, 80(F), 81(F) oxynitriding 86–87, 87(F) plasma generation 84, 85(F), 90–91, 98–101 process control 75–76, 84, 90, 105, 139–142 process gas flow 92, 139–142 process gas ratios 77–78 process gases 83 process principles 72–73, 73(F), 77–80 of pure irons 160 sputter cleaning 104–105, 104(F), 164–165 sputtering 142–150 of stainless steels 129–136 stop-off procedures 105–106, 164–165, 165(F) surface degradation 81 surface reactions 78–80 surface stability 80 thermocouples 91–92 troubleshooting 188–191 vacuum pumps 92–93, 93(F), 94(F) work cooling 101–102

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

254 / Practical Nitriding and Ferritic Nitrocarburizing

Irons cast, salt bath ferritic nitrocarburized 214, 215(F), 216(F) ferritic nodular, salt bath nitrided 57(F) nitrided case formation 31–37, 33(F) pure 159–160

J Jack, D.H. 67 Jeffries, Zay 6 Jones, Claude 10, 72 Joseph Lucas Ltd. 195, 219

K Ketcham, W.J. 6–7 Kinetic studies, of compound zone formation 142–150 Klockner Ionen GmbH 9, 72 Kolene Corp. 55, 207, 208 Krupp Steel Works 4, 5, 6, 14, 127

L Leyland Vehicles Ltd. 224 Light microscopy. See Microscopic examination Lightfoot, B.J. 67 Ludlum Steel Co. 8

M Machlet, Adolph 3–4, 6, 13, 14, 67 Maraging steels 160–161, 183(F) Martin, Stuart 10, 72 Masking. See Selective nitriding Mass spectrometry, for process gas control analysis 140–141 Massachusetts Institute of Technology 8, 9, 66 McQuaid, H.W. 6–7 Melonite process 55, 202, 203–207 Metallographic examination. See Microscopic examination Microhardness. See Hardness testing Microscopic examination of automotive crankshafts 227, 228(F) etching 177–180 microscope selection 180–181 nitrided case microstructures 181(F), 182(F), 183(F) sample preparation 168–177 Microstructures, nitrided iron and steel 31–37, 32(F), 35(F), 181(F), 182(F), 183(F) Molding resins 171(T), 172(T)

N Nitralloy steels 5, 14, 20–21, 125, 126(T) Nitrex Metal, Inc. 68 Nitride networking 5, 16, 34, 34(F), 40(F), 80–81, 80(F), 81(F), 154–155 Nitrided case. See also Case depth; Surface hardness alloying elements, effect of 128(F) examination methods 167–183 on forging die 154(F) structure of 4(F), 32(F), 35(F), 65(F), 181(F), 182(F), 183(F) Nitriding. See also Floe process; Fluidized-bed nitriding; Gas nitriding; Ion nitriding; Salt bath nitriding; Selective nitriding advantages 13–22, 19(F) applications 153–161 current status of technology 11 examination methods 167–183 furnace equipment and control systems 39–51 historical background 3–11, 13–14 metallurgical considerations 1–2 microstructures 31–37, 32(F), 35(F), 181(F), 182(F), 183(F) process principles 23–29 process requirements 2–3, 14–21 steels for 125–137, 153–161 troubleshooting 185–191 U.S. vs. German processes 5–6 Nitriding potential 26, 67 Nitrocarburizing. See Ferritic nitrocarburizing Nitrogen diffusion 24–27, 33–34, 139–142 See also Solubility limit Nu-Tride process 207, 208–217

O Oils. See Surface contaminants Optical light microscopy. See Microscopic examination Oxidation, resistance to 21 Oxygen probes 51 Oxynitriding 86–87, 87(F), 116. See also Ferritic oxynitrocarburizing

P Paint residue. See Surface contaminants Paints, stop-off 164–165 Paschen curves 74, 74(F), 96, 96(F) Phase transformations 17(F), 18, 18(F), 24, 26–27, 28–29 Photo spectrometry, for process gas control analysis 139–140, 140(F)

