Practical Nitriding and Ferritic Nitrocarburizing - D. Pye (ASM, 2003) WW

Practical

NITRIDINGand Ferritic Nitrocarburizing

David Pye

ASM InternationalMaterials Park, Ohio 44073-0002

www.asminternational.org

FM_Nitriding.qxd 10/6/03 2:04 PM Page i

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


Copyright © 2003by

ASM International®

All rights reserved

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 WAR-RANTIES, EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF MER-CHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE GIVEN IN CONNECTION WITH THISPUBLICATION. Although this information is believed to be accurate by ASM, ASM cannot guarantee that favorableresults will be obtained from the use of this publication alone. This publication is intended for use by persons havingtechnical skill, at their sole discretion and risk. Since the conditions of product or material use are outside of ASM’scontrol, 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 inamount than the purchase price of this product or publication in respect of which damages are claimed. THE REM-EDY HEREBY PROVIDED SHALL BE THE EXCLUSIVE AND SOLE REMEDY OF BUYER, AND IN NOEVENT SHALL EITHER PARTY BE LIABLE FOR SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGESWHETHER OR NOT CAUSED BY OR RESULTING FROM THE NEGLIGENCE OF SUCH PARTY. As with anymaterial, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specifictesting 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, inconnection with any method, process, apparatus, product, composition, or system, whether or not covered by letterspatent, copyright, or trademark, and nothing contained in this book shall be construed as a defense against any allegedinfringement 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; BonnieSanders, Manager of Production; Nancy Hrivnak, Jill Kinson, and Carol Polakowski, Production Editors; and ScottHenry, 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 2003671.3'6—dc21

2003056298

ISBN: 0-87170-791-8SAN: 204-7586

ASM International®

Materials Park, OH 44073-0002www.asminternational.org

Printed in the United States of America

FM_Nitriding.qxd 10/6/03 2:04 PM Page ii



Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

PART 1 Nitriding

CHAPTER 1 An Introduction to Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Metallurgical Considerations and Process Requirements . . . . . . . . . . . 1The Pioneering Work of Machlet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Parallel Work in Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Developments in the United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Other Early Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Current Status of Nitriding Technology . . . . . . . . . . . . . . . . . . . . . . . . 11

CHAPTER 2 Why Nitride? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Key Process Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

CHAPTER 3 How Does the Nitriding Process Work? . . . . . . . . . . . . . . . . . . . . . 23

The Liberation of Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24Dissociation of the Gas at the Selected Nitriding Temperature . . . . . 25Why Ammonia Is Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27Preheat Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

CHAPTER 4 Microstructures of Nitrided Iron and Steel . . . . . . . . . . . . . . . . . . 31

Influence of Carbon on the Compound Zone . . . . . . . . . . . . . . . . . . . 32Controlling Compound Zone Thickness . . . . . . . . . . . . . . . . . . . . . . . 32What Happens Below the Compound Zone? . . . . . . . . . . . . . . . . . . . . 35

iii

Contents

FM_Nitriding.qxd 10/6/03 2:04 PM Page iii



Can Plain Carbon Steel Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . 35Calculating the Compound Zone Thickness . . . . . . . . . . . . . . . . . . . . 36Other Factors Affecting Surface Case Formation . . . . . . . . . . . . . . . . 36

CHAPTER 5 Furnace Equipment and Control Systems . . . . . . . . . . . . . . . . . . . 39

Essential Furnace Design Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40Types of Nitriding Furnaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42Determining Appropriate Furnace Design . . . . . . . . . . . . . . . . . . . . . . 43Retort Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43Retort Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44Sealing the Retort to Prevent Ammonia Leaks . . . . . . . . . . . . . . . . . . 44Safety Precautions When Using Ammonia . . . . . . . . . . . . . . . . . . . . . 46Furnace Heating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47Process Control and Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . 49

CHAPTER 6 Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Salts Used and Process Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . 53Types of Salt Bath Nitriding Processes . . . . . . . . . . . . . . . . . . . . . . . . 54Salt Bath Nitriding Equipment and Procedure . . . . . . . . . . . . . . . . . . . 55Using a New Salt Bath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55Bath Replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58Bath Testing and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58Bath Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61Operating the Salt Bath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62Safety Precautions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62Design 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 . . . . . . . . . . . . . 66Reduction of the Compound Zone by the Two-Stage Process . . . . . . 66Other Methods for Controlling Compound Zone Formation . . . . . . . 67Case Depth of Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

CHAPTER 8 Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

History of Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71How the Ion Nitriding Process Works . . . . . . . . . . . . . . . . . . . . . . . . . 72Glow Discharge Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74Process Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75Other Uses for Plasma Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

iv

FM_Nitriding.qxd 10/6/03 2:04 PM Page iv



What Happens in the Ion Nitriding Process . . . . . . . . . . . . . . . . . . . . 77Gas Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Reactions at the Steel Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78Surface Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80“Corner Effect” and Nitride Networking . . . . . . . . . . . . . . . . . . . . . . . 80Degradation of Surface Finish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81Control of the Compound Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82Process Gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83Process Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84Plasma Generation Philosophies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85Oxynitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

CHAPTER 9 Ion Nitriding Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Cold-Wall Continuous dc Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . 89Hot-Wall Pulsed dc Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . 94Work Cooling after Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . 101Other Considerations for Ion Nitriding Equipment

and Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102Summary: Advantages of Plasma Nitriding . . . . . . . . . . . . . . . . . . . . 107

CHAPTER 10 Nitriding in Fluidized Beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Heating Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111Nitriding in the Fluidized-Bed Furnace . . . . . . . . . . . . . . . . . . . . . . . 114Oxynitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116Operating the Fluid Bed for Nitriding . . . . . . . . . . . . . . . . . . . . . . . . 117Measurement of the Gas Dissociation . . . . . . . . . . . . . . . . . . . . . . . . 117

CHAPTER 11 Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Size Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119Shape Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120Control of Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120Distortion in Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121Stock Removal Prior to Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122Postmachining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

CHAPTER 12 Steels For Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

Steel Selection Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125Requirements for a Nitriding Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

v

FM_Nitriding.qxd 10/6/03 2:04 PM Page v



Can Stainless Steels Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129Plasma Nitride Case Depths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

CHAPTER 13 Control of the Process Gas in Plasma Conditions . . . . . . . . . . . . 139

Analysis by Photo Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139Analysis by Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140Difficulties Associated with Gas Analysis . . . . . . . . . . . . . . . . . . . . . 141Kinetic Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142

Appendix: The Role of Sputtering in Plasma Nitriding . . . . . . . . . . . . . 142Experimental Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

CHAPTER 14 Processing with Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Hot-Work Tool Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153High-Speed Steel Cutters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158Pure Irons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159Low-Alloy Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160Maraging Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160Higher Alloyed Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

CHAPTER 15 Stop-Off Procedures for Selective Nitriding . . . . . . . . . . . . . . . . 163

Methods for Selective Gas Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . 163Methods for Selective Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . 164Methods for Selective Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . 164

CHAPTER 16 Examination of the Nitrided Case . . . . . . . . . . . . . . . . . . . . . . . . 167

Hardness Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167Etching of the Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177Safety Precautions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179Optical Light Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180

CHAPTER 17 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

Gas Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188

vi

FM_Nitriding.qxd 10/6/03 2:04 PM Page vi



PART 2 Ferritic Nitrocarburizing

CHAPTER 18 What Is Meant by Ferritic Nitrocarburizing? . . . . . . . . . . . . . . . . 193

Process Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193Early History of Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . 195Why Ferritic Nitrocarburize? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

CHAPTER 19 Salt Bath Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . 201

Low-Cyanide Salt Bath Ferritic Nitrocarburizing . . . . . . . . . . . . . . . 202Salt Bath Nitrocarburizing plus Post Treatment . . . . . . . . . . . . . . . . 207Kolene Nu-Tride Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208Other Methods for Salt Bath Nitrocarburizing . . . . . . . . . . . . . . . . . . 217

CHAPTER 20 Gaseous Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . 219

Development of the Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219Process Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220Gaseous Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221Properties of Gaseous Ferritic Nitrocarburized Components . . . . . . 221Industrial Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222Safety Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

Appendix: Gaseous Nitrocarburizing—A Suitable Alternative for the Heat Treatment of Automotive Crankshafts . . . . . . . . . . . . 223

Process Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224Typical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

CHAPTER 21 Equipment for Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . 231

Salt Bath Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231Atmosphere Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233Plasma-Assisted Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . 233Ferritic Oxynitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

CHAPTER 22 Preparation for Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . 241

Gas Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241Enhanced Plasma Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

vii

FM_Nitriding.qxd 10/6/03 2:04 PM Page vii



CHAPTER 23 Evaluating the Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

Case Depth Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245Case Hardness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245What If the Formed Case Has Low Hardness Values? . . . . . . . . . . . 246Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

viii

FM_Nitriding.qxd 10/6/03 2:04 PM Page viii



Usually one seeks out a career, and, although I chose my career, I neverrealized how that choice was to impact my life. Although I chose heattreatment as a career, I did not make a conscious choice for nitriding; thesubject 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 evaluatethe nitriding process for the DeHavilland Aircraft Group. It was at thattime 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. Yeteach time that I have researched the subject, I have found very fewresource materials available. Unlike carburizing, for example, the subjectof nitriding has had very few reference books or “cook books” written onthe subject. A few books include a chapter or two on nitriding, and someconference papers are available in proceedings volumes. Up to now, how-ever, there has not been a practical “how to” or “why to” book availableon the subject.

In 1991, Rodney Allwood of the ASM Education Department urged meto 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 Iwas able to put together the notes for that course. This was the foundationfor a book on nitriding. Mrs. Veronica Flint of the ASM Reference Publi-cations Department challenged me to write such a book and, with hertremendous patience and persistence, forced me to find the time to put pento paper.

Many books are dedicated to husbands, wives, children, or even dogsand cats (we have 8!). I can only dedicate this book to the young heat treateror metallurgist who is coming into the industry and to my colleagues whohave, without exception, given me tremendous encouragement.

I want to remember my colleagues in South Africa, where to a largeextent I learned my trade. Without that 20 years experience in SouthAfrica, I could not do what I am doing today. I also want to remember my

ix

Preface

FM_Nitriding.qxd 10/6/03 2:04 PM Page ix



colleagues at DeHavilland Propellors with whom I worked from 1956 to1963, especially Wally Simms, our foreman (now deceased), who had theforesight and tenacity to fight to bring apprentices into the world of heattreatment and the DeHavilland Propellers (now British Aerospace)apprentice program; Bill Oddey, our heat treatment superintendent; JoeAspinall, our metallurgist; and Alex Thexton, our chemist (who sold mehis briefcase, which I still have), who is now somewhere in Australia. Wewill always be grateful to Paul Huber at Seco Warwick, who brought myfamily 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 beforeme 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 givethanks to Dr. Siegfeid Stramke in Germany. I especially would like to rec-ognize Dr. Reinar Grun, who has put up with me, debated with me vigor-ously, and helped me learn the subject. He is a man whom I hold in greatrespect in the field of plasma physics.

I would like to acknowledge both George Totten and Maurice Howes forbeing mentors to me and for their very positive encouragement. Their sup-port has been a privilege and continues to be a very rewarding experience.

I would like also to acknowledge Grace Pye. She always believed that Iwas capable of “going on my own.”

It took my wife Lynn to push me to the successful completion of thisand 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 herpatience, dedication, and support. This book could not have come togetherwithout her. She typed the original manuscript from my almost illegiblehandwriting and “chicken scratching.” Thank you for your support of meand for listening to my heat treatment and furnace stories. You really are agift to me.

I also would like to recognize Valerie Sales, my mother-in-law(deceased) who interpreted my hand drawn sketches and turned them intoillustrations, 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 andefforts to reviewing the manuscript. I would also like to acknowledge theeditorial contribution of Joseph R. Davis, Davis & Associates, who helpedpolish the chapters and prepare them for the production process.

Purpose

Nitriding and ferritic nitrocarburizing in some ways can be seen asCinderella processes in comparison to the other surface modificationtechniques, as it was around but often ignored. Though the begining ofnitriding can be traced to Adolph Machlet who applied for a patent in

x

FM_Nitriding.qxd 10/6/03 2:04 PM Page x



1908 followed by Dr. Adolph Fry’s patent in 1922, it was considered inmany areas as “a new process” when Drs. Wehnheldt and Berghausdeveloped the ion nitride process in 1932. Many fine metallurgists havecontributed to the growth in our knowledge of the subject of nitriding. Ihave attempted to bring together all of their work by capturing the accu-mulated knowledge about the process in a book that will contribute to thecontinued growth of the nitriding process. The bell has not struck mid-night on the subject of nitriding, however. Its technology, the refinementof its control, and most of all, our understanding of the process and meth-ods will continue to grow. I expect the technology will be adapted forapplication to nonferrous metals, such as aluminum and the refractorymetals.

This book is intended to assist all of the practitioners of the technologyin the day-to-day process operation of nitriding and ferritic nitrocarburiz-ing. The contents are based upon my lifetime of experience and theknowledge gained from my peers. I hope that you, the reader, will gainsome useful knowledge about the subject of nitriding and its derivativeprocess techniques.

The chapters in this book address many important questions related tothe nitriding process:

• Of the many nitriding methods, which one is for you? There are manydifferent and valid reasons for choosing each relevant nitridingprocess technique, be it the reduction in thickness of the compoundzone, 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 manufac-ture 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 notremoved?

• 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 per-sonal experience (“what works for you”) and possibly by the age oldmethod of trial and error.

This book is a practical approach to the subject and not an academic orscientific work, although the subject is of a scientific nature. As with anyother surface treatment process, such as carburizing or carbonitriding, theprocess of nitriding draws on many disciplines such as physics, chemistry,mechanical engineering, and electrical engineering. The “art” of nitriding

xi

FM_Nitriding.qxd 10/6/03 2:04 PM Page xi



also requires individual ingenuity and dogged determination, temperedwith patience to accomplish the process and produce a metallurgicallysound part. The process cannot be guessed. It can offer many metallurgi-cal benefits, but it needs to be managed and controlled in order to producethe desired and acceptable surface metallurgy. While it is a very simpleprocess, there is little widespread understanding of it. This book is myattempt 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 tech-niques, and failure evaluations have been both seen and experienced bymyself and shared by others. My years as a consultant have exposed me tomany industrial problems and process techniques. All of these experiencescontributed significantly to this publication.

There are many, many steels available to the engineer who designs thepart and chooses the steel. The problem that the engineer is faced with is“How do I source my information on steel and metallurgical processingtechniques?” The simple answer (which may sound glib) is “With greatdifficulty.” It is hoped that this publication will provide the reader, whomight be a heat treater, metallurgist, or design engineer, with a clearerinsight 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 influ-ence process costs related to nitriding. Nine-tenths of an iceberg is belowthe surface of the water. The top one-tenth could be likened to represent-ing both the material cost and the process cost. However, inappropriatematerial selection and a lack of understanding of process techniques (andtheir results) can greatly inflate these costs. The submerged nine-tenths ofthis iceberg is the labor cost, machining costs, equipment costs, and timethat is lost if the part does not function properly and must be scrappedbecause of improper heat treatment procedures and material selection.

The combination of steel selection and choice of heat treatment cantherefore “make or break the product.” This book is offered to help you“make the product” and not “break the product.”

David PyePye Metallurgical Consulting Inc.Meadville, PA

xii

FM_Nitriding.qxd 10/6/03 2:04 PM Page 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]

In Europe

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

In Japan Neutrino Inc. 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.

CHAPTER 1An Introduction to

Nitriding

THE NITRIDING PROCESS, first developed in the early 1900s, con-tinues to play an important role in many industrial applications. Alongwith the derivative nitrocarburizing process, nitriding often is used in themanufacture of aircraft, bearings, automotive components, textile machin-ery, and turbine generation systems. Though wrapped in a bit of “alchemi-cal mystery,” it remains the simplest of the case hardening techniques.

The secret of the nitriding process is that it does not require a phasechange from ferrite to austenite, nor does it require a further change fromaustenite to martensite. In other words, the steel remains in the ferrite phase(or cementite, depending on alloy composition) during the complete proce-dure. This means that the molecular structure of the ferrite (body-centeredcubic, or bcc, lattice) does not change its configuration or grow into theface-centered cubic (fcc) lattice characteristic of austenite, as occurs inmore conventional methods such as carburizing. Furthermore, becauseonly free cooling takes place, rather than rapid cooling or quenching, nosubsequent transformation from austenite to martensite occurs. Again,there is no molecular size change and, more importantly, no dimensionalchange, only slight growth due to the volumetric change of the steel sur-face caused by the nitrogen diffusion. What can (and does) produce distor-tion are the induced surface stresses being released by the heat of theprocess, causing movement in the form of twisting and bending.

Metallurgical Considerations and Process Requirements

Nitriding is a ferritic thermochemical method of diffusing nascentnitrogen into the surface of steels and cast irons. This diffusion process isbased on the solubility of nitrogen in iron, as shown in the iron-nitrogenequilibrium diagram (Fig. 1).

01_Nitriding.qxd 9/30/03 9:58 AM Page 1

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

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

The solubility limit of nitrogen in iron is temperature dependent, and at450 °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 pre-dominantly epsilon (ε) phase. This is strongly influenced by the carboncontent of the steel; the greater the carbon content, the more potential forthe ε phase to form. As the temperature is further increased to the gammaprime (γ ′) phase temperature at 490 °C (914 °F), the “window” or limit ofsolubility begins to decrease at a temperature of approximately 680 °C(1256 °F). The equilibrium diagram shows that control of the nitrogen dif-fusion is critical to process success (Fig. 1).

A number of operating process parameters must be adhered to andcontrolled in order to successfully carry out the nitriding process. Mostof these parameters can be controlled with relatively simple instrumenta-tion and methods. Examples of process parameters for gas nitridinginclude:

• 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 theexception of induced residual stresses. Another benefit of nitriding is thatit acts as a stabilizing process by providing an additional temper to theprocessed steel.

2 / Practical Nitriding and Ferritic Nitrocarburizing

Iron-nitrogen equilibrium diagram. The δ-phase, not shown on this dia-gram, exists from 11.0 to 11.35% N at temperatures below approxi-

mately 500 °C (930 °F).

Fig. 1


Control of the process parameters is necessary to ensure formation ofan acceptable metallurgical case. Without control, repeatability of themetallurgical requirements cannot be guaranteed.

The process control factors are those elements that will ensure a con-trolled 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, includ-

ing 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 met-allurgical engineer for the American Gas Company in Elizabeth, NJ. Herecognized that the surface hardening technique of carburizing led to dis-tortion problems due to extended periods at elevated temperatures, fol-lowed by severe quenching into either water or oil.

Through experimentation, Machlet soon discovered that nitrogen wasvery soluble in iron. Nitrogen diffusion produced a relatively hard surfacein simple plain irons or low-alloy steels and significantly improved corro-sion resistance. This was accomplished without subjecting the steel to ele-vated temperatures and, more importantly, without cooling the steel rap-idly to achieve a hard wearing surface. It could now cool freely within theprocess chamber, while still under the protection of the nitrogen-basedatmosphere, 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 nas-cent nitrogen necessary for the process. It was not long before Machletrealized that he needed to control the decomposition accurately. He didthis by using hydrogen as a dilutant gas to reduce the amount of availablenascent nitrogen, thus controlling to some extent the formed case metal-lurgy. His reasoning behind the control of the process gas was recognitionof what is now known as the “white layer” or “compound zone.” Figure 2shows a simple construction of the nitrided case. It should be noted thatthis schematic is not to scale.

The first patent for the development of the nitriding process was appliedfor in March 1908 in Elizabeth, NJ. The patent was finally approved inJune 1913, some five years after the initial application. Machlet had beenworking for a number of years on the process prior to his patent application

Chapter 1: An Introduction to Nitriding / 3


and continued to develop both the new process and his understanding ofthe resulting process metallurgy. The patent was for “The Nitrogenizationof Iron and Steel in an Ammonia Gas Atmosphere into which an Excess ofHydrogen Has Been Introduced” (Ref 1).

Although Machlet’s development and patenting of the new nitridingprocedure was technologically important, his work remained largelyunrecognized and faded into obscurity. Even today, very few nitridingpractitioners know who he was and what he accomplished. Most metallur-gists who are familiar with the nitriding process know the work of theGerman researcher Adolph Fry, who is recognized as the “father of nitrid-ing.” While Fry’s work was more publicized and his methods were taughtat many fine metallurgical academic institutions, it was Machlet who firstpioneered 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 wasvery soluble in iron at an elevated temperature. He also recognized veryearly in his work that alloying elements strongly influenced metallurgi-cal and performance results. Fry first applied for his patent in 1921, threeyears after the First World War ended. His patent was granted in March1924 (Ref 2).

He used a technique similar to that of Machlet, where the nitrogensource had to be cracked by heat to liberate nitrogen for reaction and dif-fusion. Like Machlet, Fry used ammonia as the source gas, but he did notuse hydrogen as a dilutant gas. Thus was developed the single-stage gasnitriding process as it is known today.

Fry then investigated the effects of alloying elements on surface hard-ness. He discovered that the nitriding process produced a high surfacehardness only on steels containing chromium, molybdenum, aluminum,


Fig. 2 Schematic of a typical nitrided case structure

Compound zone,dual phase

Diffusion zone consistingof formed nitrides

Core material

Transition zone fromdiffusion zone to corematerial


vanadium, and tungsten, all of which form what are known as “stablenitrides.”

He also discovered the critical nature of process temperature in terms ofcase depth and surface metallurgy. Processing the steel at higher tempera-tures placed the surface at risk to form what is known today as “nitridenetworks” (a saturated solution of nitrogen in the immediate surface of theformed case).

Because steels with higher alloy contents were not readily available fornitriding, Fry became responsible for developing a group of steels forKrupp known as the “Nitralloy” group. These steels, specifically designedas nitriding steels, soon became internationally recognized. Even todaythe 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 groupof nitriding steels under the licensed guidance of Krupp Steels. Thesesteels were also marketed under the brand name of Nitralloy. The com-pany was Thomas Firth and John Brown Steelworks, more commonlyknown as Firth Brown Steels. The steels from Firth Brown were known asthe “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 com-positions). The En 41 series contained aluminum, which produced a muchhigher surface hardness after nitriding. Aluminum has a strong affinity fornitrogen, forming very hard aluminum nitrides that are quite stable inamounts up to 1.0% Al. Much above 1.0%, aluminum has no effect on theresultant nitriding hardness.

Differences Between the U.S. and German Processes. The princi-pal differences between the process developed in the United States andthat developed in Germany were that:

• The U.S. process used hydrogen as a dilutant gas to control the nitrid-ing potential of both the gas and steel, which in turn controlled thefinal surface metallurgy.

• The Germans manipulated the process through alloying and improvedon such aspects as core hardness and tensile strength.


Table 1 British standard nitriding steelsComposition, %

Designation(a) C Si Mn P Cr Mo Ni V Al

En 40 A 0.20–0.35 0.10–0.3 0.40–0.55 0.05 max 2.90–4.00 0.60–0.80 0.40 max ... ...En 40 B 0.20–0.30 0.10–0.35 0.40–0.65 0.05 max 2.90–3.50 0.40–0.70 0.40 max 0.10–0.30 ...En 40 C 0.30–0.50 0.10–0.35 0.40–0.80 0.05 max 2.90–3.50 0.70–1.20 0.40 max 0.10–0.30 ...En 41 A 0.25–0.35 0.10–0.35 0.65 max 0.05 max 1.40–1.80 0.10–0.25 0.40 max ... 0.90–1.30En 41 B 0.25–0.45 0.10–0.35 0.65 max 0.05 max 1.40–1.80 0.10–0.25 0.40 max ... 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


Machlet’s process was not widely accepted in the United States, as itwas perceived to have little if any commercial value to U.S. industry. Bycontrast, the Germans exploited Fry’s process in the early years followingWWI. The German process enjoyed great success throughout Europe inthe aircraft, textile, railroad, automotive, and machine tool industries.

During the mid to late 1920s, information about Fry’s success began tofilter through to American industrialists, prompting the Society of Manu-facturing Engineers (SME) to take a strong interest in the German devel-opments. 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 meetingthat Jeffries suggested to Fry that he attend the forthcoming annual SMEconference in Chicago and present a paper on the process techniques andapplications. Fry could not attend, so his friend and colleague PierreAubert made a presentation on his behalf. The presentation helped bringabout 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 parametersand the effects of alloying on the nitriding response of steels. Some of themore 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 aseries of investigations to evaluate the new nitriding process. The studieswere completed during a two-year period, which concluded with a presen-tation of their findings in 1928 (Ref 3). In general, the investigatory workfocused on process temperature. The temperatures selected ranged from540 to 650 °C (1000 to 1200 °F). The upper temperature was significantlylower than the temperatures employed by Machlet, which ranged from480 to 980 °C (900 to 1800 °F). McQuaid and Ketcham concluded thathigher nitriding temperatures had an effect on core hardness of alloy steelsbut little effect on the ability to nitride at those temperatures. They alsofound that higher process temperatures increased the risk of formingnitride networks, particularly at corners, due to the higher solubility ofnitrogen in iron. When present, nitride networks cause premature failureat the steel surface by cracking and exfoliation.

McQuaid and Ketcham began an exhaustive series of investigationsinto the new process of nitriding as developed by Machlet and Fry. Thestudies 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



• 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 steelswere much improved while undergoing salt spray tests and that practicallyany steel can be nitrided, including plain carbon steel and pure iron.

McQuaid and Ketcham were also the first early metallurgists to studythe white layer or compound zone. They concluded that the “white struc-ture” is composed of a nitride, either iron nitrides or a complex nitridelayer, involving both iron and alloying elements. A further conclusion wasthat the white layer or compound zone was extremely hard but very brittleand that the layer should be avoided if possible (though no specific guide-lines were offered). They also studied the effect of decarburization onnitrogen diffusion and the mechanical strength of the nitrided case. Theirresults showed that the steel to be nitrided should clearly be free of surfacedecarburization; otherwise, the nitrided surface will exfoliate and peelaway from the substrate. They concluded that rough machining or someother operation to ensure complete removal of any decarburized surfacelayer should be performed before carrying out any nitride operation.

Robert Sergeson was associated with the research laboratories of theCentral Alloy Steel Corporation in Canton, Ohio. He presented a paper inJuly 1929 that reviewed the work of Dr. Fry on steels containingchromium, aluminum, molybdenum, vanadium, and tungsten (Ref 4).

In unison with McQuaid and Ketcham, Sergeson concluded that processchemistry and process control in nitriding were much simpler than in car-burizing. He also reviewed the effect of reheating on the case after nitridingand found that, with increasing temperature, case hardness stability wasmuch better than for carburized and quenched alloy steel. He noted that thesurface hardness value for a chromium-aluminum steel began to decreaseat only 525 °C (1000 °F), and only slightly. He worked with many moresteels and compared the effect of temperature on both nitrided alloy steelsand carburized and quenched alloy steels, yielding similar results. Theprocess equipment that he used for his nitriding experiments was notunlike many modern gas nitriding furnaces, despite their improved materi-als of construction and computerized process control (Fig. 3).

Sergeson examined the effect of both temperature and process gas flowon alloy steels and found that if the ammonia gas flow rate was increasedat 510 °C (950 °F), little difference resulted in the immediate surfacehardness and case depth. He also found that as process temperatureincreased, case depth increased but surface hardness decreased.

His work covered alloy steels with chromium and aluminum and inves-tigated the effects of varying aluminum and nickel contents. He concludedthat nickel was not a nitride-forming element, but that it tended to retard



the nascent nitrogen diffusion if present in significant quantities. Moredetailed information on alloying effects can be found in Chapter 12,“Steels for Nitriding.”

V.O. Homerberg and J.P. Walsted. Professor Homerberg was an asso-ciate professor of metallurgy at the Massachusetts Institute of Technologyand consulted for the Ludlum Steel Company. Mr. Walsted was an instruc-tor at the same university at the time that they presented their findings onthe nitriding process (Ref 5). They studied the effects of temperature up to750 °C (1400 °F), with its resulting increase in case depth but reduction ofsurface hardness. In addition, they studied the effects of decarburizationon a nitrided surface and concluded that surfaces must be free of decarbur-ization prior to nitriding.

They reviewed Fry’s process technique and the decomposition ofammonia under heat. Once again, the equipment used for their experi-ments 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 technol-ogy, a persistent phenomenon was observed: an ever-present white layeron the nitrided steel surface. The white layer was identified as a multi-


Refractoryinsulation

Air circulatingfan

Fan drive motor

Door lift mechanism

Furnace door

Exhaust ammoniagas outlet tube

To atmosphereexhaust

Process deliverygas (ammonia)

Load preparationtable

Furnacethermocouple

Nitride processchamber

Process chamberthermocouple tube

Ammonia gasinlet tube

Fig. 3 Schematic of a simple ammonia gas nitriding furnace


phase compound layer of ε and γ ′ phases. Much recognition was given toDr. Carl F. Floe of the Massachusetts Institute of Technology, who notonly performed major research regarding identification of the layer and itscharacteristics, but also developed a process technique to reduce the layerthickness (Ref 6). Today that technique is known as the Floe process, orthe two-stage process.

The Floe process is carried out as two distinct events. The first portionof the cycle is accomplished as a normal nitriding cycle at a temperatureof 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 producethe nitrogen-rich compound at the surface. Once the cycle is complete, thefurnace temperature is increased to approximately 560 °C (1030 °F), withgas dissociation increased to 75 to 85% (i.e., an atmosphere that contains15 to 25% ammonia). Very careful gas flow control of the ammonia and itsdissociation must be maintained during the second stage of the process.The two-stage process is used to reduce formation of the compound zoneonly; 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 theuse of molten salt as a nitrogen source. The salt bath process uses the prin-ciple of the decomposition of cyanide to cyanate and the liberation ofnitrogen within the salt for diffusion into the steel surface. Salt bath nitrid-ing is described in greater detail in Chapter 6.

The ion, or plasma, nitriding process, which is based on the familiarchemistry of gas nitriding, uses a plasma discharge of reaction gases bothto heat the steel surface and to supply nitrogen ions for nitriding (seeChapter 8 for details). The process dates back to the work of a Germanphysicist, Dr. Wehnheldt, who in 1932 developed what he called the“glow discharge” method of nitriding. Wehnheldt encountered severeproblems with the control of the glow discharge. He then partnered with aSwiss physicist and entrepreneur Dr. Bernhard Berghaus. Together theystabilized the process and later formed the company Klockner IonenGmbH, specializing in the manufacture of ion nitriding equipment.Although the ion nitriding process developed by Wehnheldt and Berghauswas used successfully by German industrialists during World War II, itwas not used extensively because it was considered too complex, tooexpensive, and too unreliable to guarantee consistent and repeatableresults. Not until the 1970s did the process gain industrial acceptance, par-ticularly in Europe.

The significance of the glow discharge process was that it did not relyon the decomposition or cracking of a gas to liberate nascent nitrogen onthe steel surface. The process was based on the ionization of a singlemolecular gas, which is nitrogen, and the liberation of nitrogen ions. Theprocess offered a shorter cycle time due to the steel surface preparationand the gas ionization. Nitriding was not now restricted to steels that



required specific nitride-forming elements. Today ion nitriding is carriedout on virtually all steels and cast irons as well as refractory metals, alu-minum (not yet on a commercial basis), and sintered ferrous materials. Aschematic of an ion nitriding furnace layout is shown in Fig. 4.

Ion Nitriding Production in the U.S. After WWII and through the late1950s, the General Electric Company of Lynn, Massachusetts, operated alaboratory known as the Electromechanical Engineering and PhysicsUnit. Dr. Claude Jones and Dr. Derek Sturges, along with Stuart Martin,developed the first ion nitriding unit in the United States and applied theprocess to a variety of materials and parts. Their ion nitriding units metall normal nitriding standards and were accepted by the U.S. Navy. Sum-maries of the properties, applications, and advantages associated with theprocess were published in 1964 (Ref 7) and 1973 (Ref 8).

Other Uses of Plasma Technology. Ion nitriding is not the only heattreatment process that utilizes the glow discharge phenomenon. Simplyput, if one uses the appropriate process gases and the proper furnace, theglow discharge technique can be applied to plasma-assisted ferritic nitro-carburizing, 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 ofASM Handbook.

Plasma technology is not new. One has only to observe the NorthernLights to witness a natural plasma. Lightning is also a natural plasma.


Simple schematic of the layout of an early plasma (ion) nitriding fur-nace system

Fig. 4

Vacuum processvessel

Work piece

Process gasmanifold(nitrogen,

argon)

Vacuumpump

Powersource

+

–


Neon signs are a plasma glow, and “lumina storms” found in gift shops arejust 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 mechanicalrequirements. Nitriding often is applied to low-alloy steels to “harden” thesteel and improve corrosion resistance. In addition to conventional nitrid-ing, the following processes have been developed:

• Oxynitride process, during which a controlled postoxidation treatmentis carried out to further enhance the surface corrosion resistance

• Ferritic nitrocarburizing (a controlled process using nitrogen and car-bon 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 some-times misunderstood, process. Both the gas and salt systems have run analmost parallel course since the early part of the 20th century. The processhas found its place in both low- and high-tech applications and is becom-ing better understood by process technicians, metallurgists, applicationsengineers, furnace designers, and academics.

Many developments in process techniques are being driven by environ-mental concerns and legislation. This has resulted in the introduction ofmore effective, efficient, and economical methods and equipment.Improvements can be seen in the development of gaseous methods, saltbath methods, fluidized-bed methods, and plasma processing techniques.

REFERENCES1. A. Machlet, U.S. Patent 1,092,925, 24 June 19132. A. Fry, U.S. Patent 1,487,554, 18 March 19243. H.W. McQuaid and W.J. Ketcham, Some Practical Aspects of the

Nitriding Process, reprinted from Trans. ASST, Vol 14, 1928, SourceBook on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., AmericanSociety for Metals, 1977, p 1–25

4. R. Sergeson, Investigation in Nitriding, reprinted from AmericanSociety for Steel Treaters (ASST) Nitriding Symposium, 1929, inSource 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. Unter-weiser and A.G. Gray, Ed., American Society for Metals, 1977, p 56–99



6. C.F. Floe, A Study of the Nitriding Process Effect of Ammonia Disso-ciation on Case Depth and Structure, reprinted from Trans. ASM, Vol32, 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 GlowDischarge, Met. Prog., Feb 1964, p 94–98

8. C.K. Jones, D.J. Sturges, and S.W. Martin, Glow-Discharge Nitridingin Production, reprinted from Met. Prog., Dec 1973, Source Book onNitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society forMetals, 1977, p 186–187



CHAPTER 2Why Nitride?

THE UNIQUE ADVANTAGES of the nitriding process were recognizedby 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 Statesbecause of its moderate hardness values for plain carbon, cast iron, andlow-alloy steels. The very long nitriding process cycle times required toreach the same case depths achieved by more conventional methods suchas carburizing were considered a disadvantage. For example, to achieve acase hardening depth of 1.0 mm (0.040 in.), the nitriding process requires90 h, compared to 4.5 h for carburizing (Fig. 1, 2).

As described in Chapter 1, the patent for gas nitriding was first appliedfor by Adolph Machlet and was for the nitrogenization of iron and steel in

Quench

Total case depthrequired = 1.0 mm (0.040 in.)

4.5 h1700

1475

927

800

Time

Tem

pera

ture

,°F

Tem

pera

ture

,°C

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

Fig. 1




an ammonia gas atmosphere diluted by hydrogen. The hardness results thathe achieved were not high by the standards of today or even by the Germanstandards of the day. Hardness results generally measure the success of theprocess 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 30to 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 partbeing treated. What was not recognized was the excellent corrosion resist-ance 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 synonymouswith the nitriding process (and which exhibited considerably higher hard-ness values than those obtained by Machlet). German industrialists beganto control the steel analysis for nitridable steels, licensing it to varioussteelmakers in other countries. The attitude toward the use of the processin America began to change after the 1927 presentation of Fry’s paper atthe SME conference in Chicago.

Key Process Considerations

Several factors helped nitriding to gain acceptance:

• Compared to other case hardening methods, nitriding is a relativelylow-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 relativelylow temperature compared to the more widely recognized surface treatmentmethods. The temperature employed is in the region of 500 °C (925 °F),


Free cool

Total case depthrequired = 1.0 mm (0.040 in.)

90 h

925 495

Time

Tem

pera

ture

,°F

Tem

pera

ture

,°C

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

Fig. 2


although it varies according to the steel being treated. Temperature selec-tion is based on the final tempering temperature of the steel during its pre-heat-treatment procedure. Should the nitriding temperature be at or abovethe tempering temperature, then the core hardness of the pretreated steelwill diminish according to time and temperature. It is most important tomaintain the core hardness of the pretreated steel to provide adequate sup-port to the diffused case. Without good core support, the case can fail.

Surface treatment processes that require higher temperatures than thenitriding 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 sur-face hardening techniques, along with case depth characteristics. Casedepths accomplished by carburizing are usually considerably deeper thanthose accomplished with nitriding. Some carburized case depths can be up

Chapter 2: Why Nitride? / 15

Quench

2.5 h

1475

1700

800

925

Time

Tem

pera

ture

,°F

Tem

pera

ture

,°C

2 h

1475

1600

800

870

Time

Tem

pera

ture

,°F

Tem

pera

ture

,°C

Processing time-temperature plots. (a) Illustration of carburizing to pro-duce 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

(a)

(b)


to approximately 6.35 mm (0.250 in.) (requiring an extremely long cycleat the process temperature). The downside of the deep case, of course, isthe occurrence of grain growth, which is inevitable due to the extendedcycle time at elevated temperatures for the carburizing procedure. Nitrid-ing takes place in the ferrite region on the iron-carbon equilibrium dia-gram (Fig. 5), which means that grain size in both the surface and the coreis not affected. However, the same deep case achieved by carburizing can-not 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 tem-peratures and will readily diffuse. The higher the temperature to which thesteel is elevated, the faster and deeper the nitrogen diffusion occurs. How-ever, 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 ofnitrogen in iron will occur, which can cause the problem known as nitridenetworking. This condition will embrittle sharp corners, leading to spallingor even exfoliation.

No Quench Requirement. With conventional surface diffusion processtechniques such as carburizing and carbonitriding, the steel must be at asuitable austenitizing temperature (depending on the steel composition)after carburizing, followed by quenching (or rapid cooling) to transform theaustenite phase (face-centered cubic, or fcc, lattice) into martensite (body-centered tetragonal, or bct, lattice) (Fig. 6). One should also be cognizant of


Diffuses carboninto the steel

surfaceProcess

temperatures1600-1950°F(870-1065°C)Case depth:

medium

Pack Gas Salt

Carburize

Ion

Diffuses carbonand nitrogen intothe steel surface

Processtemperature1550-1650°F(845-900°C)Case depth:

shallow

Diffuses boron intothe steel surface

Processtemperatures 1400-2000°F(760-1095°C)Case depth:

shallow

Diffuses nitrogeninto the steel

surfaceProcess

temperatures 600-1020°F(315-550°C)Case depth:

shallow

Diffuses carbonnitrogen, sulfur,

oxygen(individually

or combined) intothe steel surface

Processtemperatures1050-1300°F(565-705°C)Case depth:

shallow

Pack Gas Salt

Nitride

IonGas Salt

Cabontride

Ion Gas Salt

Ferritic nitrocarburize

Thermochemical diffusion techniques

Ion Pack

Boronize

Gas

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

Fig. 4



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 7.00

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800 3270

3090

2910

2730

2550

2370

2190

2010

1830

1650

1470

1290

1110

930

750

570

390

210

30

Carbon, wt%

Tem

pera

ture

,°C

Tem

pera

ture

,°F

1495°C1538°C

1394°C

1227°C

6.69%

2.08%

2.11%

0.68%

0.77%

4.30%

4.26%1154°C

1148°C

912°C

(727°C)

770°C

(α-Fe)Ferrite

(δ-Fe)

738°C

Liquid

Austenite+

cementite

Cementite(Fe3C)

A3 Acm

A1

Ferrite+

cementite

Solubility ofgraphite inliquid iron

(γ-Fe)Austenite

Additional carbondissolves intostructure

Heating to hightemperature

Face centeredcubic structure

Rapidcooling

Room temperaturebody centered cubic structure Room temperature

body centered tetragonal structure

Iron atoms

Carbon atoms

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

Fig. 5

Crystal lattice changes that take place during high-temperature heattreatment processes such as carburizing. Ferrite is bcc structure;

austenite, fcc; martensite, bct. Source: Ref 2

Fig. 6


Austeniteand

phasechange

Quench(thermal shock)

Martensitestart line

Phase change(stress induced)

Time

Tem

pera

ture

Hea

t up

(str

ess

relie

ve)

the size of the lattice structure between the two phases of austenite andmartensite. Remember that austenite is an enforced condition in an alloycarburizing steel, as is martensite. Martensite cannot be created without thetransformation from austenite by rapid cooling. Also, in order to formmartensite it is necessary to have carbon present in sufficient amounts whenthe steel is rapidly cooled (Fig. 7). A rapid cooling rate is unnecessary fornitrided steel after completion of the process cycle. The process chamberfree cools under the ammonia atmosphere down to a suitable temperature ofapproximately 200 °C (400 °F) and then is purged with clean, dry nitrogen.Unlike the ferritic nitrocarburizing, carbonitriding, and carburizing proce-dures, the nitriding process does not involve a critical cooling rate.

Minimal Distortion. Distortion is one of the most persistent problemsfacing heat treaters and engineers. Distortion manifests itself during thefinal heat treatment process in the form of either:

• Shape distortion: a change in geometrical form, such as curving,twisting, or bending

• Size distortion: a change in workpiece volume due to either growth orshrinkage


Austenite(fcc)

14 atoms

Quench

Martensitestart

Ferrite (bcc)9 atoms

Martensite (bct)9 atoms

Time

Tem

pera

ture

Schematic illustration of the phase transformation taking place whenhardening steel with sufficient carbon present. Crystal lattice: bcc,

body-centered cubic; fcc, face-centered cubic; bct, body-centered tetragonal

Fig. 7

Fig. 8 Phase changes in relation to thermally induced stress on quenching


Distortion can be minimized but not eliminated. Stresses are induced intothe steel through rolling and forging as well as machining procedures. Theseinduced 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 bekept to a minimum, then it is necessary to introduce an interim stress-relieving procedure after the preheat treatment and rough machining andbefore the final machining and nitriding process. The stress relief processwill act as a stabilizing process on the final nitriding process. Anothermethod of stabilizing is to cryogenically treat the steel (particularly analloy or tool steel). This will transform any retained austenite into untem-pered martensite that will be tempered by the nitriding process, ensuringbetter dimensional stability during and after nitriding (Fig. 9). If the heattreater exercises careful control over the heating rate during the heat-upcycle, stress relieving can be minimized.

The reverse side of the coin becomes apparent after the appropriatesoaking period has been completed and it becomes necessary to cool thesteel rapidly, as during hardening of either a through-hardened steel or acarburized or carbonitrided steel. Thermal stress patterns are then inducedinto the steel due to geometric section changes, differential cooling rates,and phase changes in the steel.


Nitride

(No phase change)(Additional temper)

(Dimensionally stabilized ifretained austenite is present)

Time

Tem

pera

ture

Hea

t up

(stre

ssre

lieve

) (nophase

change)C

ool down

Fig. 9 Benefits of the nitriding process

Before nitriding After nitriding

Fig. 10 Effect of growth due to nitriding


In addition, when another element or combination of elements is addedto create changes in surface chemistry, volume changes will occur in theform of size change and uniform growth. For example, a ring will grow interms of outside and inside diameters, the net effect being that the borediameter decreases (Fig. 10).

Nitriding is a relatively low-temperature process and produces a shal-lower case than does carburizing. Most importantly, no quench isinvolved. On completion of the process, the steel can be cooled down nat-urally under its atmosphere or force-cooled using clean, dry nitrogen.Generally the steel is allowed to cool naturally, thus reducing the risk ofthermal gradients due to sectional changes.