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org Index / 255

Piston rods, oxynitrided 87(F) Plasma nitriding. See Ion nitriding Plasma-assisted chemical vapor deposition 76–77, 157 Plasma-assisted nitrocarburizing. See Ion ferritic nitrocarburizing Preheat treatment 28–29 Process chambers. See Retorts Programmable logic controllers 39, 39(F), 49, 76, 101, 139–140, 140(F), 235, 235(F)

Q–R QPQ process 55, 205(F), 207–208, 208(F), 209–210, 209(F), 210(F), 211(F) Quench requirements 16–18 Refractory firebrick, for retorts 44 Resins, molding 171(T), 172(T) Retorts construction 43–44 maintenance 44 sealing 44–46, 45(F) Rocker arms, wear resistance 206–207, 207(F) Ryzhov, N. 139

S Safety precautions for ammonia use 46–47 etchants 179, 180(F) gaseous ferritic nitrocarburizing 223 salt bath nitriding 62–63 vapor degreasing 179–180 Salt bath ferritic nitrocarburizing corrosion resistance 208, 208(F), 209(F), 217 dimensional stability 209 early history 195–196, 196(F), 197(F) engineering properties 215–217 fatigue strength 217, 218(T) furnaces 231, 232(F) low-cyanide 202–207 lubricity 216, 217(F) Melonite process 55, 202, 203–207 metallurgical results 211–215, 216(F) Nu-Tride process 207, 208–217 post treatment 207–208 process control 210 process parameters 201–202, 209–210 QPQ process 205(F), 207–208, 208(F), 209–210, 209(F), 210(F), 211(F) surface cleanliness 241–242 surface hardness 206, 206(T) Sursulf process 217 Tenoplus process 217 tensile strength 206, 206(T) trade names 202(T)

wear resistance 206–207, 207(F), 215–216, 217(F) Salt bath nitriding advantages 54 bath aging 55–58 bath maintenance 61–62, 188 bath replacement 58 bath testing and analysis 58–61 early developments 9, 54 equipment 55, 56(F), 58, 63 furnace types 56(F) procedure 55–63 process types 54–55 safety precautions 62–63 salts 53–54, 58, 60–61, 187 stop-off procedures 164 surface contaminants 57, 59 troubleshooting 187–188 Sample preparation cutting 168–170, 170(F) mounting 171–174, 171(T), 172(F), 172(T), 173(T) polishing 176–177, 176(F) pregrinding 174–176, 174(F), 175(F) vapor degreasing 170, 179–180 Seals, and ammonia leaks 44–46, 45(F) Selective nitriding 105–106, 163–165 Sensors 51 Sergeson, Robert 7–8 Sintered steels 223 Society of Manufacturing Engineers (SME) 6, 14 Solubility limit, nitrogen in iron 2, 139–142 Sputter cleaning 104–105, 105(F), 243 Sputtering, and ion nitriding 142–150 Stainless steels corrosion resistance 36, 129, 134(F) hardness 35, 129 nitridability 129–137 for retorts 43 salt bath ferritic nitrocarburizing 214, 214(F) type 304, salt bath ferritic nitrocarburized 216(F) type 316, salt bath ferritic nitrocarburized 214(F) type 422, ion nitriding of 130–131, 132(F), 134(F) type 440A, ion nitriding of 131–134 type 440C, tempering curve 133(F) type 630 (17–4 PH), ion nitriding of 135–136, 136(F) Steels. See also Alloy steels; British standard nitriding steels; Carbon steels; Highspeed steels; Maraging steels; Nitralloy steels; Sintered steels; Stainless steels; Tool steels compositions 126(T) compound zone thickness 143(F), 144(F), 145(F), 146(F), 148(F), 149(F) nitrided case formation 31–37 selection 125–126