If the engineer can live with extended cycle times, the nitriding processcan offer many distinct advantages:

• Reduced distortion • Improved part cleanliness • Reduced final machining time • Improved dimensional stability

If one considers the cost of scrap due to distortion and time spentstraightening, nitriding is not an expensive process compared to carburiz-ing. 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 simplybecause there is no opportunity for retained austenite to form, as can occurin carburizing and quenching. Over time, retained austenite will decom-pose to untempered martensite. In carburizing, retained austenite willleave mixed phases in the formed case, which can lead to dimensionalinstability. 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 amountof 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 processcan only be applied to special steels that contain specific alloying ele-



ments with an affinity for nitrogen and which form stable nitrides. Asurface-treated steel often is considered inadequate if a high hardnessvalue has not been achieved.

Pure iron and low-alloy steels will nitride. However, they will exhibitmaximum hardness values of about 35 HRC (by gas nitriding). In theearly days of process development, Fry developed the Nitralloy group ofspecial alloy steels that produce high hardness values after nitriding. TheNitralloy steels contain alloying elements such as chromium, molybde-num, vanadium, tungsten, and aluminum.

Because these steels tended to be more expensive than the more con-ventional case hardening steels, reluctance toward their use quickly devel-oped. This attitude began to dissipate once larger lots of these steels wereproduced and their benefits (minimal growth and distortion, high hard-ness) were recognized.

The high hardness values achieved in the Nitralloy steels are due to theaffinity of the alloying elements to form stable nitrides at designatedprocess temperatures. The resulting hardness value is a function of theamount of these elements present. Considerably higher hardness valuesare 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 thatcontain 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 tra-ditional case hardening techniques, nitrided steels offer improved corro-sion and oxidation resistance. The nitrided surface of an alloy steel or toolexhibits increased resistance to saltwater corrosion, moisture, and water.However, this does not apply to stainless steels; in fact, their resistance tocorrosion will be reduced. This is because chromium has an affinity foroxygen, readily forming chrome oxide on the stainless steel surface. Thechrome oxide acts as a barrier to nitriding. For diffusion to occur, the sur-face must be passivated, thus reducing corrosion resistance.

Although nitriding improves the corrosion properties of alloy steels, theimprovement is not permanent. Surface degradation or pitting will even-tually occur, albeit not as rapidly as might occur had the steel not beennitrided.

The core properties of nitrided alloy steels usually do not change, butthis does not apply to lower-alloy steels. Nitriding normally is performedat a temperature below that of the final tempering temperature of the steelafter the prehardening sequence.

Care must be taken during the prehardening and tempering procedure tominimize surface decarburization; any surface decarburization must beremoved prior to nitriding. It is usually good practice to have an adequatecore hardness value in order to support the final nitrided case and a tem-pered martensite core.



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

Concluding Remarks

The nitriding process was first put to commercial use in the automotiveindustry on transmission gears, particularly spiral bevel pinions and spiralbevel gears because of their natural tendency to try to “straighten” at theheat treatment process temperature. These types of gears were notoriousfor distortion on the gear teeth. Current spiral bevel machining designpractice is to cut distortion into the gear. This technique requires an inti-mate knowledge of the steel combined with extremely good control of theheat treatment process.

Nitriding has become more widely accepted because of its ability toserve many applications that previously were not considered possible oreven worthwhile. Engineers and metallurgists use nitriding creatively,making it a viable and commercially acceptable process. As our under-standing of the benefits associated with the process evolves, nitriding willcontinue to grow in use and popularity.

REFERENCES1. D. Pye, Diffusion Surface Treatment Techniques: A Review, Ind.

Heat., March 2001, p 39–442. K.G. Budinski, Diffusion Processes, Surface Engineering for Wear

Resistance, Prentice Hall, 1988, p 78–119



CHAPTER 3How Does the

Nitriding Process Work?

SEVERAL PROCESS PARAMETERS must be considered in order toensure 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 thedecomposition (or dissociation) of ammonia gas supplied via an externalbulk storage system or individual bottles connected to a manifold (Fig. 1).

Ammonia gas begins to decompose when heat is applied, usually froman external source within the furnace. At the usual nitriding temperaturesof 500 to 570 °C (930 to 1060 °F), ammonia is in an unstable thermo-dynamic state and decomposes in the following manner:

2NH3 ↔ 2N + 3H2 (Eq 1)

D

AB 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.

Fig. 1




Typically, three reactions take place at the steel surface when the steel isat the set process temperature:

NH3 → 3H + N (Eq 2)

2N → N2 (Eq 3)

2H → H2 (Eq 4)

The atomic nitrogen and hydrogen components shown in Eq 2 are unstableand will unite with other like atoms to form molecules as shown in Eq 3and 4. It is while they are in the atomic state that diffusion takes place.

The released nitrogen diffuses into the steel at the nitriding processingtemperature, but very slowly, to the point where it is not economicallypractical or effective. The temperature of 500 °C (930 °F) is considered tobe an “economical” temperature. Nascent nitrogen has an affinity for steeland iron and will readily diffuse into both materials at elevated tempera-tures. The higher the temperature, the faster and deeper the nitrogen diffu-sion. An economical temperature is one that produces an optimum casedepth 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 theinstant of decomposition, the liberated nitrogen will exist as nascent oratomic nitrogen and as such can be absorbed by the steel.

Nitrogen has an atomic diameter of 0.142 nm and is dissolved in iron ininterstitial positions in octahedral voids of the cubic lattice that have amaximum diameter of 0.038 nm in bcc alpha (α) iron and a maximumdiameter of 0.104 nm in fcc gamma (γ) iron. Nitriding of pure iron at tem-peratures 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 roomtemperature 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 innitriding practice.

The process of the smaller nitrogen atoms passing between the iron-base crystals as heat is applied up to a suitable process temperature isknown as “interstitial diffusion.” This process is shown schematically inFig. 2.



Dissociation of the Gas at the Selected Nitriding Temperature (Ref 1)

At temperatures around 500 °C (930 °F), the stability of ammonia isquestionable, leading to dissociation rates higher than 98% and thus toformation of a protective gas without any nitriding effect. Despite the ben-eficial catalytic effect of the workpiece surfaces and the furnace wall, thedissociation of ammonia is an extremely slow process. Therefore, ammo-nia-based nitriding atmospheres for steel treatment rarely contain less than20% ammonia and frequently up to 50%, and thus their degree of dissoci-ation is far from equilibrium. The remaining ammonia content is decisivefor the effect of nitriding, during which nitrogen diffuses into the steelaccording to the reaction (Ref 2):

NH3 → N(α) + (3/2)H2 (Eq 5)

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

aN = KN

pNH3 (Eq 6)p3/2

H

where KN is the equilibrium constant of nitrogen, and pNH3and p3/2

H are thepartial pressures of ammonia and hydrogen, respectively.

The nitrogen transfer is comparatively low, and the release of hydrogenfrom the ammonia molecules determines the process rate. Therefore, asmentioned earlier, nitriding times are rather long, up to 120 h.

Chapter 3: How Does the Nitriding Process Work? / 25

Schematic of interstitial diffusion during the nitriding process. The workis 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 apart of the furnace atmosphere.

Fig. 2


The nitriding effect of a nitriding atmosphere is defined by its degreeof dissociation, with high degrees of dissociation always indicating anear-equilibrium state where the nitriding effect is low. In processing, itis common to use a constant degree of dissociation (e.g., 30%), but some-times a two-stage procedure involving variations in temperature and dis-sociation degree (the Floe process, described in Chapter 1) is followed.However, the measured ammonia content is not equivalent to the actualdegree of dissociation. In ammonia dissociation according to Eq 1, twomolecules of ammonia dissociate to one molecule of nitrogen and threemolecules of hydrogen. This increase in volume dilutes the ammoniacontent, as do additional gases, and must be taken into account in deter-mining the actual degree of dissociation.

The nitriding effect can be determined more easily by means of thenitriding potential:

Np = pNH3 (Eq 7)p3/2

H2

The nitriding potential is generally used for describing the nitridability ofan ammonia atmosphere. Its control allows the formation of predictablecase structures and depths for a range of steels, including some tool andstainless grades (Ref 4). Higher Np values produce higher surface concen-trations of nitrogen and steeper concentration gradients. Lower potentialsallow 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 toapproximately 0.099 wt%. Thus, low solubility values of nitrogen in α-iron, combined with the high partial pressure values for gaseous nitrid-ing that are needed for nitrogen absorption, make use of bottled nitrogenimpossible. Gaseous molecular nitrogen will not dissociate into nascentnitrogen when cracked by heat at process temperature, and thus no diffu-sion of molecular nitrogen will occur. The steel surface, in conjunctionwith heat, will act as a catalyst for the decomposition of ammonia intonascent nitrogen for diffusion into the steel surface.

The carbon in the steel plays a part in the reaction because the carbon ispresent in solution with the iron in austenite. Nitrogen has a stabilizingeffect on austenite, which will transform very slowly. Thus when marten-site 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



nitriding is limited to the ferrite region on the iron-carbon diagram. Thebcc lattice structure remains unchanged. Growth, however, will be experi-enced due to an increase in the volume of the immediate steel surface andby the nitrogen diffusion (Fig. 3).

What this suggests is that the observable growth will be primarily tosurface volume rather than to the core, as no phase changes occur in eitherthe core or the immediate surface. Nascent nitrogen is being diffused atthe surface and dispersed into the immediate surface, increasing the sur-face volume.

The atomic nitrogen and hydrogen as described in Eq 1 is consideredunstable and will begin to unite with like atoms to form a molecule asshown in Eq 2 and 3. During the atomic state is when diffusion into thesteel begins to take place.

Because gaseous molecular nitrogen will not dissociate into nascentnitrogen when cracked by heat at process temperature, no diffusion ofmolecular nitrogen will occur. The steel surface, in conjunction with heat,will act as a catalyst for the decomposition of ammonia into nascent nitro-gen and for diffusion into the steel surface.

Provided that all the appropriate conditions are met, nascent nitrogenwill begin to diffuse into the steel surface at a rate determined by theprocess temperature. In other words, the process temperature can be aslow as 350 °C (600 °F) and will react with the alloying elements to beginto form the solid solution of stable nitrides. Conversely, the temperaturecan be as high as 600 °C (1050 °F). However, although the reaction of thenascent nitrogen will still occur, it will also strongly affect the thickness ofthe surface compound layer and give rise to the risk of nitride networking.

Distortion

Do not be misled into thinking that the nitriding process causes abso-lutely no distortion. It is a question of the definition of distortion. Distor-tion resulting from the nitriding process takes the form of uniform growth


Growth

Growth

Growth

Growth

Growth

Growth

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

Fig. 3


at the immediate surface. The amount of growth is determined by the fol-lowing 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 resultfrom:

• Mixed residual phases due to an excessively slow cooling rate on pre-hardening (presence of retained austenite)

• Residual stress patterns due to machining and the omission of inter-mediate stress relieving

• Stacking of components atop one another in the process chamber• Selection of the nitriding temperature when residual phases are pres-

ent. If the process temperature is higher than the tempering tempera-ture, 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 sup-port the nitrided case during its operation. If the core is not preheat treatedand is left in, for example, the annealed condition, nitriding will be verydifficult to accomplish. Even if it were possible to nitride an annealedsteel, the workpiece would not perform well within its operating environ-ment simply because a load on the case could exceed the core strength ofthe steel. This is one of the reasons why it is necessary to preheat treat thesteel (harden and temper to produce tempered martensite).

Annealed steel is in the ferritic condition; that is, the atomic structure ofthe ferrite phase is in the bcc lattice structure. On heating the steel for thehardening operation, the phase will change to austenite, which is a 14-atomfcc lattice configuration (see Fig. 6 and 7 in Chapter 2). On quenching forthe hardening operation, the lattice structure changes to that of martensite,which has reverted back to a 9-atom structure, but now in the bct configu-ration. 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 toreduce the core hardness back to the appropriate hardness value that willbest support the nitrided case during operation.



If during the hardening operation austenite remains with the untem-pered martensite, that retained austenite must be transformed to untem-pered martensite. This can be achieved during either the tempering or thenitriding operation. Remember, the nitriding process temperature isslightly lower than the primary tempering temperature of the steel. Thetempering procedure will decompose the retained austenite into a tem-pered martensite condition. In addition, the nitriding process temperaturewill also act as a stabilizing procedure for the steel and enhance dimen-sional stability.

What hardness will be the correct hardness? The answer requires deter-mining what type of loading will be placed on the case. The core hardnesswill also be determined by the steel chemistry, the austenitizing tempera-ture, and the final tempering temperature.

REFERENCES1. J. Grosch, Heat Treatment With Gaseous Atmospheres, Steel Heat

Treatment Handbook, G.E. Totten and M.A.W. Howes, Ed., MarcelDekker, 1997, p 663–719

2. C.A. Stickels and C.M. Mack, Overview of Carburizing Processes andModeling, Carburizing: Processing and Performance, G. Krauss, Ed.,ASM International, 1989, p 1–9

3. B. Edenhofer and H. Pfau, Self-Adaptive Carbon Profile Regulation inCarburizing, Heat Treatment and Surface Engineering: New Technol-ogy 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 Guess-work out of Nitriding, Advanced Materials & Processes, Dec 1994, p 52–54




CHAPTER 4Microstructures of

Nitrided Iron and Steel

FORMATION OF THE NITRIDED CASE begins through a series ofnucleated growth areas on the steel surface. These nucleating growth areaswill eventually become what is known as the “compound layer” or, morecommonly, the “white layer.” This layer is usually very hard and brittleand comprises two intermixed phases. The layer does not diffuse into thesteel, 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 nucleationprocess, thus causing a mixed phase formation at the steel surface, withthe diffused and reacted nitrides forming beneath the surface. The regionimmediately beneath the white layer is called the “diffusion zone.” Thisregion is made up of stable nitrides formed by the reaction of nitrogenwith nitride-forming elements. The area below the diffusion zone is thecore of the steel, which consists of tempered martensite. All threeregions—the white layer, diffusion zone, and core—are shown in Fig. 2.

The nitrides begin their formation by the nucleation of γ ′ at the immedi-ate steel surface interface with the nitriding atmosphere. This nucleation

Compound zone,dual phase

Diffusion zone consistingof formed nitrides

Core material

Transition zone fromdiffusion zone to corematerial

Fig. 1 Typical nitrided case




process progresses and continues until the subsequent nucleation of ε atthe steel surface interface (Fig. 3). Note that the nitrogen diffusion ismuch 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 surfacestructure. In a typical nitriding steel with a carbon content of approxi-mately 0.4 wt%, the formation of γ ′ to ε will be roughly equal using thegas 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 con-tent, the greater the γ ′-phase.

The thickness of the compound zone is a function of time, tempera-ture, and pressure (usually atmospheric pressure). The amount of carbonin the steel has a small effect on thickness. Again, however, carbon con-tent considerably affects the composition of the compound layer, deter-mining whether the layer will be predominantly γ ′, ε, or equal amountsof each phase. If the compound layer thickness is a critical issue, thenthe steel must be selected carefully to accomplish the required surfacemetallurgy.

Controlling Compound Zone Thickness

The thickness of the compound layer is controllable to a large extent bymanipulating the process technique. As discussed subsequently, it can becontrolled by dilution, the two-stage Floe process, or by ion nitriding.


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

Fig. 2


Whether the compound zone is detrimental to the formed case willdepend entirely on the application of the part and its working environ-ment. This can only be established by knowing how the compound zonewill perform under operating conditions. Is it advantageous or detrimentalto form an either γ ′-rich or ε-rich compound zone or to have equalamounts of each phase in the zone, or even to have only a minimal com-pound layer thickness? Compound layer formation is perhaps the mostcontroversial issue related to the nitriding process and has been the subjectof many debates, discussions, and research papers (Ref 3). See, for exam-ple, the bibliography published in conjunction with Ref 4, which includes33 references that describe the influence of compound layer thickness,phase composition, chemical composition, and microstructure on thebehavior of nitrided steels.

Dilution. When hydrogen is added to the flow of ammonia during gasnitriding, the nitrogen component of the ammonia is diluted. This reducesthe amount of available nascent nitrogen for diffusion into the steel surface.

Chapter 4: Microstructures of Nitrided Iron and Steel / 33

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

Fig. 3


Controlling the nitrogen availability allows control of the compound layerthickness 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 proce-dure: one step at the normal nitriding temperature (500 °C, or 930 °F) withnormal ammonia gas dissociation (between 15 and 30%), and a secondstep 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 muchthinner compound layer. The principle of the procedure is to reduce theamount of available nitrogen for surface diffusion and to ensure a rapiddiffusion by raising the process temperature. The temperature for the sec-ond stage must be carefully selected. Selection of higher process temper-atures poses the risk of grain-boundary networking with iron nitrides atthe periphery of the grain boundaries, leading to premature componentfailure at sharp corners (Fig. 4). When considering use of the two-stageprocess, it may be prudent not to employ the higher temperature. Thehigher temperature does not influence the final hardness value, only thediffusion rate.

Ion nitriding is gaining much recognition. The process relies on thecreation of a gaseous plasma under vacuum conditions. The process gasescan be selected in whatever ratio suits the required surface metallurgy. Inother words, the formation of the compound layer can be single phase,dual phase, or diffusion zone only. The surface metallurgy can be manipu-lated to suit both the application and the steel. Ion nitriding has opened thedoor to many applications that were not possible with conventional nitrid-ing techniques.


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

Fig. 4


What Happens Below the Compound Zone?

The compound layer will always be thicker on plain carbon steel thanon alloy steel for the same cycle time and temperature. However, the met-allurgy will be completely different in terms of both surface compoundlayer and diffusion zone. Beneath the compound zone at nitriding temper-ature, the nitrogen dissolves into the α-iron and also reacts with certainalloying elements—such as aluminum, molybdenum, chromium, tung-sten, vanadium, and silicon—if they are present.

All of these alloying elements form nitrides in steel. This area of nitrideformation below the compound zone is known as the diffusion zone (Fig. 1, 2). The nascent nitrogen will begin to react with those elementsthat will readily form stable nitrides. The nitrides exhibit very high hard-ness values, particularly the aluminum nitrides.

Can Plain Carbon Steel Be Nitrided?

Contrary to popular belief, plain carbon steel or even plain iron can benitrided (Fig. 5). However, the compound zone will be much thicker thanfor alloyed steel with equal carbon content. This is because the nitridesformed by the alloying elements will contain more nitrogen than thoseformed with iron, or iron in the plain carbon steel. In addition, the hard-ness values will be much lower. Plain carbon steels that have been pre-treated to a tempered martensite structure typically will exhibit hardnessvalues of 400 to 700 HV (49 to 60 HRC) under a 200 gf microhardnessload. Alloy steels can exhibit hardness values from 700 to 1000 HV (60 to69 HRC) under the same load. Surface hardness values for stainless steels,particularly when surface chemistry reactions are controlled (as in theplasma nitriding process), can reach 1500 HV (78 HRC).


The metallographic appearance of AISI 1015 (UNS G10150) steel aftera 2 h vacuum nitrocarburizing treatment in an ammonia/methane mix-

ture with 1% oxygen addition

Fig. 5


Only in a plain carbon steel (one without nitride-forming elements) willa supersaturated solution of α-iron occur. It is quite possible for the ironnitrides to segregate on slow cooling from the nitriding temperature or viaan aging process at a low-temperature set point.

Another often overlooked feature of the nitriding process (even on a plainiron or plain carbon steel) is enhancement of the steel surface corrosioncharacteristics. By contrast, the corrosion resistance of stainless steels willbe reduced, due to breakdown of the surface chrome oxide barrier to enablenitrogen 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 • Preheat temperature• Temperature • Pressure• Gas composition

A simplified Harris formula can be used as a rough guide for calculatingthe 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 nitridingchamber volume. If the nitrogen is added to the ammonia, thus potentiallyincreasing the nitriding potential of the ammonia, or if hydrogen is addedto the ammonia, then the nitriding potential is reduced and the gas turnoverrate 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 contami-nants. This means that the work surfaces should be “clinically” clean—acondition commonly achieved by vapor degreasing the surface after hard-ening and tempering and prior to nitriding. Surface contaminants, whichcan cause formation of a nonuniform, or “spotty,” case, include:

• Cutting fluids • Paint• Oils for surface protection • Decarburization• Fingerprints

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



decomposition will occur, causing formation of a prohibitive surface bar-rier that prevents nitrogen diffusion.

Oils should also be removed from the steel surface so that they do notdecompose and leave a carbon residue. The carbon will form a barrier atits point of deposition on the surface, thus resisting the nitriding effect andresulting 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 decom-pose, leaving behind a carbon residue on the steel surface. The materialhandler should wear lint-free cotton gloves, or other gloves that will notreact or leave a deposit on the steel surface.

Paint residue or even marking ink will set up a nitride-resistant barrieron the steel surface. Once again, absolute surface cleanliness is mandatory.

Decarburization is a surface problem. The presence of a decarburizedlayer will inhibit the uniform formation of a good nitrided surface. Ifdecarburization has occurred from mill rolling and insufficient machinecleaning, or if the preheat treatment operation was carried out in a decar-burizing atmosphere, then a nonuniform case will form. The result is usu-ally visible in the form of an “orange peel” effect on the steel surface, orthe case will exfoliate from the core steel.

Hot-rolled steel should be machined with at least 10% of the bar diame-ter or 10% of the thickness removed. If the steel is in the form of rolledplate, it is imperative that the machining be uniform. Nonuniform machin-ing will only contribute to the possibility of a partial decarburized layer onone side and will also induce a residual stress pattern that will relieveitself during the nitriding procedure.

REFERENCES1. 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, 19913. T. Bell, B.J. Birch, V. Korotchenko, and S.P. Evans, “Controlled

Nitriding in Ammonia-Hydrogen Mixtures,” presented as Heat Treat-ment 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, ASMInternational, 1992, p 878–883

5. C.E. Floe, A Study of the Nitriding Process Effect of Ammonia Disso-ciation on Case Depth and Structure, ASM Conference Transactions,1944




CHAPTER 5Furnace Equipment and

Control Systems

THE NITRIDING FURNACE is simple in design. Other than the mate-rials used, little has changed in furnace construction. Principal designchanges involve the control of process parameters, with developmentsmade in the use of computerized systems employing programmable logiccontrollers (PLC) for time, temperature, and gas flow control (Fig. 1).Good gas circulation is necessary to prevent process gas stagnation withinthe process chamber (also called the retort).

Typical PC/PLC screen configuration for nitriding furnace control.

Courtesy of Plateg USA, Meadville, PAFig. 1




Essential Furnace Design Criteria

A well-designed gas nitriding furnace provides accurate, uniform con-trol of process temperature, gas flow, and gas circulation. The furnaceshould be equipped with a good, simple mechanical handling system tomaneuver the process chamber in or out of the furnace.

Accurate Temperature ControlAs the steel is likely to be prehardened and tempered, the tempering

temperature would be higher than the nitriding process temperature. It isnecessary to maintain temperature uniformity throughout the furnaceheating chamber within at least ±5 °C (10 °F). That generally is accom-plished by an internal air-circulating fan, although the circulation systemcan be external to the process chamber. Remember, nitriding cycles areusually lengthy, especially for deep case depths. Large process tempera-ture fluctuations must not occur during the process, particularly anyupward swing in temperature. A sudden temperature increase will lead tothe possibility of reduced core hardness, as well as nitride networking(usually on corners) due to the higher solubility of nascent nitrogen athigher temperatures (Fig. 2).

Gas Circulation in the Process ChamberGood 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:

• Overpressure circulation• Internal fan circulation

Micrographs of white nitride layers developed on vacuum-melted AMS6470 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. Buildupof 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 hwith 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

Fig. 2(a) (b)



Overpressure circulation simply means ammonia gas flows in andout due to the pressure difference. This process relies on gas delivery pres-sure and, most importantly, on part geometry. Complex geometries cancause gas stagnation, leading to the possibility of nonuniform case forma-tion and sometimes even no case formation. Examples of complex geome-tries include bushes, sleeves, and cavities.

Good gas circulation is imperative to prevent the possibility of stagna-tion in blind holes, cavities, and shielded areas within the process chamber(i.e., areas where one part is shielding another part). If gas stagnationoccurs and nascent nitrogen is not replenished within the hole or cavity,then a shallow nitrided case may form. The overpressure method of gascirculation requires careful loading of components.

Internal fan circulation is unquestionably the best method for pre-venting gas stagnation (Fig. 3). Gas circulation by a fan nearly guaranteescomplete accessibility of the process gas throughout the process chamber.

Another benefit of the internal fan system is good temperature unifor-mity. 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 stresspattern in the nitrided component. Internal fan circulation is not necessar-ily restricted to the bell-type furnace shown in Fig. 3; a horizontal furnaceor even a pit-type furnace design can be used.

Chapter 5: Furnace Equipment and Control Systems / 41

Heating bell

Retort

Work basket

Cooling bell

Exhaust fan

Heating elements

Work support

Ammonia supply

Bell-type furnace with heating bell

Exhaust

Oil seal

Circulating fan

Bell-type furnace with cooling bell

Fig. 3 Schematic of bell-type furnace containing an internal fan for gas circulation


Types of Nitriding Furnaces

Any type of furnace can be used for the nitriding process, provided thatit has the integrity to maintain an atmosphere without leaking and meetthe required process control conditions. This means that even if a furnaceis 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 isolat-ing the process gas from the shop environment. This is accomplished byprocessing the work in a sealed (gastight) furnace or in a sealed processretort or chamber.

Insulation

A furnace that is repeatedly heated up and down through a temperaturerange both uses and loses energy. As much of the energy (heat) as possibleneeds to be contained within the process chamber, and external heat lossesneed to be kept to a safe working minimum temperature. Therefore, thefurnace 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 toinsulation and heating elements are discussed below. There is no idealinsulation material. The type of insulation depends on whether a processretort is used and how the retort goes into the furnace. Poor mechanicalhandling may damage the heating system and the insulation materials.

Relatively Low Temperatures. Because nitriding furnaces usually arenot subjected to high temperature levels, wear and tear on the insulationand furnace elements is very low. The heating elements are more oftenthan 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 forthe application.

Long Useful Life. The furnace is not subjected to adverse temperaturefluctuations, thus prolonging the life of the heating system and insulationmaterial. The author personally knows of a used vertical air-circulating



furnace that in 1963 was converted into a nitriding furnace, with newrefractory brick and heating elements added. The furnace is still in opera-tion 40 years later, with no further element or refractory replacement. Ifthe refractory lining and heating elements are subjected to mechanicaldamage, of course, then their life expectancy will be reduced. However,unless a nitriding furnace is subjected to poor mechanical handling, itrequires low maintenance.

Energy Costs. The quality and thickness of the insulation will deter-mine the energy cost of the cycle. The insulation design should take intoaccount heat losses to the outer furnace casing and the insulation thicknessand quality. Higher quality insulation may have a higher initial cost, butthat 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, orproduction work?

• Part loading: How large are the parts to be treated, and what is themaximum load that is to be charged into the process chamber? This isimportant, as it will determine the heat/energy requirements for thefurnace, chamber, and load.

• Frequency of operation: Will the furnace be operated on a jobbing basisor will there be guaranteed loads with regular frequency of loading?

• Furnace availability: Will the furnace be used for other operationssuch as tempering, subcritical annealing, or stress relieving?

Retort Construction

Steels. Heat-resistant type 309 or 310 stainless steel is quite suitable asa retort material. The most satisfactory material is Inconel; however, it isexpensive. The primary requirement is that the material must not catalyzeammonia dissociation at the internal surfaces of the process chamber.Materials that fall into this category are mild or plain carbon steel, andtypes 302 and 304 stainless steel.

If used for the process chamber, mild steels and plain carbon steels willdraw away the nitriding reaction from the workpiece surface to the pointthat minimal nitriding, if any, will take place. They also lack the necessarymechanical strength at the process operating temperature, quickly distort-ing and failing prematurely. Types 302 and 304 stainless steel also lackmechanical strength and will distort, as well as deteriorate rapidly on theinside due to corrosion resulting from the breakdown of surface chromeoxide caused by the nitriding reaction.

Temperature-resistant glass, in theory, is probably the best retort con-struction material because of its nonreactive nature with ammonia gas.



However, its fragility would result in mechanical damage, making glassimpractical for use in a production nitriding environment.

Insulating refractory firebrick can be used to line the process cham-ber, provided that it is dense and heat resistant, with low porosity and lowiron content. The mortared joints must be very thin and contain minimalporosity; otherwise, the ammonia process gas will begin to leak. If theiron content of the brick is high, the iron will react with the ammonia toform iron nitrides and begin to erode from the joint, particularly if there isa fan circulation system in the process chamber.

Enameling. The container can be enameled but, like temperature-resistant glass, is subject to chipping by mechanical damage and is not apractical solution for the nitriding process chamber.

Deoxidized copper can be used, but lacks strength at high tempera-tures. The copper will not take part in the process reaction, being com-pletely impervious to nascent nitrogen or anhydrous ammonia. Copper isused as a masking medium to prevent nitriding from taking place onselected portions of steel parts.

Retort Maintenance

Even if stainless steel or Inconel is used, the retort will not be mainte-nance free. No matter what the material of construction, the internal retortwalls require a periodic program of regeneration.

Regeneration of the retort is necessary when gas dissociation becomesdifficult to control and when workpieces exhibit inconsistent metallurgicalresults. This indicates that the free nitrogen is reacting with the internalwalls of the process chamber, due to deterioration of their chrome oxidesurfaces. Regneration is simply accomplished by heating the emptyprocess chamber to about 900 °C (1650 °F) and holding at that tempera-ture for a few hours. This procedure will “burn” out the nitrogen and causethe chrome to oxidize, thus forming chrome oxide. After the processchamber cools, the surfaces are cleaned by either shot-blasting or sand-blasting. The frequency of regeneration correlates with retort usage andwill vary from plant to plant. Once every 3 to 5 years is typical.

Trays and fixtures require the same considerations as the process retortin terms of construction materials and maintenance. Each time the nitrideoperation is run, the trays and fixtures are exposed to the atmosphere andthe process temperature. Therefore, a planned maintenance proceduremust 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 onthe retort flange. Mild steel or stainless steel studs, nuts, and washerssecure the sealed top cover. If mild steel fasteners are used, dip the thread



into a mixture of oil and flake graphite or molybdenum disulfide to ensureeasy removal after processing. Otherwise, the threads of both the stud andnut will seize solid by the end of the process cycle.

Other sealing methods include oil seals (commonly used on the lift-offbell furnace configuration), clamp arrangements, and vacuum seals. Greatcare must be taken to ensure that the process chamber is gastight.

Checking and Identifying Seal Leaks. The new seal will usually leakslightly on the first cycle without serious consequences. After that, the sealhas “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 moltensulfur. Once lit and manually moved around the flange perimeter, a cloud


Dissociator

AmmoniaTank

Heating Elements

Seal

Exhaust

Gas Outlet

Fan Motor

Nitriding Furnace

DissociationPipette

Seal

Gas Inlet

Nitriding Retort

WorkBasket

Fan

Fig. 4 Nitriding process retort with a seal that fits on the retort flange


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

Safety Precautions When Using Ammonia

Ammonia is an extremely pungent gas, which can make breathingimpossible if large volumes of ammonia escape. Even if small volumes ofammonia gas leak from the process chamber, people near the leak sourcemay experience loss of breath and extremely painful eye irritation. It isimperative to read the Material Safety Data Sheet or International Chemi-cal Safety Card when using ammonia gas.

Ammonia gas is also very flammable. If small leaks occur in the seal ofan internal process chamber, the ammonia can ignite and burn vigorouslyespecially in the presence of oxygen. If part of the sealing method is madefrom aluminum, then there is a strong possibility that the aluminum portionof the seal will melt, thus increasing the fire risk. Large concentrations ofanhydrous ammonia can cause serious injury, even death, due to violentexplosions if the leaking ammonia is exposed to a naked flame. Great caremust 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; undersome conditions it can produce both a toxic and violently explosivegas mixture.

• Ammonia gas that may have leaked from the process retort or gasstorage bottle can be violently explosive and flammable. Ensure ade-quate air ventilation around the nitriding furnace.

• Keep a rebreathing apparatus system in the process vicinity. If anhy-drous ammonia leaks it is violently pungent and can only beapproached while wearing a rebreathing apparatus.

• Avoid the ammonia gas coming into contact with the skin or sores, asthis will be very painful.

• Fit an afterburner system or scrubber or both on the exhaust line sothat the ammonia is decomposed before being dispersed into theatmosphere.

• Ensure that everyone concerned with the nitriding process knowswhere the main safety shutoff valve is located.

• Ensure that the safety shutoff valve is in position and can be turnedwith relative ease. If there are intermediate valves on the line and it isa key valve, make sure that the key is chained to the shutoff valve.

• If the process uses only the dissociation burette measuring system forprocess gas control, check city regulations regarding the disposal ofwater contaminated by ammonia. Do not put the ammonia water downa city drain.



• If a leak does occur on an exposed delivery pipe, remember thatwater will absorb 70 times its own volume of ammonia. In otherwords, ammonia is very soluble in water. Simply wrap soaking wettowels or rags over the pipe to absorb the ammonia until professionalhelp is available. Ammonia is also a very dense gas and will settle to the lowest point. The area can simply be sprayed with a waterhose; however, advise city authorities if the ammonia water goes intoa drain.

• If a bad ammonia leak occurs, evacuate the area immediately, advisethe appropriate authorities, and then ventilate the area thoroughly.

• Clearly mark all exit paths and exit doors. Be sure that all personnelunderstand completely how to evacuate the affected work area.

• A safety shower and eyewash station should be close to the furnaceoperating area. If a person comes in contact with large volumes ofammonia gas, it is necessary to flood the body with an excess amountof water. If no safety shower is available, then wash the affected partwith large volumes of water from the nearest source. The same appliesto eye contact with ammonia; the eyes must be flushed with clean,uncontaminated water. If no eyewash station is available, then use asqueeze bottle filled with water to flush the eyes.

• Make sure that all personnel know how to contact both a responsibleperson and, if necessary, a physician. Maintain up-to-date telephonenumbers and post them in an accessible and convenient place.

• Do not, under any circumstances, try to neutralize the ammonia leakor spill with an acid. A violent reaction can occur.

• Most importantly, read the Material Safety Data sheets. It is in yourown interest and that of everyone else who works with you to befamiliar with the safety aspects of handling ammonia gas.

Caution Regarding Safety When Using Ammonia. This section isintended only to alert the reader to the dangers of ammonia. It does notclaim to address all possible safety concerns. The user, in conjunctionwith equipment and material suppliers and appropriate government agen-cies, must develop, institute, and maintain health and safety practices suf-ficient and in compliance with regulatory requirements.

Furnace Heating

A furnace can be heated electrically or by natural gas. The choice ofheating medium depends on:

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



Electric heating elements usually are made of a nickel-chromium(Nichrome) material and are not normally exposed to the ammonia atmo-sphere. If ammonia exposure occurs, the element will corrode rapidly. Elec-tric heating elements usually are encapsulated or located in the furnace thatholds the process retort.

If the furnace is heated by natural gas, the burners usually are located ina remote heater box and the heated combustion air is distributed to theprocess retort. Alternatively, the combustion products can heat a series ofexternal 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 ambienttemperature (in °F); and 3412 is the conversion factor from Btu’s (Britishthermal units) to kW·h.

Note that the calculated figure is defined as “that energy required toheat the complete load up from ambient temperature to the process operat-ing temperature in an hour.” The calculated figure can then be furtherdivided by the number of hours in which it is both practical and economi-cal to raise the furnace temperature. Remember, the calculation must befor the gross load going into the furnace.

Normally the furnace designer will calculate the heat input calculationbased on the maximum furnace operating temperature. This will accountfor 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 doesnot account in any way for heat losses through the furnace insulation orthrough apertures such as seals, thermocouple holes, fan-drive shaft holes,or any other aperture.

The calculation for gas heating energy requirements for grossworkload is (Ref 4):

Btu’s = Specific heat × Load mass × Temperature

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



ing it necessary to divide the resulting value by whatever time factor isconvenient. The calculated value does not account for furnace heat lossesthrough refractories or apertures cut in the furnace. The “add on” kW orBtu’s will depend on the refractory value of the furnace insulation mate-rial 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 instru-mentation such as thermocouples and a time-temperature recorder.Advanced control instrumentation uses a computer for process controland a dual PC/PLC system (see Fig. 1). This system arrangement mustconform 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 isusually not as expensive as for more conventional control methods—except, of course, for the cost of programming. The level of programmingdepends 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 deter-mination of the number of control zones, the furnace will require at leasttwo thermocouples per zone: a set-point temperature thermocouple and anover-temperature thermocouple. The over-temperature thermocouple isnecessary to prevent temperature overshoots and furnace temperature“runaways.” If the over-temperature thermocouple is removable, alwaysremember to reload it.

Fitting a Load Thermocouple into the Retort. A hole is cut into theretort and a pipe with a single gastight sealed end is welded into place inthe box. The thermocouple is inserted into the open end of the tube. The



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

Monitoring Gas DissociationGas dissociation is simply the amount of decomposition of the ammo-

nia gas (Ref 5). During the nitriding process, ammonia gas is continuallydecomposing due to the catalytic effect of the steel and the retort. The gasbreaks down into its elemental form of hydrogen and nitrogen, and thusthe exhaust gases consist of ammonia and hydrogen:

(2NH3 + H2 + N2)

Of the above three gases, only ammonia is soluble in water, to the point thatwater 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 inthe nitriding box is first admitted into A by opening taps C and D. After theair has been expelled, taps C and D are closed. During the measurement,tap E is opened and the water immediately absorbs the undissociatedammonia. The water takes up precisely the volume previously occupiedby the ammonia, but the remaining N2-H2 gas (dissociated ammonia) doesnot dissove in water.

This is a simple and effective method of measuring the gas dissociation.Other methods include using an oxygen-measuring unit and recalibratingto feed in percent dissociation of ammonia or hydrogen.

The rate of dissociation at which the process should operate is usuallyaround 30%. This, of course, will depend on whether a single-stage or


B0

25

50

75

100

25%

0

25

50

75

100

50%

0

25

50

75

100

75%

Water

NH3 + N2 + H2

N2 + H2N2 + H2

N2 + H2

H2O

H2O

H2O

E

A

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

C

D

Fig. 5 Dissociation pipette (burette) schematic. See text for discussion.


two-stage process is being conducted. For the two-stage (Floe) process,the ammonia gas dissociation is at a lower value, generally (but notalways) around 15%.

Oxygen Probe ControlMany 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 UnitedStates, both of the Aga Gas Company, reported their computer controlmodel in a paper that discusses the equilibrium of hydrogen and oxygenwith water vapor (Ref 7). In order for the control system to be effective,the flow of the process gas components must be accurately metered usingmass flow controllers. These units will meter very precise gas flows to theprocess.

Nitriding SensorsDevelopment of the nitriding sensor has significantly contributed to

nitriding process control (Ref 8, 9). This system has been designed usingthe principle of a solid-state electrolyte, measuring variations in themagnetic and electrical properties of a steel as temperature increases andsurface chemistry changes. The sensor is placed in the furnace process-ing chamber and then calibrated according to the desired nitriding result,the parameters of the parts being treated, and the surface area of theworkpieces.

The control system allows the technician to control the thickness of thecompound layer and case depth. If water vapor or oxygen is present withinthe nitriding atmosphere, the oxygen partial pressure can be used as anindirect measure of the nitriding potential of the process atmosphere. Theresulting generated signal from the sensor indicates the partial pressure ofthe oxygen and the degree of dissociation occurring. Initial part clean-liness is essential, because any surface contamination will affect the sen-sor results.

REFERENCES1. Gas Nitriding, Heat Treating, Cleaning, and Finishing, Vol 2, 8th ed.,

Metals Handbook, American Society for Metals, 1964, p 149–1632. “Safety Requirements for the Storage and Handling of Anhydrous

Ammonia,” ANSI K61.1, American National Standards Institute3. G. Totten, G.R. Garsombke, D. Pye, and R.W. Reynoldson, Heat

Treatment Equipment, Steel Heat Treatment Handbook, G.E. Tottenand 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 HeatTreatment Handbook, G.E. Totten and M.A.H. Howes, Ed., MarcelDekker, 1997, p 663–720



6. V.O. Homerberg and J.O. Walstead, A Study of the Nitriding Process:Part 1, ASST Nitriding Symposium, American Society for SteelTreaters, 1929

7. L. Sproge and S. Midea, Analysis and Control of Nitriding and Nitro-carburizing Atmospheres, Carburizing and Nitriding with Atmos-pheres, ASM International, 1995, p 303–307

8. B. Edenhofer and J.W. Bouwman, Vacuum Heat Treatment, Steel HeatTreatment Handbook, G.E. Totten and M.A.H. Howes, Ed., MarcelDekker, 1997, p 483–526

9. Marathon Sensors Inc., Cincinnati, OH, www.marathonsensors.com



CHAPTER 6Salt Bath Nitriding

AN ALTERNATIVE to the gas nitriding process was sought in the mid-1930s that would produce a more uniform and better metallurgicallyformed case. Notable chemical companies conducted many experimentsto find an appropriate alternative method. A liquid, researchers thought,would fulfill the uniformity requirement through surface contact of theliquid to the steel. The depth and quality of case would be determined bythe chemical composition of the liquid. A heat source would be necessaryto drive the nitrogen into the steel surface.

A cyanide-based salt, it was discovered, fulfilled the requirements ofthe process, and salt bath nitriding began. Salt bath nitriding is in essencethe same process as gas nitriding; only the medium is different. Salt bathnitriding offers extremely good case depth uniformity.

Salt bath nitriding utilizes the melting of a salt containing a rich nitro-gen source. When heat is applied from either an internal or externalsource, 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 tem-perature in the molten salt, similar reactions begin to occur as in gasnitriding; that is, controlled amounts of nitrogen are released to diffuseinto the steel surface.

Salts Used and Process Advantages

Nitriding Salts. Early on, a cyanide-based salt, such as NaCN, was usedwith other salts. A typical nitriding salt mixture would be (Ref 1):

• NaCN, 30% • KCl, 39%• Na2CO3 or K2CO3, 25% • Moisture, 2%

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

• NaCN, 60% • KCl, 24%• K2CO3, 15% • Moisture, 1%




Advantages. The salt bath nitriding technique gained popularity in theearly 1950s because it required a lower capital investment than gasnitriding. Operating costs also were considerably lower, making it a cost-effective method of nitriding steel. An added benefit was that salt bathnitriding was carried out at a slightly higher temperature than gas nitrid-ing: 565 to 585 °C (1050 to 1085°F). This meant faster processing time.

To summarize, the advantages of using salt bath nitriding equipmentwere:

• 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, manyderivatives have appeared (Ref 1). The two major European companies thatpioneered the salt bath process were the Cassel Division of Imperial Chem-ical Industries (ICI) in England and the Durfferit Division of Degussa inGermany.

The early process was developed by ICI, which was eventuallyacquired 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, Degussadeveloped the two-component process of forming both nitrides and somecarbides in the steel surface (i.e., nitrocarburizing). The German processwas introduced as the Tufftride process. Tufftride became popular inEurope in the early 1950s and was introduced to the United States in theearly 1960s. It was quickly accepted as a reliable procedure and identifiedby 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, whichis corrosion resistant.

Because of emerging environmental concerns, Degussa pioneered theuse of a low-cyanide salt. Kolene Corporation developed a low-cyanide



salt process, now known as the Melonite process (Ref 2), which offered analternative to the Tufftride process. The process uses a cyanide-free salt,the composition of which is specified under the SAE designation AMS2753. Another derivative of the Melonite process by Degussa is known asthe QPQ (“Quench-Polish-Quench”) process (QPQ is a trademark of theKolene Corp.).

Both the Melonite process and the QPQ process require the part to bequenched into a molten oxidizing bath to neutralize any residual cyanidethat might be present. After the quench procedure, the part is mechanicallypolished, 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 con-tinuous system. Either equipment system can be electrically or gas heated(Fig. 1).