© 2003 ASM International. All Rights Reserved. Practical Nitriding and Ferritic Nitrocarburizing (#06950G)

www.asminternational.org

256 / Practical Nitriding and Ferritic Nitrocarburizing

Stop-off procedures 163–165 Stress relieving. See Distortion Sturges, Derek 10, 72 Sulfinuz process 195, 201 Surface contaminants and case formation 36–37 in ferritic nitrocarburizing 241–243 in gas nitriding 185, 186–187 in salt bath nitriding 57, 59 Surface hardness. See also Case depth; Nitrided case and alloying elements 4–5, 7–8, 35, 127–129, 127(F), 160, 161, 203(F), 206 automotive crankshafts 227 diffusion techniques compared 16(F) early results 14 high-speed steel cutters 156–157 maraging steels 161 Melonite process 203–204, 203(F), 204(F), 206, 206(T) pure irons 160 stainless steels 161 values 20–21 Sursulf process 217

T Temperature control 49–50, 154–155 factors 69(T), 137(T), 237(T) requirements 14–16

Tenoplus process 217 Thermocouples 49–50, 91–92 Thin-film deposition 76–77, 157 Timken Detroit Axle Co. 6 Tool steels compositions 126(T) D2, microstructure 216(F) extrusion dies 155–156, 155(F), 156(F) forging dies 153–155, 154(F) H13, microstructure 182(F), 216(F) salt bath ferritic nitrocarburizing 211, 213(F), 216(T) salt bath nitriding 188 Townsend discharge 74, 74(F), 96(F) Transition zone 4(F), 31(F) Trays and fixtures maintenance 44 for salt bath ferritic nitrocarburizing 197(F) Troubleshooting 185–191 Tufftride process 54, 55, 196, 201, 203 Two-stage process 34, 66, 67(F). See also Floe process

U–Z University of Aachen 72 Vapor degreasing 36, 170, 179–180 Walsted, J.P. 8 Wehnheldt, Dr. 9, 71, 72 White layer. See Compound zone (layer)

ASM International is the society for materials

engineers and scientists, a worldwide network dedicated to advancing industry, technology, and applications of metals and materials. ASM International, Materials Park, Ohio, USA www.asminternational.org This publication is copyright © ASM International®. All rights reserved. Publication title

Product code

Practical Nitriding and Ferritic Nitrocarburizing

#06950G

To order products from ASM International: Online Visit www.asminternational.org/bookstore Telephone 1-800-336-5152 (US) or 1-440-338-5151 (Outside US) Fax 1-440-338-4634 Mail

Customer Service, ASM International 9639 Kinsman Rd, Materials Park, Ohio 44073-0002, USA

Email [email protected]

American Technical Publishers Ltd. 27-29 Knowl Piece, Wilbury Way, Hitchin Hertfordshire SG4 0SX, In Europe United Kingdom Telephone: 01462 437933 (account holders), 01462 431525 (credit card)

www.ameritech.co.uk Neutrino Inc. In Japan Takahashi Bldg., 44-3 Fuda 1-chome, Chofu-Shi, Tokyo 182 Japan Telephone: 81 (0) 424 84 5550 Terms of Use. This publication is being made available in PDF format as a benefit to members and customers of ASM International. You may download and print a copy of this publication for your personal use only. Other use and distribution is prohibited without the express written permission of ASM International. No warranties, express or implied, including, without limitation, warranties of merchantability or fitness for a particular purpose, are given in connection with this publication. Although this information is believed to be accurate by ASM, ASM cannot guarantee that favorable results will be obtained from the use of this publication alone. This publication is intended for use by persons having technical skill, at their sole discretion and risk. Since the conditions of product or material use are outside of ASM's control, ASM assumes no liability or obligation in connection with any use of this information. As with any material, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specific testing under actual conditions is recommended. Nothing contained in this publication shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters patent, copyright, or trademark, and nothing contained in this publication shall be construed as a defense against any alleged infringement of letters patent, copyright, or trademark, or as a defense against liability for such infringement.