The process time is generally around 60 to 90 min at temperature (parttemperature), followed by a quench, then into the oxidizing bath operatingat a process temperature of 400 °C (750 °F), followed by cooling with awater quench. At this point the process may be considered complete, orthe part can be mechanically polished. After polishing, the part can beimmersed in another oxidizing bath to complete the procedure.

The metallurgical results are the same as with the conventional gasnitriding process: A compound layer is formed at the steel surface with adiffusion zone area immediately below the compound layer. The surfacecompound layer thickness will be determined by the steel composition.Usually case depths are confined, but not limited, to shallow depths ofapproximately 0.13 mm (0.005 in.). Analysis of the compound layer onplain carbon steel showed it to contain a nitrogen concentration of around6%, with very small amounts of carbon. If the procedure used is the Mel-onite process, which includes a post-oxidize treatment, then a shallow sur-face oxide layer will be seen microscopically on the immediate surface ofthe part (Fig. 2, 3).

Using a New Salt Bath

A newly made-up nitriding salt bath should not be used immediatelyon 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, thesalt 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 thesteel.

Chapter 6: Salt Bath Nitriding / 55


The bath must be raised to operating temperature and “aged” by oxidiz-ing its cyanide content through decomposition of the cyanide to cyanate.The aging period depends on the surface area of the salt bath. For exam-ple, a bath that is 750 mm (30 in.) in diameter by 750 mm (30 in.) deepwill take approximately 12 to 16 h to age. It is important to note that nowork should be processed through the bath during the aging process.


Principal furnace types for liquid salt bath nitriding. (a) and (b) Exter-nally heated. (c) and (d) Internally heated, with immersed alloy elec-

trodes and metal liner or submerged electrodes with ceramic tile lining

Fig. 1


As with gas nitriding, workpieces must be cleaned prior to salt bathnitriding to prevent any oil, grease, paint, and so forth from contaminatingthe bath. In addition, the steel surfaces must be free of oxides or other contaminants.

A cyanate decomposition percentage up to 25% normally will produceacceptable surface composition and hardness results. If the carbonate con-tent is high, the temperature of the bath should be lowered to around 440


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

Fig. 2

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.

Fig. 3


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 bathshould be discarded once every 3 to 31/2 months and replaced with a newbath. Again, care must be taken to age the new bath.

Bath Replacement

The bath is ready for replacement if it is becoming difficult to maintaina cyanate level around 25% or if signs of corrosion are evident on the steelsurface. Be careful not to confuse the corrosion with that caused by post-washing of the steel.

The cyanide to cyanate control level is not the same for all steelsprocessed through the nitriding bath and will vary for various steels. Sodiumcyanide (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 astandard solution of silver nitrate (AgNO3) (Ref 3). The test methodsdescribed in the following paragraphs are simple titration methods ofanalysis and not complex. It is essential, however, before testing that theequipment used in the sampling be thoroughly clean and that the salt sam-ple taken from the bath also be clean and free of contaminants such assludge or graphite, which could be on the molten salt surface.

Materials and EquipmentThe 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. Eachgraduation on the burette is equal to 0.5 mL (0.5 cm3)

• One sample spoon designed to hold approximately 1 g of powderedsample

• 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)



It is advisable to maintain a shift logbook of titrations of the nitriding bathto 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 intothe molten operating bath and remove immediately. A film of salt willfreeze and adhere to the rod. Scrape off and collect about 10 g in aclean mortar. Crush to a fine powder.

2. Measuring sample: Measure out 1 g. Transfer the sample from thespoon to the 250 mL glass beaker and fill to the halfway mark withdistilled water, preferably warm.

3. Dissolving sample: It is preferable to bring the solution to a gentleboil; however, it can be dissolved by stirring. Add a small pinch oflead carbonate (1/4 to 1/2 g). Disregard the small amount of blackresidue. Allow the solution to settle or precipitate the lead sulfide(PbS).

4. Decanting (satisfactory for ordinary shop practice): Permit the solu-tion to settle for 5 to 10 min until all solids precipitate to the bottom ofthe beaker. Decant the clear liquid into the second beaker and proceedwith titration.

5. Filtration (for maximum accuracy): Filter through 11 cm No. 2 What-man filter paper or the equivalent. To the filtrated solution, add 3 or 4drops of indicating solution (10% KI).

6. Operation of the burette: By pressing the bulb, you automatically fillthe burette. The burette will always fill to the zero mark and anyexcess will return to the bottle. The tube should always be filled beforeeach test. Pressing the inch clamp will release the chemical reagentthrough the glass delivery tip into the beaker.

7. Titration: Titrate by slowly adding the chemical reagent solution fromthe burette until the contents of the beaker turn cloudy. At first, a slightdiscoloration or cloudiness may form. Ignore this. Continue titratinguntil the solution is entirely cloudy and opalescent (resembling thecolor of yellow lemonade). This is the “end point.”

8. Determining % NaCN in the bath (1 g sample): The reading on thegraduated burette is the percentage of NaCN in the salt bath tested.

Important Factors for Successful ResultsWash all grease and oils from parts to be nitrided. Oils may contain

sulfur, which is detrimental and will reduce efficiency. If work iswashed in an alkaline cleaner containing sodium hydroxide or similaralkali, the work should be rinsed in hot distilled water and dried beforenitriding.



Determination of Sodium Carbonate and Sodium CyanateSodium 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 mLErlenmeyer flask, and add 100 mL of distilled water.

• After the sample has dissolved, add 25 mL of a neutral 19% bariumchloride solution. Stop flask tightly, and let stand for 2 h.

• Filter and wash the precipitate with neutral 1% barium chloride untilthe washings give no color with phenolphthalein.

• Place the filter paper and precipitate in the original flask. Add 50 mLof 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 carbon-ate 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 meas-ured in grams. The normality of a solution is the number of gram equiva-lent weights of solute per liter.

Sodium Cyanate. The procedure for determination of sodium cyanatein the bath consists of the following steps:

• Accurately weigh an approximate 1 to 2 g sample of the bath andtransfer to a 650 mL Kjeldahl flask. Use a 1 g sample when checking anitriding bath and a 2 g sample when checking a carburizing bath.

• Add 100 mL ammonia-free distilled water, 2 or 3 selenized Hengargranules, and 5 mL of concentrated sulfuric acid. Digest under thehood or in a Kjeldahl digesting unit until the volume reaches approxi-mately 25 mL. Allow to cool. This digestion drives off all the cyanidenitrogen as hydrogen cyanide and fixes the cyanate nitrogen as ammo-nium bisulfate or ammonium sulfate.

• Add 350 mL of ammonia-free distilled water and 50 mL of ammonia-free 72% sodium hydroxide to the digested sample in the Kjeldahl flask.

• Immediately connect to a Kjeldahl distilling unit, being sure to includea Kjeldahl trap before entering the condenser. Place the tip of the deliv-ery tube into a 500 mL Erlenmeyer flask containing 50 mL of saturatedboric acid solution with methyl red indicator added. The tip of thedelivery tube must be below the surface of the boric acid solution.



• Distill off 150 to 200 mL of liquid, disconnect, and wash down thecondenser 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 cyanatepercentage:

% 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 ofsolute per liter. Note: To avoid running appreciable blanks, reagent-gradechemicals and ammonia-free water should be used for all solutions.

Bath Maintenance

Maintenance of the salt bath and related equipment is critical and quitesimple. The procedures can be broken down into daily, weekly, and monthlyactivities (Ref 1):

Daily maintenance consists of:

• Analyzing the bath for cyanide and cyanate content at the start of eachshift

• Checking that the temperature control instrument is in proper workingorder

• Checking the cleanliness of the bath by using a perforated scoop anddesludging the bath before the start of the shift

Weekly maintenance consists of:

• Checking the cleanliness of the bath around the top (that is, the partabove salt level)

• Removing the salt pot and checking the integrity of the external sur-faces, especially if the salt bath is gas fired. Watch for signs of excessscale and “ballooning” at the bottom of the pot, which means that thepot wall is thinning and the salt is too heavy for the pot at that wallthickness.

Monthly maintenance consists of:

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



• Determining that the burner ignition system is operational and that theflame rod is clean

• Checking that no traces of nitriding salt are present on the cleanoutport

• Checking that the elements are fully immersed if the furnace is electri-cally heated using an immersed heating element

• Checking that the elements show no signs of deterioration or indica-tions of hot spots if using an externally heated salt bath. Deteriorationof the elements is easy to see.

• Checking that the ammeter phases are in balance. If there is an imbal-ance, the elements should be checked for uniform wear or potentialbreakages.

Operating the Salt Bath

As discussed, furnace operating personnel must ensure that the bath isfully aged, regenerated, and of the correct salt analysis. Once these param-eters have been established, the bath is ready for operation. As with anysalt bath procedure, the steel workpiece must be thoroughly preheated andfree of any surface moisture—either on the steel, the support fixture, orthe fastening wire—before immersing it into the molten salt. On immer-sion, the workpiece will freeze the immediate salt surrounding it andcause the salt to solidify on the steel surface. The “cocoon” of salt sur-rounding the steel acts as an insulator to shield the steel from excessiveheat. This will reduce the risk of distortion through thermal shock.

Due to its nature of being molten when under heat, the salt will slowlyoxidize at the salt/air interface and cause a breakdown of the cyanide tocarbonate. 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 withhigh-temperature salts (Ref 3). Some of the more important considera-tions are:

• Always preheat to both reduce thermal shock and remove moisturefrom the steel part surface.

• Do not mix nitrate salts with cyanide-based salts, as this poses a seri-ous 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.



• Make sure that secure storage is available for the storage of cyanide-based 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 gasnitriding. However, the salt bath furnace tends to be a more compact unitthat occupies less floor space than its gas nitriding counterpart.

In addition to batch-type equipment, continuous systems are availablefor salt bath nitriding. Most systems are encapsulated, with a protectivefront and an observation panel.

REFERENCES1. Liquid Nitriding, Heat Treating, Vol 4, ASM Handbook, ASM Interna-

tional, 1991, p 410–4192. J. Easterday, Salt Bath Nitriding Proceedings (Detroit), Kolene Corp.,

Oct 19953. Cassel Manual of Heat Treatment and Case Hardening, 7th ed., Impe-

rial Chemical Industries, Ltd., Jan 1964




CHAPTER 7Control of the

Compound Zone orWhite Layer

THE COMPOUND ZONE is more commonly known as the whitelayer, simply because when the nitrided sample is sectioned through thecase, and then polished and etched with a standard solution of nital (2 to5% nitric acid and alcohol), the immediate surface etches out white inappearance above the nitrided case. The zone is called “compound” dueto the presence of more than one phase (Fig. 1). Two phases generally arepresent in the compound zone: the epsilon (ε) phase, which has a chemi-cal formula of Fe2-3N, and the gamma prime (γ ′ ) phase, which has achemical formula of Fe4N (Fig. 2). Depending on which phase domi-nates, spalling or chipping can occur during service (Ref 2). Choice of

Formation of the nitrided case. Courtesy of Pye Metallurgical Consult-ing, Inc.

Fig. 1




the necessary surface metallurgical phase should be determined by thepart 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 thesurface of a nitrided part will indicate the presence of the white layer. Ifthere is a white layer (or any remaining white layer after grinding orchemical finishing), the drop will deposit copper onto the surface. If thereis 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 forhistorical background). He developed what was originally known as theFloe process, now more commonly called the two-stage or sometimes thedouble-stage process. The purpose of the two-stage process is to reducethe thickness of the compound zone on the immediate surface of the steelby reducing the nitriding potential at an elevated temperature. The firststage of the process ensures rapid formation of the white layer, and thesecond stage arrests the formation of the white layer without allowing thediffusion 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 approxi-mately 30% at 495 °C (925 °F). This is followed by raising the temperatureduring the last third of the cycle to 565 °C (1050 °F) with a dissociationrate of 75 to 80%. In some applications, formation of the surface com-pound zone is desirable. If the layer is not required, it can be removedmechanically (by grinding), or chemically (using a cyanide solution devel-oped by Bell Helicopters) (Ref 4).


Fig. 2 Formation of the compound zone. Source: Ref 1


Other Methods for Controlling Compound Zone Formation

Nitriding Potential. Control of the compound zone can be achieved bycontrolling 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 thatat nitriding potentials that completely avoid white layer formation, onlysteels that contain aluminum will nitride satisfactorily. Steels that dependon chromium as the primary nitride-forming element will nitride veryslowly.

The white layer constituents must be stable to provide a fine dispersionof chromium nitrides in ferrite. Strict control of the nitriding potentialrestricts white layer growth.

Building on Adolph Machlet’s early work with dilution techniques (seethe later discussion in this chapter), Thomas Bell of the University of Liv-erpool began work in the early 1970s on dilution of the ammonia gas byhydrogen (Ref 2). His objective was to establish the mechanism of com-pound layer formation and to determine how the layer thickness can bemanaged. By monitoring the nitriding atmosphere (an ammonia-hydrogenmixture), and knowing the input gas (whether ammonia to enrich thenitrogen content or hydrogen to dilute the nitrogen content), he discoveredthat the nitriding potential can be controlled to produce components with awhite layer thickness of no more than 0.004 mm (0.00016 in.).

The dilution process was first developed by Machlet when he patentedhis process in 1913 (see Chapter 1 for further details). The patent wasgranted for the nitrogenization of low-alloy steels and cast irons, usingammonia gas diluted with hydrogen. Machlet had observed compoundlayer formation without realizing its significance, yet recognized that thelayer could be controlled by dilutant gases such as hydrogen. The sole

Chapter 7: Control of the Compound Zone or White Layer / 67

Fig. 3 Typical two-stage process


purpose of the dilution procedure, using hydrogen, is to reduce the nitrid-ing potential of the ammonia gas. Some nitriding potential will remain,however, and thus a very shallow compound layer may form on the imme-diate surface, albeit microns thick.

Careful control of the dilution gas in relation to the process gas is neces-sary in order to ensure the required surface metallurgy. Based on thismethod, a Canadian company (Nitrex Metal, Inc.) manufactures computer-controlled furnace equipment to control compound layer formation. Theequipment seems to be enjoying good success (Ref 6).

Ion Nitriding. By ionizing molecular gases and preparing the steel sur-face in a completely different manner compared to the gas nitriding proce-dure, the compound layer can be controlled to the point of elimination.The procedure relies on the ability of the control parameters to change asrequired by:

• Process gas ratios• Process operating pressure levels• Process temperature

With the exception of plain low-carbon steels, most steels can be satisfac-torily 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 diffusionof 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 inrelation 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



case depth? In the late 1930s and early 1940s, F.E. Harris determined asimple formula for the effect of diffusion at temperature (Ref 7). The for-mula is based on the square root of time at a particular temperature multi-plied 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 fol-lowing table:

Temperature

°C °F Temperature factor (K)

460 865 0.00150470 875 0.00155 475 885 0.00172 480 900 0.00195 500 930 0.00210510 950 0.00217 515 960 0.00230525 975 0.00243540 1000 0.00262

These temperature factors are based on a commercially available nitrid-ing steel without the addition of aluminum and do not account for whatmight be considered abnormally high alloying contents (such as might befound in stainless steels). The factors also do not account for surface con-dition and variations in the particular cast/melt. The factor values areapproximate and should be used only as a guide. Similar guides for esti-mating case depth for plasma nitriding and ferritic nitrocarburizing can befound in Chapters 12 and 21, respectively.

Remember, the higher the alloying content of the steel is, the slower thediffusion rate of nitrogen into the steel surface. The lower the alloyingcontent is, the faster the diffusion rate.

Temperature uniformity within the process retort is critical to uniformcase depth as well as uniform case metallurgy. If temperature variationsoccur within the process retort, both the compound layer and the casedepth also can vary.

REFERENCES1. M.A.J. Somers and E.J. Mittemeijer, Härterei-Technische Mitteilun-

gen, Vol 47 (No. 5), 19922. T. Bell et al., Controlled Nitriding in Ammonia-Hydrogen Mixtures,

Heat Treatment ’73, The Metals Society, Dec 1973, reprinted in SourceBook on Nitriding, American Society for Metals, 1977, p 259–265

Chapter 7: Control of the Compound Zone or White Layer / 69


3. D.B. Clayton and K. Sachs, Reduction of “White Layer” on the Sur-face of Nitrided Components, Heat Treatment ’76, The Metals Society,May 1976, reprinted in Source Book on Nitriding, American Societyfor 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 WithoutWhite-Layer Formation, Heat Treatment ’73, The Metals Society, Dec1973, reprinted in Source Book on Nitriding, American Society forMetals, 1977, p 248–254

6. W.K. Liliental, G.J. Tymowski, and C.D. Morawski, Typical NitridingFaults and Their Prevention Through the Controlled Gas NitridingProcess, Ind. Heat., Jan 1995, p 39–44

7. F.E. Harris, Case Depth, Met. Prog., Aug 1943, p 265–272



CHAPTER 8Ion Nitriding

GAS IONIZATION is a method of causing a gas to develop an electricalimbalance. At a particular pressure and electrical potential, the gas willglow—not unlike the gas used in neon light tubes. The phenomenon of gasionization was first investigated in the 1800s, when studies made in the upperreaches of the Arctic Circle in Norway showed the aurora borealis to be low-pressure ionized air that exhibits a characteristic spectrum of the rarer gasespresent in the upper atmosphere. Observable in an indefinite region of thenight sky around the magnetic pole, the ionization is influenced by the mag-netic 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 thenitriding process that utilizes it is known as ion nitriding. Ion nitriding isalso 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 processingtool by Dr. Wehnheldt, a German physicist, but he was unable to control itas a nitriding process due to the instability of the glow discharge. Thistype of instability can be seen in “lumina storm” novelty lamps. The dis-charge dances away from the center ball and around the inner glass case.




Shortly after Wehnheldt’s discovery, he met the Swiss physicist andentrepreneur Dr. Bernhard Berghaus (Ref 1). Together they were able todevelop, control, and market the process as a gas nitriding alternative thatoffered acceptable control of the compound zone. Development of theplasma 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 processwas General Electric. G.E. engineers Dr. Claude Jones, Derek Sturges,and Stuart Martin successfully investigated the glow discharge method ofnitriding and were able to use the process on a wide variety of materialsand components (Ref 2).

In Germany, the work of Wehnheldt and Berghaus led to formation ofthe company Klockner Ionen, which commercialized the process. Thecompany designed and built ion nitriding equipment, and licensed otherinternational companies to build the equipment and develop the process as well.

During the mid-1970s, scientists at the University of Aachen in Ger-many worked on better methods of controlling the glow discharge andother associated phenomena such as arc discharging. The proceduredeveloped at Aachen was that of pulsed dc current technology, which sim-ply means interrupted power to the point of power shutdown. This tech-nique offered many advantages to process engineers in terms of control ofthe nitriding procedure.

How the Ion Nitriding Process Works

The process is based on the phenomenon of current flowing betweentwo electrodes placed in a sealed gaseous environment. The gas within thetube acts as an electrical conductor and carries the current from one side tothe other as it would if it were a wire conductor (Fig. 1). The gaseousatoms become excited and are propelled along a very short “mean freepath” and collide with one another. When this occurs, energy is releasedand a “glow” occurs—hence the name “glow discharge nitriding.” Thecolor of the glow is determined by the type of gas used.

At normal atmospheric temperature and pressure, the resulting energyis too low to have any significant use as an energy source in terms of pro-viding 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 molec-ular impact is great but is infrequent due to the long molecular mean freepath that each gas molecule has to move to find another molecule toimpact. Thus the plasma can then generate heat, but in an amount insuffi-cient to heat the workpiece surface for plasma processing. In other words,



because of the infrequency of molecular impact, the resulting energy stillcannot be used as a heating medium.

There is no ideal pressure value, but there is a range in which the pres-sure can be adjusted to suit the operating parameters of the particularmaterial 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 deter-mine 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 partwill become intermittent. Where there is no glow on the steel surface,there will be no nitride formation. This means that the voltage has failed tosupport the plasma ignition at that particular pressure. In addition, high-temperature areas can occur at sharp corners, often resulting in localizedoverheating and even burning. If the pressure is too low (i.e., toward highvacuum), then the glow seam around the part will be very “foggy,” result-ing in ineffective nitriding of the internal surface areas of holes.

Pressure is one of the principal areas of control. If the pressure condi-tions are correct and the voltage is too high, then a discharge will occurmuch like a lightning strike. This occurrence is known as the arc dischargeregion. If the voltage is too low and the pressure too high, the glow willdisappear and nothing will happen.

Chapter 8: Ion Nitriding / 73

Influence of pressure on the glow discharge. (a) A sealed glass tube con-taining a gas at normal atmospheric temperature and pressure (100 kPa,

or 15 psi) conducts an electrical current. The gas glows brightly but does notrelease 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 lowenergy. 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.

Fig. 1


Glow Discharge Characteristics

To understand the principles of the glow discharge, it is necessary torefer to the Paschen curve (Fig. 2). The Paschen curve is a comparison ofinput voltage in relation to current density of the steel part surface, and thevarious events that take place depending on voltage in relation to currentdensity. By understanding the Paschen curve, one can determine processvoltage requirements.

Nonmaintained Region. The nonmaintained region of the Paschencurve states that if a voltage is applied to a gas, then the electrons withinthat gas can be charged to the point where an electrical ignition of the gaswill occur. This can be likened to the spark that occurs when an automo-bile spark plug is charged with high voltage. The air between the plug gapis electrically charged to the point where a spontaneous spark will occur.In the process chamber the electrons from the ionized, ignited gas will beaccelerated toward the cathode from the anode. Once the molecular colli-sion begins to occur due to the gas ionization, energy (heat) will be gener-ated at the work surface. When the process voltage is increased, then anappropriate increase will occur in the current density. The gas ionizationwill progress into the next phase of the Paschen curve.

Self-Maintained Region (Townsend Discharge). The self-maintainedregion on the curve is the area in which more electrons can be released byfurther gas ionization, which will perpetuate even further gas ionization.The area can be considered to be a chain reaction.


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

Fig. 2


Transition Region (Corona). The current density will increase if thecurrent limiting resistance is reduced, thus causing a voltage drop betweenthe cathodic workpiece and the anodic internal vessel wall. Within thisarea, voltage stability cannot be maintained.

Subnormal Glow Discharge. In this region the glow discharge isbeginning to ignite and a very fuzzy glow will be seen.

Normal Glow Region. It is at this point on the Paschen curve that aglow seam will completely cover the work. Its thickness will be deter-mined by the chamber vacuum pressure and the process voltage.

Abnormal Glow Region. It is in this region that the glow seam willcompletely cover the steel work surface uniformly, following the geomet-ric profile of the workpiece. One must now adjust the process pressure toensure penetration of blind holes and cavities. However, the ratio of holediameter to hole depth must be considered:

1× hole diameter : 4× hole depth

The blind hole depth can equal four diameters. This ratio will follow forsmall 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 forsmall hole penetrations, but for large holes from 50 mm (2 in.), the rulewill not apply; the glow seam can be “forced” down the hole by theadjustment of pressure. It is in this region of abnormal glow that plasmanitriding takes place and where ideal conditions exist.

Arc Discharge Region. As the current density increases, a noticeableincrease in the voltage drop will occur, causing an appreciable increase inthe power density at the work surface. As power density increases, thetemperature of the steel work surface rises to the point of serious over-heating, resulting in metallurgical problems such as grain growth, local-ized melting, and pitting. As the power intensity builds, the potential foran arc will occur. This arc, visible through the process chamber sightglass, looks like a lightning strike.

Process Control

Process parameters requiring good control are:

• Current density• Power• Process chamber pressure• Gas composition



Instrumentation for achieving control of these parameters is the per-sonal computer/programmable logic controller (PC/PLC) combination,ideal for the plasma nitriding process.

Control Characteristics. When a partial pressure is created in aprocess chamber and a constant voltage is applied to the chamber with theappropriate cathode potential and anode, then the gas in the partial pres-sure will ionize. If a steel part is placed in contact with the cathode so thatit is at the same electrical potential, then not only will the partial pressuregas “glow,” it also will create heat in the steel part at the cathode potentialdue to the kinetic energy generated by ionic bombardment.

If the process gas used in the process chamber is diffusible in steelwhen 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 areno heating elements within the furnace, simply because the thermalenergy is created by the ionic bombardment of the workpiece by the freegas ions. The workpiece is heated by kinetic energy from ionic bombard-ment. However, some systems do use internal elements to assist in tem-perature uniformity with large, densely packed loads. When using a sys-tem with internal heating elements, care must be taken in positioning theworkpieces to prevent localized overheating. Otherwise, nonuniform sur-face 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 usedfor 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-filmdeposition 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 care-fully prepared steel surface, resulting in hardness and corrosion resistance.The thin-film coating is usually (but not always) preceded by a plasmanitriding procedure, which produces a hard substrate to which the thin-film material can bond. An example of thin-film processing is plasma-assisted chemical vapor deposition (PA-CVD) using chromium, tungsten,



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 ofnitrogen and hydrogen. The reverse will happen on cooling the decom-posed gas. In ion nitriding, unlike gas nitriding where ammonia gas isused, the gases are brought in as separate gases:

N2 + H2

Because the gases are not in a combined form, the metallurgist is able tovary the nitriding potential by varying the proportions of the individualgases. By varying the hydrogen-to-nitrogen ratio of the elemental gases, thecompound zone (white layer) formation can be controlled. When the nitro-gen 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 intothe following elemental gas ratios (Fig. 3):

1 Nitrogen : 3 Hydrogen

2NH3→ N2 + 3H2


Fig. 3 Illustration of the ammonia molecule 2NH3 and its decomposition

←


This formula for ammonia shows three hydrogen molecules to one nitrogenmolecule, and the ratio is a fixed ratio. If the ratio of nitrogen gas to hydro-gen gas is varied, any ratio can be selected to create any particular surfacemetallurgy. The ratio can be a low nitrogen to hydrogen ratio, or a highnitrogen to hydrogen ratio, depending on the surface metallurgy require-ments. In other words, a fixed gas chemistry produces a fixed surface metal-lurgy, and a variable gas chemistry allows a variable surface metallurgy.

Variable gas proportions permit creation of the appropriate surface met-allurgy 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 thematerial being treated.

Reaction I. Ionized and neutral nitrogen atoms are produced by ener-gized electrons:

e– → N2 = N+ + N + 2e–

Reaction 2. Iron and other contaminants are removed from the surfaceof the work by an action known as sputtering. The impact of the nitrogenions bombarding the work surface dislodges the contaminants, which areremoved by the vacuum pumping system. Contaminant removal can beloosely described as atomic cleaning and allows nitrogen to diffuse intothe work surface:

N+ → Work surface = Sputtered Fe and sputtered contamination

Reaction 3. As a result of the impact of the sputtered iron atoms, caseformation begins at the work surface of iron nitrides:

Sputtered Fe + N = FeN

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



blasting where the carrier medium is the air blast and the cleaning mediumis the steel shot. In this instance the gas ion is like the steel shot, and theelectrical imbalance is like the air blast. The surface becomes atomicallycleaned.

Reaction 4. At the work surface the breakdown of FeN begins underthe influence of continual plasma bombardment. The plasma causes insta-bility of the FeN, which begins to break down into the ε-phase, followedby 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)


Glow discharge ion nitriding mechanisms (Koelbel’s models). Note thevoltage 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 theplasma will glow brightest.

Fig. 4


Thus, if one can control the reactions at the steel surface, one can con-trol the surface metallurgy. Hardness profiles show control of the forma-tion of the compound zone and the diffusion zone.

Surface Stability

Each of the surface layers achieves excellent dimensional stability sim-ply because the temperature to generate plasma is not dependent upon aconventional heat source. The plasma energy is the heat source, and theprocess temperature can be adjusted to suit the steel by manipulation ofplasma voltage and pressure. However, use of a lower process tempera-ture 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 diffu-sion heat treatment process, including plasma nitriding, gas nitriding, andcarburizing (Fig. 5). Because nitrogen is diffusing from all angles of thecorner, a normal reaction of nitrogen saturation occurs at the corner (par-ticularly on gear teeth). If allowed to proceed unchecked, a supersaturatedsolution of nitrogen can form at the corners, leading to nitride networkingthroughout the corner region. The net result is that the corner can becomevery brittle and prematurely fail by chipping or spalling (Fig. 6, 7). If the


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

Fig. 5


nitrogen is controlled to reduce the nitriding potential to suit the partapplication, then the risk of nitride networking is considerably reduced ornearly eliminated.

Degradation of Surface Finish

Sputter cleaning and ion nitriding generally enhance the surface condi-tion of the workpiece. Ionic bombardment of the steel surface reducesturning lines and high points. Of course, care must be exercised whenusing denser gases such as argon for ionic bombardment. Argon will tendto etch the surface to the point where the surface will become pitted.When aggressive sputter cleaning is necessary, it is strongly recom-mended that the argon be blended with hydrogen to a ratio of 95% H2 to5% Ar up to a maximum of 90% H2 to 10% Ar.


Chipping at the pressure point of a gear tooth. Courtesy of Pye Metal-lurgical Consulting, Inc.

Fig. 6

Extrusion die with surface exfoliation. Courtesy of Pye MetallurgicalConsulting, Inc.

Fig. 7


Control of the Compound Zone

Plasma nitriding can control compound zone formation on many differ-ent 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, impact-resistant, 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 mixedphase compound zone of both ε and γ ′ (usually around 50% each). Highinternal stresses result from the different phase volumes, which meansthat the crystal interfaces are inherently weak. The thicker the compoundzone, the weaker it becomes, causing weak crystal boundaries within thezone to fail even under small loads. This is particularly evident on partssuch as H13 hot-work tool steel for aluminum extrusion dies (Fig. 8a). Ifthe compound zone is allowed to develop into a thick layer, the die willbend when loaded by the aluminum billet (Fig. 8b). As the die bends, theloaded face becomes concave and the back face becomes convex. Thecompound zone on the loaded face will undergo compression and crack.This also will occur if the effective case depth of the nitrided layerexceeds 0.25 mm (0.010 in.).

A similar effect will occur on an H13 hot-work press forging die if thecase is too deep and the compound layer too thick. As the press loads thedie face, the surface of the die becomes compressed and the formed casedeforms at the surface, leading to surface cracking and exfoliation. Thisis not serious if the exfoliation occurs on areas of the die that are not inuse, but when exfoliation occurs on a working face it becomes a majorproblem.

Cracks can initiate quite easily due to impact loading or incorrect pre-heat treatment. If the core hardness mechanical properties are lower thanthe load conditions of the treated steel, then the core is likely to fail to sup-port the nitrided case on the surface, no matter how well the nitriding has



been done. Therefore, the thinner the compound zone (to the point ofelimination), 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 gasesprimarily used for ion nitriding. Argon can be used, but only to assist incomponent cleaning (known as sputter cleaning) before the nitridingsequence. Methane can also be used to deliver controlled amounts of car-bon to influence control of the ε-phase in the compound zone. Once again,care should be exercised in using methane; too much carbon can activelypromote a dominant ε compound phase.

Accurate control of the gas delivery into the process retort ensuresaccurate control of the nitriding metallurgy. Precise metering of each gasis accomplished by mass flow controllers.


Cross section through an extrusion die. (a) Bearing face and reliefclearance 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 tearon the case.

Fig. 8


Process Parameters

In conventional gas nitriding, time, temperature, and dissociation (orsalt analysis for salt bath nitriding) are the key process parameters thatmust 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 parameterscan be controlled to ensure repeatable process conditions.

Temperature uniformity during the nitriding process is critical. With-out it, the formed case will be nonuniform. Some latitude in temperaturetolerance can be allowed, but usually no more than ±5.5 °C (10 °F) of theprocess set point. Remember, the higher the process temperature, thegreater the risk of nitride networks due to the potential for an excess ofnitrogen in iron. This also means that nonuniform case depths will occuracross the temperature range within the process retort. For example, a tem-perature difference of 35 °C (60 °F) over a 24 h cycle period can result in acase depth variation of up to 0.038 mm (0.0015 in.). Obviously, the longerthe process cycle time, the greater the case depth variation. This principleapplies 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 willvary with the pressure (Fig. 9). With this type of system, the process heatgeneration normally is derived from the kinetic energy created on ionicbombardment. This means that the furnace does not have heating ele-ments as are often seen in conventional furnaces.

Pulsed dc can be likened to entering a darkened room and switching onthe light. On leaving the room, the light is switched off. The peak voltagefrom the power source is constant, but the duration of the power to thelight is variable. Pulsed dc power supplies have the ability to switch thepower on and off as required by component geometry. The pulsed dc peakvoltage can be varied according to the part and chamber configuration(Fig. 10). The equipment will be discussed in greater detail in Chapter 9.



Advantages

Here are a few of the advantages of the plasma generation technique fornitriding:

• 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


Fig. 9 Continuous dc power plasma nitriding

Pulsed dc plasma showing Pmax, Pmin, and Ptemp (where P = power) inrelation to power input versus time. Note that the voltage can be

adjusted, as can the duration of the pulse. Courtesy of Plateg USA

Fig. 10


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

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

Oxynitriding

The oxynitriding process has grown in popularity over the past twodecades, especially in pulsed plasma nitride processing. The pulsedplasma nitride unit offers the capability of a controlled backfill of mois-ture in the form of either a vapor or oxygen-bearing gas. The purpose ofthe oxynitriding process is to form a controlled oxide layer on the surfaceof the treated steel. Once the oxide barrier has been formed, there is aresistance to corrosion. The degree of corrosion resistance will be deter-mined by the thickness of the oxide layer, which in turn is determined byboth time and temperature. The oxynitriding process can be performed ingas, salt, or plasma.

The procedure is done on completion of the nitride cycle, when the con-trol program moves into the cooldown mode. It is during the cooldown pro-cedure that oxygen is fed into the process chamber (Fig. 11). The net effectis that the nitrided surface is deliberately oxidized to provide a corrosion-resistant, oxygen-rich surface layer. Upon completion of the cooldown, thefurnace bell is opened and the oxynitrided work is removed.

Figure 12 shows a plain carbon-manganese steel piston rod that hasbeen nitrided, followed by the controlled oxynitride procedure. The centerrod shows the rod before the start of treatment. The two rods at the righthave been oxynitrided, and the two rods at the left are untreated; all fourhave been subjected to salt spray testing.

The process gases used for the controlled oxynitriding procedure areenvironmentally friendly and pose no threat to the ecology or the immedi-ate environment. The surface finish of the steel after the procedure is darkblue, almost matte black. The oxidation layer is usually around 1 µmthick, but can be varied according to the cooldown time and the time heldat an elevated temperature (around 900 °F, or 480 °C). Generally the pro-cedure is to go into cooldown immediately after nitriding and commencethe controlled oxidation treatment. It is a simple, effective procedure thatadds no significant time to the overall process.



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


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 UNSG41400 and H41400 chromium-molybdenum steels. Courtesy of Plateg USA

Fig. 12

Fig. 11 Schematic illustration of the oxynitriding process


REFERENCES1. F. Hombeck, Forward View of Ion Nitriding Applications, Ion Nitrid-

ing, T. Spalvins, Ed., ASM International, 1987, p 169–1782. C.K. Jones, D.J. Sturges, and S.W. Martin, Glow Discharge Nitriding

in Production, Met. Prog., Dec 1973, reprinted in Source Book onNitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society forMetals, 1977, p 186–187

3. Heat Treating, Vol 4, ASM Handbook, ASM International, 1991



CHAPTER 9Ion Nitriding Equipment

ION NITRIDING EQUIPMENT falls under two categories: cold-wallcontinuous direct current (dc) technology and hot-wall pulsed dc technol-ogy. Both are described in this chapter along with other important consid-erations for ion (plasma) nitriding equipment and processing.

Cold-Wall Continuous dc Plasma Nitriding

The cold-wall continuous dc plasma system is perhaps the simplest ofthe plasma nitriding furnace systems (Fig. 1). It consists of a simple vac-uum chamber encapsulated with a water jacket such as that found in a

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

Fig. 1




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

• Vacuum pumping system• Power source• Gas supply• Process controller

Process ParametersPlasma 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 bya personal computer/programmable dedicated logic controller (PC/PLC)system.

Cold-Wall FurnaceThe primary component of the cold-wall furnace is the furnace process

chamber, constructed much the same as a conventional vacuum furnaceand consisting of inner and outer vessels. The inner vessel, or vacuumvessel, is usually fabricated from stainless steel, and the outer water jacketis usually manufactured from carbon steel. A water-cooling area betweenthe two vessels conducts any heat losses from the inner vacuum vessel tothe water and to a heat exchanger. The vessel sidewall usually is fittedwith 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 wellas the thermocouple feedthrough. The power feedthrough is designed toallow continuous power flow to the cathode feedthrough.

Plasma Generator Power PackThe plasma power generator is usually a solid-state unit designed to

produce continuous dc power from line voltage through to the variable



voltage control to the furnace cathode power feedthroughs situatedwithin the process chamber. Because the power feedthrough allows con-tinuous power from the plasma generator, its insulating characteristicsmust 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 thecathodic workpiece.

Some older power generator designs used a straight dc power system.Glow stability was sometimes a problem when plasma nitriding complexpart geometries and holes, particularly blind holes. This led to the use ofhigh-process voltages in order to address the nitriding of complex geome-tries, which in turn created problems arising from the high voltage in rela-tion to the arc discharge region on the Paschen curve (discussed later in thischapter). Should the arc be initiated, there is a serious risk of both metallur-gical and mechanical damage to the part, caused by overheating, possibleburning, and possible stock removal at sharp corners on the workpiece.Therefore, some form of arc detection and suppression electronic-controlsystem is mandatory.

The power supply is used to set up a bias between the workpiece andthe vacuum chamber wall. The disadvantage of these units is that whenoperating in the lower regions of the glow, it is difficult for the glow topenetrate along the form of the part being treated, particularly in partswith blind holes or complex geometries. In such cases, both pressure andvoltage must be varied; however, if high voltage and pressure levels areselected, there is a serious risk that the arc discharge region could bereached and arcing could occur.

Heating ElementsA cold-wall furnace normally has no heating elements. Heat into the

part is generated by the kinetic energy developed by the ionic bombard-ment and is controlled simply by voltage and current density regulation.In some cases the furnace manufacturer will design a furnace with supple-mentary elements to assist the plasma heating. These elements wouldmost likely be found within the furnace process chamber and usually areelectrically isolated to prevent them from being nitrided.

Furnace ThermocouplesPerhaps even more important than control of process temperature is

control of part temperature. Unlike more conventional heat treatmentmethods, temperature generally is measured at the part rather than thechamber. In conventional heat treating, the temperature generally is meas-ured by a thermocouple located within the process chamber. It is oftenincorrectly assumed that what the thermocouple is measuring is the cham-ber temperature. However, the thermocouple measures only the tempera-ture at the point of the thermocouple. It does not measure temperature atany other point within the process chamber.

Chapter 9: Ion Nitriding Equipment / 91


Many factors within the furnace chamber influence temperature unifor-mity. Therefore, part temperature—not chamber temperature—must bemeasured. The part and process temperature are measured by consideringthe thinnest part and the thickest part within the process chamber. The parttemperatures are usually held to within a tolerance band of ±5 °C (10 °F).

If the thermocouple cannot be attached to the workpiece, then it shouldbe attached to dummy test coupons that are representative of the work-piece cross-sectional area and the material being treated. The thermocou-ple must be isolated from the furnace anode.

As in conventional process temperature measurement, the signal fromthe thermocouple is generated in millivolts and transmitted back to eithera process controller or a data logger to record the process control parame-ters. The signal transmission method can be either conventional hard wireor fiber optics.

Temperature uniformity throughout the process chamber is mandatoryto ensure uniform surface metallurgy, surface hardness, core hardness, and case thickness. Without temperature uniformity, serious metallurgicalconditions can result.

Gas FlowGood process gas-flow input control is critical. Some of the earlier sys-

tems used flowmeters. Although this method worked to some extent, itwas 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-flowcontroller, an electronic device that allows precise flow control. Use of themass-flow controller is not limited to plasma nitriding; it can also be usedin gas nitriding, particularly when using the dilution method.

Gas flow can affect nitriding quality. The required gas flow remainsconstant only if the work surface area also remains constant. Most heat-treat shops cannot guarantee constant same-surface area loads; therefore,the gas flow requirements will vary from load to load, and will vary inrelation to the type of surface metallurgy required.

Depending on the process retort size, the usual gas flow consumptionrate 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 con-sumption 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 PumpChamber pressure is controlled by a simple mechanical vacuum pump, a

rotary-vane vacuum pump (Fig. 2, 3), or a combination system of a mechan-ical pump and a roots blower. With an operating pressure of approximately10 to 500 Pa (1.4 × 10–3 to 7.3 × 10–2 psi or 0.075 to 3.75 torr), the rootsblower improves the system pumping speed and compression. Lower-vac-uum diffusion pumping systems and cryogenic systems would be used only



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


when a very clean inner chamber is required and low residual gas values arenecessary.

Pump sizing is an important aspect of furnace design, as this will deter-mine the pumpdown time of the process chamber. There is no point inoversizing the vacuum pump to pump down in less than 10 min.

The vacuum pump-out port on the furnace chamber is usually located inthe hearth. The furnace hearth and the vacuum pump are connected by aflexible stainless steel connector.

Maintenance of the vacuum pump is critical but simple. The vacuumpump oil must be checked weekly, and changed every 3 to 6 months. Thecorrect gas ballast setting is mandatory. Remember, one of the processgases is hydrogen, which is a very soluble gas as well as highly flammableand explosive if mixed in the correct combination with oxygen. If the gasballast setting is done incorrectly, an explosion or fire could result. Note:Refer to the manufacturer’s operating and maintenance manual for thecorrect method of setting the vacuum pump gas ballast. If the vacuumpump should fail, the unit will not plasma nitride and production will stop.

Cathode and AnodeThe vacuum vessel acts as the anode potential, and the furnace hearth is

attached to a specially designed and insulated power feedthrough, whichin turn is connected to the dc power source. The principal concern with the


power feedthrough is that it be insulated with a very dense ceramic mate-rial with high insulation characteristics. The insulator is then covered witha 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 somerespects to the cold-wall continuous dc system. The major physical differ-ence between the two is that the hot-wall system is fitted with an insulatedheating bell furnace around the process chamber. This means that work-piece heating can now be separated from plasma generation, becauseplasma is not necessary to heat the parts. Plasma is used only for work-piece surface preparation and to ionize the process gas. Figure 4 shows atypical schematic layout of a hot-wall plasma furnace. Table 1 comparesthe cold-wall and hot-wall plasma nitriding systems.


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

Fig. 3



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

Fig. 4

At what temperature is plasma started?

Why is plasma started at those temperatures?

Why is there a difference in plasma generationvoltages?

How does the heatup rate compare?

Why pulse the power input?

What happens to the heat?

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

Room temperature

The cold-wall furnace uses a constant dc sys-tem, which requires plasma voltages around600 to 800 V. Mechanical and metallurgicaldamage to the workpiece surface may occur byprocessing so close to the arc discharge region.

To heat and maintain the workpiece tempera-ture, the required power (kW) corresponds to acurrent density on the workpiece of approxi-mately 10 A/m2 (1 A/ft2) at this partial pressureand voltage.

The cold wall usually requires more time forheatup.

With a constant voltage input, there is a con-stant heat output. Reducing voltages to reducetemperature changes the current density andthe cathode fall voltage distance (glow seam);therefore, other parameters also must bechanged.

With a cold-wall system the released electronis hotter than the ion. This means that the elec-tron goes back to the furnace wall (anode) andcreates heat. Heatup of the wall will continue,necessitating water cooling of the wall to dissi-pate heat.

Increasing the operating pressure causes a cor-responding increase in current density, followedby an increase in part surface temperature.

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

The hot-wall furnace utilizes a partial pressurecondition using hydrogen or nitrogen as a ther-mal conductance gas. The vacuum retort isheated only by external heaters and not byplasma voltage. This means that the input volt-age is not as high (400 to 500 V), and is awayfrom the arc discharge region.

Because the part is preheated the powerrequired to maintain the workpiece tempera-ture at this partial pressure corresponds to 1–2A/m2 (0.09–0.2 A/ft2). A lower voltage isenough to produce these currents.

The hot-wall heatup of the port is usually about15 times faster.

With a pulsed voltage, high voltage can beused without risk of overheating the part, ortaking the part to the point of arc discharge.This means that the other parameters need notbe changed.

The hot-wall furnace combined with the pulsetechnology uses external blowers to preventexcessive wall heating. The wall temperaturecan safely rise to around 650 °C (1200 °F)without concern over heat buildup.

Using the hot-wall pulsed power system withthe same pressure/temperature combinationand the same voltage/current relationship asthe cold wall, the plasma energy can be main-tained by varying the duty cycle (pulse varia-tion), even with a changing voltage and currentdensity.

Table 1 Comparison between hot-wall and cold-wall plasma ion nitriding systemsQuestion Cold wall Hot wall

Source: Ref 4


Another substantial difference is that the plasma power generationpack is now a pulsed dc generator, rather than a continuous dc unit. Atfirst, this might not seem to be a striking development. However, thePaschen curve (Fig. 5) shows that high voltages are needed to generate aplasma glow. The use of high process voltages at ambient temperatures isquite dangerous to the workpiece because of the potential for arc dis-charge. Pulse technology nullifies that risk simply by pulsing the power atsuch a frequency as to completely interrupt the power supply, thus elimi-nating the arc buildup and consequently the risk of arc discharge.

Hot-wall pulsed plasma nitriding equipment contains no internal heat-ing elements. If heating elements are installed inside the process chamber,then they will be nitrided. However, supplementary heating elements arelocated in the external heating bell furnace. The elements can be mountedeither in the insulation wall of the external bell (as with traditional meth-ods) or directly onto the external wall of the process vessel. This is a moreefficient method of preheating the process chamber and allows for betterheat 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.


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

Fig. 5


• The process chamber is backfilled with hydrogen gas, and the externalheating elements are switched on to heat the gas; the workload isheated by convection.

• Once the process temperature has reached approximately 230 °C(450 °F), the sputter cleaning procedure begins. Hydrogen ions atom-ically blast the workpiece surfaces, a more thorough and effectivemethod than aqueous cleaning.

• Holding time at the sputter clean temperature is determined by the ini-tial surface cleanliness but generally is not more than 20 to 30 min. Ifthe workpiece surfaces are badly contaminated, a gaseous mixture ofup to 10% argon and 90% hydrogen should be used. To avoid surfaceetching, do not use more than 10% argon in the mixture. Sputtercleaning is further discussed later in this chapter.

• Once the sputter cleaning operation is complete, the process chambertemperature is raised to the appropriate nitriding temperature for therequired holding time and with the appropriate gas flow to achievethe required surface metallurgy.

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


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

Fig. 6


the process retort and the water contained in the water jacket if a contin-uous dc system is used. In the hot-wall pulsed plasma system, the energyrequired provides external heating to the process retort and conductiveheat to the workload. This means that only the minimal energy neces-sary to generate the plasma glow seam is required. The continuous dcsystem, however, requires energy from the plasma and supplementalinternal elements.

Pulsed Power SupplyThe power source used in the pulsed plasma nitriding system is the

heart of the system. It must provide the operator accurate control of puls-ing time and required process voltage in order to create the plasma neces-sary for surface preparation and nitriding completion with regard to boththe steel and the part geometry. The plasma generator supply must:

• Create the physical conditions for the abnormal glow discharge (seethe 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 condi-tions is somewhat limited simply because the first three conditions arelinked together. When the user has a mixed load of different part geome-tries and sizes, it becomes difficult to handle because small parts with ahigh surface area-to-volume ratio can quickly overheat. The minimumpower input necessary for the abnormal glow discharge has to be bal-anced. For example, cooling of the chamber wall may create nonuniformtemperatures.

The cold-wall unit is usually cooled by a water jacket. This means thatheat generated in the process chamber is taken away through the water toa heat exchanger. This is a wasteful method of cooling. The hot-wall sys-tem uses a series of external blowers that draw in shop air to cool theexternal wall of the process vessel as required (Fig. 7).

Arc formation must be detected and interrupted as soon as possible. Thetime between development of the arc and its discharge is a matter of milli-seconds, and it will damage the workpiece through rapid overheating atthe point of contact. The steel may burn, or the arc discharge can causelocalized grain growth and sometimes stock removal.

Figure 8 schematically shows the power characteristics of a dc powersupply. 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 val-ues. The area below the line Ptemp is equal to the energy input necessary tobalance the energy losses of the system and to hold the temperature in theworkload at the desired value.



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

• The power should not be of a sine-wave type. That means the powerwill only be reduced and not completely shut off or isolated. It is thusnecessary to have a defined power-on, power-off system—that is, adefined square form so that the power jumps from zero voltage intothe allowed gap of the abnormal glow discharge region.

• The length of the pulse should be shorter than the development time ofthe 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 aneasy ignition for the next pulse—for example, less than 1 µs.

• The ratio of pulse to pause should be variable over a wide range tocontrol the power input by the plasma in the workload so that it will bepossible 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 duringthe pulse within the gap. The sum of the areas under each pulse is equal toarea under the line Ptemp. In the figure the pulse time (the width of thesquare wave) is constant. The pause time can vary to balance the energylosses Ptemp. By this method, the temperature adjustment is separated from


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

Fig. 7


the other process parameter. Typical pulse time values are between 5 and100 µs, while pause time can vary between 5 to 200 µs (Ref 5).

Pulsed dc power gives the furnace user three additional process controlvariables:

• 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, sothe control parameters could be described as voltage, frequency, and ratio


Power characteristics (a) Continuous dc. (b) Pulsed dc. Abnormal glowdischarge (see Fig. 5) occurs between Pmin and Pmax. The power

required for plasma nitriding, Pplas, is in this region. Ptemp is the time average ofpower required to maintain the workload at the desired temperature. In (b) thepulse widths are regulated so the area under the pulses equals the area underPtemp. Source: Ref 5

Fig. 8


of on time to off time. This amounts to variable pulsed dc power in rela-tion to time on and time off, as well as the ability to vary the chamberinternal pressure. These additional controllable variables allow the opera-tor to better control the process and to address such part geometries asblind holes, cavities, and complex shapes. Occurrences such as arc dis-charge and hollow cathode potential (discussed later in this chapter) arestill 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 bemade of PC/PLC computer technology. Considering the state-of-the-artcommunication technology in industry today, the complete furnace sys-tem and any furnace problems can be remotely monitored, saving costlydowntime, 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 isalmost completely self-diagnostic and self-managing and is able to com-pare current process information with historical data.

Work Cooling after Plasma Nitriding

There are five methods of work cooling from the process temperature toan acceptable exposure temperature after plasma nitriding. Selectiondepends on the primary design of the furnace system. The cooling methodsdescribed in this section apply to both cold-wall and hot-wall furnaces.

Free-Cool Method. Free cooling is achieved by simply turning off theplasma power and allowing the work to cool from the process temperaturedown to below 150 °C (300 °F) under partial pressure conditions. Thetime required depends on workpiece mass and surface area. The disadvan-tage of this method is that it is very inefficient. There is insufficient



process gas within the furnace chamber under vacuum conditions to alloweffective 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 thechamber to a partial pressure with a nonreactive gas—usually clean, drynitrogen (not metallurgical grade). The work will cool down somewhatfaster than with the vacuum free-cooling method because there is a con-vection gas within the process work chamber. However, there is no meansof distributing acquired heat from the workpiece into heated gas and dis-charging it to the atmosphere.

Cooling under Positive Pressure. This method of cooling relies onsufficient nitrogen being introduced into the work chamber so that thechamber pressure exceeds atmospheric pressure. Gas agitation is now byan internal recirculation fan. This method is also more efficient than thefree-cooling method.

Cooling Using a Combination of Nitrogen Backfilled Gas in Con-junction with a Water-Cooled Heat Exchanger. This method of cool-ing is the most efficient method available to the plasma nitride furnaceuser. The furnace is fitted with a finned-tube copper heat exchanger withambient-temperature water passing through the heat exchange coils as thebackfilled gas passes over the heat exchanger. Movement of the coolinggas is made possible by the internal recirculation fan.

Postoxidation Treatment. The fifth option is to cool down to a specifictemperature and conduct a postoxidation treatment (oxynitriding) toenhance the surface corrosion properties of the workpiece.

Postcooling. Once cooling has taken place, the process chamber can thenbe opened safely without risk of oxidation or discoloration of the work sur-face. The workpiece should be a matte gray color. However, if a postoxida-tion treatment has been carried out, the color of the workpiece surface will beblue 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 thehollow cathode effect, sputter cleaning, furnace loading, pressure/voltagerelationships, workpiece masking, and furnace configuration options.

Hollow CathodeHollow 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, freeelectrons (energy) are trapped in the area beneath the glow seam. The glowseam holds the free electrons within the hole. The electrons as energy thenbegin to migrate through the wall of the hole. This raises the temperature of



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

Supplementary Internal Heating ElementsSupplementary internal heating elements are sometimes used in the

cold-wall system. However, it must be remembered that everything withinthe vacuum chamber will be nitrided, including the heating elements.

The hot-wall furnace operates in a completely different manner ini-tially. The furnace is evacuated by the vacuum pump to the appropriatevacuum level and backfilled with hydrogen to a partial pressure. Theexternal hot-wall heaters are started and the part is heated by convectiongases and not by plasma energy. At ambient temperatures and high plasma


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

Fig. 9

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

Fig. 10


generation voltages, plasma energy is dangerous, as is the case with thecold-wall system.

With the hot-wall system, when the temperature reaches approximately230 to 260 °C (450 to 500 °F), the plasma generation voltage is startedand the part is then sputter cleaned under hydrogen gas up to the appropri-ate nitriding temperature.

Sputter Cleaning Sputter cleaning can be likened to “atomic shot blasting,” that is, clean-

ing by ionic bombardment. This is the procedure used to preclean the worksurfaces prior to nitriding. The sputter gas typically is hydrogen, the lightestgas, which cleans as well as acts as a reducing gas. Any surface oxide willbe reduced by the hydrogen to the base metal. If the workpiece surface isseriously contaminated, the hydrogen can be mixed with argon to increasethe gas density and its cleaning ability. Again, apply caution when usingargon to ensure that the mixture is not so severe as to cause surface etchingof the steel component (maximum ratio 90% hydrogen, 10% argon).

The sputtering time depends on the prior surface condition of the steelbeing treated. Generally, one would introduce several temperature stepsabove the initial convective heating at 230, 370, and 450 °C (450, 700,and 850 °F), then up to the final selected nitriding temperature. The hold-ing time at the selected sputter cleaning stage would be approximately 10 min (depending, of course, on how much cleaning is necessary). Notethat as the temperature increases, the sputter cleaning continues (Fig. 11).


Time versus temperature curve illustrating the sputter cleaningprocess as the temperature increases to the processing temperature

Fig. 11


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

Loading of the FurnaceFurnace loading is not complicated if the operator has a basic under-

standing of the process. The furnace can be loaded with a static fixture thatwill accommodate the particular parts to be nitrided. This assumes that theparts are of similar geometry and similar case-depth requirement. The realsecret of parts loading is to be aware of the potential for the hollow cath-ode effect. However, the part geometry can be mixed, provided that theworkpiece materials are similar in both chemical analysis and case-depthrequirements. If different case-depths are required and the steel analysesare radically different, it is not possible to mix the load.

Pressure in Relation to VoltageIt is most important to have good control over both the vacuum level

and the voltage as these two process parameters will strongly influence theplasma glow-seam coverage of the part (or, more accurately, they willdetermine the point of plasma ignition on the part). If the pressure is toohigh with a normal process voltage for the part geometry, the part willbegin to lose the plasma glow-seam coverage. If this occurs, the area with-out plasma ignition will not nitride (Fig. 12a and b).

MaskingA part can be masked quite simply by remembering the following:

What plasma can see, it will nitride. Mechanical masking involves wiringsteel shim stock material to the part. The thickness of the shim stock isusually 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 incontact with the hearth will not nitride. If it is necessary to nitride itsunderside, then the part can be mounted on steel points (Fig. 13).

If two parts are placed directly on top of each other, the two contactingfaces will not nitride and will remain soft. Holes can be masked by insert-ing a simple steel plug. For a threaded hole, the first two or three turns of astud can be screwed into it. There is no need to coat the stud threads.

Proprietary “paints” are available for masking that will resist the effectsof sputtering (unlike early paints that sputtered off the surface in particlesthat made their way to the inner vessel wall, changing the electrical char-acteristics of the anode over time). Nitriding stop-off paints will resist theeffects of sputtering if the proper application methods are followed. Com-plete coverage is mandatory, and there should be no brush marks.

Copper plating is commonly used to mask workpieces undergoing tradi-tional gas nitriding. In plasma nitriding, however, the deposited plate can



be sputtered off during the cleaning process and transported to the anodevessel, where they will coat the inner surface and affect its electrical char-acteristics. Thus, copper plating is not recommended for plasma nitriding.

Additional information on masking prior to plasma nitriding operationscan 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 choicedepends on:

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


Glow-seam coverage dependence on voltage. (a) Workpiece duringplasma 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 com-plete workpiece surface. Only the areas covered by the glow will be nitrided.Nitriding at the uncovered areas will be reduced or absent. (b) Workpiece duringnitriding with pulsed dc glow discharge. By using the pulsed dc with a repetitionfrequency of about 10 kHz, the complete surface is covered and uniform nitrid-ing results. The average power input is the same as in (a). The peak power of eachpulse is higher so that the region of an abnormal glow (Fig. 5) is applied duringthis pulse. Pulsed technology allows complete coverage of the surface with highpeak powers, but low average power input, so that workpiece overheating can beavoided. Courtesy of Plateg GmbH.

Fig. 12


Summary: Advantages of Plasma Nitriding

Plasma nitriding, particularly pulsed plasma nitriding, has become amature technology capable of processing workloads efficiently and repeat-ably. Here are some of the advantages of the pulsed plasma process:

• Environmental benefits: The process is clean and nontoxic and pro-duces no disposable effluent.


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

Fig. 13

Vertically configured plasma system. This system has two chambersor bells, so one can operate while the other is being loaded or

unloaded. Courtesy of Plateg GmbH

Fig. 14


• No fire risk: Hydrogen, used for sputter cleaning and as a dilution gas,presents no fire risk whatsoever because of the minuscule amounts ofoxygen remaining within the process chamber under vacuum conditions.

• No obnoxious smells: Neither nitrogen nor hydrogen will produce anyoffensive smells, nor any adverse skin reactions.

• Minimal distortion: Because the process can control the type of surfacemetallurgy created on the workpiece, compound zone thickness can bebetter controlled, leading to lower overall size growth. In addition, betteradvantage can be taken of the lower process temperatures. The processtemperature will act as a stabilizer, or an additional tempering proce-dure. Thus any retained austenite that might be present as a result of theprehardening and tempering procedure will be decomposed, leading tobetter dimensional stability.

• Clean work: The work surface is usually very clean. However, oxygenfrom the air can discolor the work surface if the vessel is opened at toohigh a temperature. The work surface can also be discolored by anoverly intensive sputter-cleaning program.

• Repeatable results: Consistent and repeatable results can be achievedthat suit the application.

• Elimination of nitride networks: Problems associated with nitride net-works can be overcome by manipulating the nitriding potential. Thisis achieved simply by reducing the amount of required nitrogen for theprocess.

• Process management: The system is almost self-managing, taking fulladvantage of PC/PLC combinations for better control. Preventive main-tenance can be better planned, and remote troubleshooting and processcontrol are possible. If a power failure occurs, the system can be pro-grammed to restart when power resumes.

• Operating costs: Plasma nitriding process requires minimal operatorsupervision (load/unload/program/initiate), provides good utiliza-tion of floor space, and results in reduced energy costs because ofshorter cycle times compared to traditional nitriding. The capitalcost of the equipment is higher than for gas nitriding but is offset bybetter plant utilization due to faster process cycles and more repeat-able metallurgy.

• Integration into cell manufacture: The pulsed plasma nitride furnacecan be integrated into the manufacturing line. Figure 15 shows a pre-cision gear manufacturing facility that has successfully integratedheat-treatment production requirements into the gear cutting line. Avariety of small helical and spiral gear pinions and planetary gears arenitrided. This system can also be built into a robotic handling system,completely automating the process line (though only with a singleproduct line, generally automotive components). Such automated sys-tems can considerably reduce operating costs.



REFERENCES1. D. Pye, Nitriding Techniques and Methods, Steel Heat Treatment Hand-

book, G.E. Totten and M.A.W. Howes, Ed., Marcel Dekker, 1997, p 7442. N. Harris, Oil Sealed Mechanical Pumps, Modern Vacuum Practice,

McGraw Hill, 1989, p 713. N. Harris, Oil Sealed Mechanical Pumps, Modern Vacuum Practice,

McGraw Hill, 1989, p 724. D. Pye, “Practical Nitriding” course notes, 19865. R. Gruen, Pulse Plasma Treatment: The Innovation for Ion Nitriding,

Ion Nitriding Proceedings, ASM International, 1987, p 143–147


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

Fig. 15



CHAPTER 10Nitriding in

Fluidized Beds

A FLUIDIZED-BED FURNACE SYSTEM can be used for the gasnitriding process. The fluidized-bed furnace is a unique metallurgical pro-cessing tool that enables the user to complete most heat treatmentprocesses, including surface treatments. The discussion here focuses onfluidized-bed nitriding.

Previous chapters have discussed gas nitriding, salt bath nitriding, andplasma ion nitriding. Each of these procedures is conducted at a processtemperature of approximately 500 °C (930 °F). The fluidized-bed furnaceuses ammonia gas for its nitrogen source, whereas the salt bath usescyanide. Fluidized-bed nitriding is similar in process technique to gasnitriding and similar in the method of heat transfer to salt bath nitriding.

The fluidized bed exhibits the same characteristics as a liquid, with oneexception: A fluidized bed is not wet. The technique of fluidization involvesthe disturbance of a bed of finely divided particles, which behave as a liquidwould behave. This is accomplished by passing a gas at sufficient volumeand pressure so as to separate microscopically the fine particles. If the vol-ume and pressure of the gas are too great, then the fine particles will be car-ried in the gas stream and leave the bed. Thus, gas pressure and volume arecritical. The bed does not require large volumes of gas, only that necessaryto separate the particles (Ref 1) (Fig. 1).

The particles in this instance are aluminum oxide. Good heat transfertakes place from the heating medium to the aluminum oxide particles,which in turn transfer the heat to the workpiece. When the bed is in opera-tion, its surface of finely divided aluminum oxide particles bubbles justlike water bubbles when air passes through it (Fig. 2).

Heating Method

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




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

• Nitrogen • Ammonia• Methane • Gas mixtures such as methane and • Endothermic gas nitrogen or ammonia and methane

Internal Resistance Heating. This method of heating (Fig. 4) is verysimilar to that of the immersed electrodes in a salt bath. The difference inthis instance is that the internal heating elements are usually sheathed. It isa very simple method of heating the fluidized bed and provides good heattransfer between the heating elements and the aluminum oxide.

Gas-Heated Fluidized Beds. Gas heating is usually less expensivethan electrical heating (depending, of course, on locality and geography).


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

Fig. 2 Liquid-like behavior of gas fluidized beds. Source: Ref 1


Once again, heating of the fluidized-bed process chamber can be eitherexternal or internal.

External Combustion Heating. When the fluidized-bed container isheated by an external gas combustion system, the container is manufac-tured from the same heat-resisting material as it would be if manufactured

Chapter 10: Nitriding in Fluidized Beds / 113

External resistance heating. (a) Fluidized-bed furnace with externalheating 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 offluidizing gas in a fluidized-bed externally heated by electrical resistance. Source:Ref 1

Fig. 3


for an electrical heating system (Fig. 5). The fluidized-bed container iscontained within the furnace combustion chamber, which in turn is sur-rounded by the installation material and the furnace casing. Many differ-ent 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 temperatureafter the workload has been removed from the furnace and a new loadintroduced makes it an attractive choice for nitriding. The bed does notrequire conditioning or purging when changing from one atmosphere sys-tem to another, making it productive and economical in terms of operatingcosts and throughput. The fluidized bed also permits easy change fromone process system to another (e.g., from nitriding to ferritic nitrocarbur-izing, or from carburizing to carbonitriding) (Ref 2). The one major disad-vantage is that it requires a fairly high volume of reactive gas to completethe process.

Temperature control is exactly the same as for a conventional heat treat-ing furnace. The bed temperature uniformity is usually well within ±5 °C(10 °F). Atmosphere control, particularly for nitriding where gas dissocia-tion is critical, is somewhat difficult. Fluidized-bed nitriding tends to rely


Fluidized-bed furnace with internal heating by electrical resistance ele-ments: (1) pivoting cover in two parts; (2) insulation; (3) refractory mate-

rial; (4) fluidized bed; (5) heating elements; (6) intake for fluidizing gas; (7) parts tobe treated. Source: Ref 1

Fig. 4


on volume of ammonia gas in relation to workpiece surface area. Becausethe atmosphere gas control is variable, the compound layer can be con-trolled—a critical factor in surface performance.

As with the conventional ammonia gas nitriding furnace, many gradesof steel can be processed effectively using the fluidized-bed technique.The types of steels that can be treated, including all the stainless steelgrades, will be described in Chapter 12. Stainless steels particularlyrequire good gas flow control and the appropriate gas dissociation.

The fluidized process cycle times, as in the gas nitriding process, aregoverned by the laws of diffusion. In other words, the diffusion rate of nas-cent nitrogen into the steel surface is the same for fluidized-bed nitriding as


Fig. 5 Immersed-element ceramic retort fluidized-bed. Source: Ref 1

External gas-heated fluidized bed with recuperator. The use of regen-erative burners where the exhaust gas temperature is only 200 °C (390

°F) achieves efficiencies similar to those of electrically heated furnaces. Source:Ref 1

Fig. 6


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

Oxynitriding

Oxynitriding can be accomplished in the fluidized-bed furnace muchlike it is accomplished in gas nitriding. This means that on completion ofthe nitriding cycle, controlled amounts of moisture are added to the processchamber.

The oxynitriding process forms a very thin surface oxide layer on theimmediate surface of the workpiece. This deliberately oxidized surfacelayer is resistant to some aspects of corrosion (though not all). In general,the process is used for applications where an expensive material such asstainless steel is being replaced by low-carbon steel with an enhanced sur-face condition. The oxynitriding process is gaining in popularity, particu-larly in the automotive industry, in both Europe and the United States.


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

Fig. 7


Operating the Fluid Bed for Nitriding

Startup of the fluid bed should follow the furnace operation manual ofthe furnace manufacturer. Once the fluidized bed has been brought to theprocess temperature, the bed is activated with the process gas, or a gaswith a sufficiently high flow rate to “unlock” the bed of aluminum oxide.This means that the bed is no longer “slumped,” but activated and bub-bling. Great care must be taken not to have the gas flow rate so high that itwill blow the aluminum oxide out of the bed.

Any previous atmosphere in the bed (e.g., carburizing) can be changedover quickly to nitriding. This is a major advantage of using a fluidizedbed instead of a salt bath. The only changeover necessary is the processgas, making the fluidized-bed furnace more productive and much easier tohandle. The least amount of process gas necessary to complete the nitrid-ing 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 thefloor-to-floor time is faster, since the bed recovery time is shorter thanwith a conventional furnace. The resulting metallurgy of the twoprocesses should be the same (Tables 1, 2).

Measurement of the Gas Dissociation

Measurement of the gas process dissociation is not the same as for gasnitriding. The gas dissociation measurement of ammonia in the conven-tional gas nitriding system is based on the solubility of ammonia in water.Water will absorb 70 times its own volume of ammonia, providing aneffective method of measuring the gaseous activity within the nitridingprocess chamber.

With a fluidized bed, the method is effective only if the exhausted ammo-nia can be captured and the dissociation measured. Such measurement is


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

Fig. 8


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 com-mercial heat treatment shop), and if the gas flow rate remains constant foreach load of work processed, then the gas dissociation will be different foreach load. Studies on various materials and various load surface areasmust be made to ensure a reasonable chance of consistent and repeatablenitriding.

ACKNOWLEDGMENTGrateful acknowledgment is given to Ray Reynoldson of Quality Heat

Treatment Pty Ltd., Turbo Drive, North Bayswater 3153, Australia, for hisvalued assistance with this chapter.

REFERENCES1. R.W. Reynoldson, Theory and Practice, Heat Treatment in Fluidized

Bed Furnaces, ASM International, 1993, p 3–92. C. Dawes and D.F. Tranter, Nitrotec Surface Treatment Technology,

Heat Treat. Met., Vol 12 (No. 3), 1985, p 70–76

Table 2 Depth of compound layer after fluidized-bed nitridingCase depth

mm in.

Material min max min max

Carbon and low-alloy steels 0.0038 0.03 0.00015 0.001Tool and die steels (structural) 0.003 0.013 0.0001 0.0005Tool and die steels (cutting) 0.003 0.0001 ...

Corrosion- and heat-resistant steels 0.0038 0.03 0.00015 0.001Ductile, malleable, and gray cast iron 0.0038 0.03 0.00015 0.001Powder metal products (ferrous) 0.0038 0.0 0.00015 0.001

Source: Ref 1


Table 1 Recommended fluidized-bed nitriding proceduresRecommended time Temperature

Material min max °C °F

Carbon and low-alloy steels 1 h 2 h 580±5 1075±10Tool and die steels (structural) 30 min 3 h 540–580 1000–1075Tool steels (cutting) 5 min 1 h 540–580 1000–1075Corrosion- and heat-resistant steels 1 h 2 h 580±5 1075±10Ductile, malleable, and gray cast iron 1 h 4 h 580±5 1075±10Powder metal products (ferrous) 30 min 2 h 580±5 1075±10

Source: Ref 1


CHAPTER 12Steels For Nitriding

MANY STEELS are commercially nitrided:

• Aluminum-containing low-alloy steels, including the Nitralloy groupwith 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 relationto the operating environment of the part that will not only ensure goodnitriding results but will also be cost effective, easy to machine or fabri-cate, and functional. Several questions must be addressed during theselection process:

• What is the product to be manufactured and how complex is the partgeometry?

• Under what type of operating conditions will the workpiece operate?Will there be compressive loads, cyclic loads, impact loads, or tensileloads?

• 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?





Table 1 Compositions of selected nitridable steelsComposition, %

Alloy steels(a) C Cr Mo Si Mn Ni V

SAE 4137 0.35 1 0.2 0.25 0.8 ... ...SAE 4142 0.42 1 0.2 ... ... ... ...SAE 4140 0.40 1 0.2 0.25 0.85 ... ...SAE 4150 0.5 1 0.2 ... ... ... ...28 Ni Cr Mo V 85 0.3 1.3 0.4 ... ... 2 0.132 Ni Cr Mo 145 0.32 1 0.3 ... ... 3.3 ...30 Cr Ni Mo 8 0.3 2 0.4 ... ... 2 ...34 Cr Ni Mo 6 0.34 1.5 0.2 ... ... 1.5 ...SAE 4337 0.38 0.8 0.4 ... ... 1.5 ...SAE 4130 0.26 1 0.2 ... ... ... ...

Low-alloy steels C Si Mn P Cr Mo Ni V Al

Nitralloy 0.20–0.30 0.10–0.35 0.40–0.65 0.05 max 2.90–3.50 0.40–0.70 0.40 max ... ...Nitralloy M 0.30–0.50 0.10–0.35 0.40–0.80 0.05 max 2.50–3.50 0.70–1.20 0.40 max 0.10–0.30 ...Nitralloy 135 0.25–0.35 0.10–0.35 0.65 max 0.05 max 1.40–1.80 0.10–0.25 0.40 max ... 0.90–1.30Nitralloy 135M 0.35–0.45 0.10–0.35 0.65 max 0.05 max 1.40–1.80 0.10–0.25 0.40 max ... 0.90–1.30

Special-purpose tool steels C Si Mn Cr Mo Ni V W

F2 1.45 ... ... 0.3 ... ... 0.3 3.0L6 0.55 ... ... 1.1 0.5 1.7 0.1 ...

Dimensionally stable tool steels(b) C Si Mn Cr Mo Ni V W

D2 1.55 ... ... 11.5 0.8 ... 1.0 ...D3 2.0 ... ... 12.0 ... ... ... ...A2 1.0 ... ... 5.0 1.0 ... 0.2 ...O1 0.95 ... 10.5 ... ... 0.1 0.5 ...O2 0.9 ... 20.4 ... ... 0.2 ... ...D6 2.1 ... ... 11.5 ... ... 0.2 0.7D2 1.65 ... ... 11.5 0.6 ... 0.1 0.5S1 0.59 ... ... 1.1 ... ... 0.2 1.9

Hot-work tool steels C Cr Mo Ni V W Co

H12 0.36 5.2 1.4 ... 0.4 1.3 ...H13 0.4 5 1.3 ... 1 ... ...H11 0.4 5 1.3 ... 0.6 ... ...H21 0.3 2.7 ... ... 0.4 8.5 ...H19 0.4 4.3 0.4 ... 2 4.3 4.3H10 0.32 2.8 2.8 ... 0.5 0.3 ...

High-speed steels C Cr Mo V W Co

T5 0.75 4 0.6 1.6 18 9.5T4 0.8 4 0.7 1.6 18 5T1 0.75 4 ... 1 18 ...T15 1.5 5 ... 5 12.5 5M42 1.08 4 9.5 1.2 1.5 8M41 0.92 4 5 1.8 6.5 5M3 1.2 4 5 3 6.5 ...M2 0.87 4 5 1.8 6.5 ...M2 1.0 4 5 1.8 6.5 ...M7 1.0 4 8.7 2 1.8 ...M1 0.83 4 9 1.2 1.8 ...

(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 prehardenand 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 searchfor the appropriate steel can begin. If several steels are suitable for a par-ticular application, price and availability become additional considera-tions (Ref 1).


Requirements for a Nitriding Steel

In the early years of nitriding, Adolph Fry at Krupp Steel (Ref 2) recog-nized that certain steels responded better from a metallurgical standpointin terms of surface hardness, core hardness, distortion, cycle time at tem-perature, and the formation of stable nitrides. Fry discovered that certainelements will respond more readily than others to form stable nitrides dur-ing the nitriding process, and this led to the development of the Nitralloygroup of steels (see Table 1 for compositions). Of the alloying elementscommonly used in commercial steels, aluminum, chromium, vanadium,tungsten, and molybdenum are beneficial in nitriding because they formnitrides that are stable at nitriding temperatures. The effects of specificalloying elements are discussed later in this chapter. Figures 1 and 2 showthe influence of alloying elements on hardness after nitriding and depth ofnitriding.

Aluminum will form very hard nitrides in the nitrided steel surface.Generally the maximum amount of aluminum permitted in the steel is inthe region of 1.5%. Above 1% Al will lead to surface cracking underextreme surface load conditions. This is because the core hardness of thematerial is usually very ductile. If a highly ductile workpiece undergoessevere loading, then there is a strong possibility that the surface of the casewill lead to crack propagation.

Chapter 12: Steels for Nitriding / 127

Effect of alloying elements on hardness after nitriding. In steels contain-ing 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

Fig. 1


Molybdenum will form stable nitrides at the nitriding temperature andwill 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 makesthem more difficult to nitride. Chromium reacts with oxygen to form achrome oxide barrier on the surface, which must be broken down by depas-sivation in order for nitriding to be effective. The higher the percentage ofavailable chromium at the steel surface, the more difficult the steel will be tonitride. The positive side of this is usually high surface hardness values.

Vanadium in a nitriding steel also is conducive to the formation of sta-ble nitrides. In addition, fine grain toughness will be exhibited within theformed case.

Tungsten enables the steel to retain its hardness at high operating tem-peratures with no loss of surface hardness. Depending on the tungstencontent and the general composition, the nitrided steel is able to operate attemperatures up to 590 °C (1100 °F) with enhanced wear characteristicsand no appreciable loss of surface hardness.

Silicon is considered to be a good nitride former. Though it is usuallypresent as either an oxidizer or a stabilizer, silicon generally is not of suffi-cient volume to be considered a strong nitride former.

Summary. As stated in Chapter 1, all steels will nitride. Steels that con-tain the above alloying elements will readily form stable nitrides. Steelsthat do not contain those elements, such as the mild steels and low-carbon


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

Fig. 2


steels, will also nitride but will have lower surface hardness because thecase formation is limited to pure iron nitride. However, the corrosionresistance of low-carbon steels will be greatly enhanced, successfullywithstanding a minimum 100 h salt spray test. Studies have shown thatwith a deep case (90 h cycle on gas nitriding), low-carbon steels exhibitedresistance 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 adverseeffects on corrosion resistance. Hardness values are generally in the rangeof 1000 HV or more, depending on the nitriding method. With ion nitrid-ing and control of the nitriding potential, some hardness values greaterthan 1400 HV have been achieved.

Austenitic stainless steels are perhaps the most difficult to nitride. Thefollowing types have been successfully nitrided using gas, plasma, andsalts: 301, 302, 303, 304, 305, 309, 310, 316, 321, and 347. However,when nitriding these steels at conventional nitriding temperatures, corro-sion resistance is seriously impaired, in some cases up to 1000%.

The material should be in the annealed condition, which will reduce therisk of blistering or flaking. The oxide film must be removed prior tonitriding, 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-basedsolution as a cleaning agent. Once depassivated, the surface should not betouched by hand; deposition of body oil from fingerprints will inhibit thenitriding effect at the point of contact.

Remember, if the core is annealed, the surface can accommodate onlyabrasion resistance. The core will be too soft to support any kind of con-stant or cyclical load on the surface. If a load is applied, the case likelywill collapse, crack, and flake.

Martensitic stainless steels will nitride without exception. This groupcan be preheat treated to give a supportive core for abrasive and impactapplications, as well as torque loads.

Once again, remember that the surface must be depassivated beforenitriding. Corrosion resistance will be adversely affected but may be pro-tected by using lower-than-normal nitride processing temperatures. How-ever, 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 considerationsapply to this group of steels as apply to the hardenable martensitic stain-less steels.

Cycles for Gas Nitriding of Stainless Steels. In general, the single-stage process is used with a process temperature in the region of 490°C(925 °F) (depending on steel composition and preheat treatments), withtime at temperature ranging from 24 to 48 h with fairly low dissociation



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

Example 1: Plasma Nitriding of AISI Type 422 Stainless SteelSurface modification of type 422 stainless steel was accomplished

using the pulsed plasma nitriding process. The workpiece design calledfor an operational service temperature up to 645 °C (1200 °F) and the abil-ity 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 0.25Manganese 1.00Phosphorus and sulfur 0.025 maxSilicon 0.7Chromium 12Nickel 0.8Molybdenum 1.10Vanadium 0.25Tungsten 1.1

The same elements that are strong carbide formers (tungsten, molybde-num, and vanadium) are also strong nitride formers. As a result, the mate-rial was suitable for nitriding.

The preheat treatment condition of the core was tempered martensitethat had been hardened and tempered to 340 to 360 HV (~35 to 37 HRC).This was accomplished via the following heat treatment procedure:

Process Condition

Preheat 1 425 °C (800 °F)Preheat 2 760 °C (1400 °F)Austenitize 1035 °C (1895 °F); 32 mm (1.25 in.) section soaked

for 60 min at the process temperatureQuench Air cooledStabilization Cryogenic treatment using liquid N2 (approximately

–70 °C, or –95 °F)Temper 595 °C (1100 °F) for 2 h at the process temperature

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



Characteristic Required value Actual value

Effective case depth, mm (in.) 0.2 (0.008) 0.2 (0.008)Case hardness, minimum 960 HV (65 HRC, 960 HV (68 HRC)

approximate)Compound zone maximum γ ′, 0.005 (0.0002) 0.0025 (0.0001)

mm (in.)

Plasma nitride process parameters used to achieve these results were:

Process Parameter

Sputter clean time 1.5 hSputter clean voltage 650 VSputter retort pressure 100 Pa (0.015 psi)Process temperature 525 °C (975 °F)Cycle time 16 h at process temperatureOperating pressure 300 Pa (0.05 psi)Operating voltage 500 VCooling Partial-pressure N2

Summary Results. The microhardness surveys of four cycles of pulsedplasma 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 sur-face 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 gasnitriding process, due to a slight roughening of the surface. This necessi-tated two further grinding operations. The deterioration in surface finishafter pulsed plasma ion nitriding was 13.3%. This was within the acceptedsurface finish requirement, requiring no further machining. Thus, in someinstances it is not necessary to grind after plasma nitriding, and the work-piece can be used directly.

The part was thermally sensitive to distortion. Precise measurementswere taken prior to pulsed plasma nitriding and the growth/distortion wasdetermined to be within the dimensional tolerance levels. The computer-controlled system regulated not only temperature levels, but also currentdensity, process pressure, gas flows, gas ratios, and power levels.

Example 2: Nitriding of AISI Type 440A Stainless SteelA martensitic type 440A stainless steel specimen was nitrided using

the pulsed plasma ion process in order to establish the minimum nitridingtemperature, bearing in mind that high as-quenched hardness values for



martensitic stainless steels begin to decrease at temperatures as low as150 °C (300 °F).

Typical composition of type 440A is:

Element Wt%

Carbon 0.7Manganese 0.85Phosphorus 0.040 maxSulfur 0.030 maxSilicon 0.9Molybdenum 0.75Chromium 17

The preheat treatment of the steel to achieve the core properties andhardness values was:

Preheat 1 370 °C (700 °F)Preheat 2 760 °C (1400 °F)Austenitize 1015 °C (1860 °F) with a 30 min soak at part temperature Quench OilStabilize Cryogenic treatment using liquid N2 for 20 minTemper 190 °C (375 °F)


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

Fig. 3


The nitriding specification was:

Characteristic Required value

Case depth 0.0375 mm (0.0015 in.) maxCase hardness 800 HV (~64 HRC)Compound zone 0.0025 mm (0.0001 in.) max γ ′

The process parameters were:

Process Parameter

Sputter time 1 hSputter voltage 650 VSputter retort pressure 150 Pa (0.025 psi)Process temperature 190 °C (375 °F)Cycle time 16 hOperating pressure 300 Pa (0.05 psi)Operating voltage 450 VCooling Partial pressure with clean, dry N2

Summary Results. The process temperature of 190 °C (375 °F) was toolow for effective nitrogen diffusion. Diffusion did occur, but not suffi-ciently to form a commercially usable case. The tempering curve was fur-ther examined for a comparable martensitic grade containing a slightlyhigher carbon content (AISI type 440C with 1.10% C) (Fig. 4). Based on


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. At260 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

Fig. 4


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 rapiddeterioration 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 1 hSputter voltage 650 VSputter retort pressure 150 Pa (0.025 psi)Process temperature 370 °C (700 °F)Cycle time 16 hOperating pressure 300 Pa (0.05 psi)Operating voltage 450 VCooling Partial pressure using clean, dry N2

Final case depth 0.05 mm (0.002 in.)Final case hardness 896 HV (~67 HRC)

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


Reduction in corrosion resistance after nitriding of type 422 stainlesssteel

Fig. 5


Example 3: Nitriding of AISI Type 630 (17-4 PH) Stainless SteelType 630 stainless steel is widely used in the aircraft and aerospace indus-

try for gear manufacture and other critical performance items. It is known asa precipitation hardening steel, providing good tensile strength and impactvalues when heat treated by solutionizing and precipitation treatments.

Pulsed plasma ion nitriding technology was considered for this steelbecause of its ability to provide controlled, repeatable metallurgy withoutaffecting the core. Ion nitrided type 630 also maintains good corrosionresistance.

Typical composition of type 630 is:

Element Wt%

Carbon 0.07 maxManganese 0.89Silicon 1.00Nickel 3.8Copper 3.9Niobium 0.3Tantalum 0.4Chromium 16.5

The preheat treatment consisted of:

Preheat 1 315 °C (600 °F)Preheat 2 455 °C (850 °F)Solutionize 1045 °C (1910 °F) (soak for 20 min at temperature)Quench OilPrecipitation harden 480 °C (905 °F)Final hardness 423 HV (~43 HRC)

The nitride process specification was:

Case depth required 0.05 mm (0.002 in.) maxCore hardness required 830 HV (65 HRC) minCompound zone required 0.005 mm (0.0002 in.) max

The plasma nitride process parameters were:

Process Parameter

Sputter time 1 hSputter voltage 600 VSputter retort pressure 150 Pa (0.025 psi)

(continued)



Process Parameter

Process temperature 480 °C (900 °F)Cycle time 4 h (at temperature)Operating pressure 300 Pa (0.05 psi)Operating voltage 450 VCooling Partial pressure using

clean dry nitrogen gas

Summary Results. The microhardness survey (Fig. 6) showed a surfacehardness of 960 HV that diminishes into the material core below the formedcase. 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 aslight increase in the core hardness of almost 1 HRC. The pulsed plasmanitriding process acted as an additional precipitation treatment (Ref 4).

Further cycles were made with similar metallurgical results. It was fur-ther observed microscopically that a much denser structure of ferritestringers emerged in the martensite matrix using Vilella’s reagent (5 mLHCl plus 1 g picric acid plus 100 mL ethanol).

Plasma Nitride Case Depths

The following temperature factor values are based on the Harris for-mula (Ref 5):

Case depth = Square root of time × Temperature factor


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

Fig. 6


Suggested temperature factors for the process cycle time in relation to therequired case depth are:

Temperature

°C °F Temperature Factor

460 865 0.00221470 875 0.00233475 885 0.00259480 900 0.00289500 930 0.0030510 950 0.0033 515 960 0.0035525 975 0.0037540 1000 0.0038

The factors are based on the steel at the selected process time. They donot pertain to nitriding of the higher alloyed steels such as stainlesssteels. The diffusion rate of nitrogen into the steel surface will reducedramatically as the alloy content increases. The factors also do not con-sider the furnace loading and load density, and positioning in relation topotential shielding of the work (e.g., hollow cathode in the process cham-ber). The factors are based on a simple nitriding steel without the addi-tion of aluminum.

The cycle times will be approximate, and serve only as a guide to thecycle time for a particular load. The process technician should keep arecord 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

REFERENCES1. 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



3. K.-E.Thelning, Nitriding, Steel and Its Heat Treatment, 2nd ed., But-terworths, 1984, p 492–544

4. D. Pye, Pulsed Plasma Ion Nitriding and Its Effects on the SurfaceModification of Stainless Steels AISI 422, 440A and 630, SurfaceModification 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



CHAPTER 13Control of the Process Gas in

Plasma Conditions

PROCESS GAS CONTROL for plasma (ion) nitriding is a matter ofestimating the flows necessary to accomplish the required surface metal-lurgy. Conventional gas nitriding systems measure the dissociation of theammonia process gas. This is not quite so simple with plasma systemsbecause the process is operating under partial pressure conditions, makingit difficult to introduce an effective sampling system into the unit. Com-pounding the problem is the fact that the process gas is at cathode poten-tial and at the work surface.

This chapter reviews several studies aimed at better understandingprocess gas control in plasma nitriding and its influence on compoundzone formation. Emphasis is placed on the effect of sputtering on thekinetics of compound zone formation. Additional information on gasratios 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 cred-ited to N. Ryzhov of Moscow State University, who developed whatappeared to be a workable control system of the gas species activity of theplasma glow seam (Ref 1). The work, which has not yet been commercial-ized, is based on the solubility limit of nitrogen in iron and the amount ofatomic nitrogen available for diffusion. The gas is evaluated by line-of-sight through an observation port in the furnace process chamber.

The system uses photo spectrometry as the principle of operation. Theobservation unit, sighted onto the plasma glow seam, develops an electricalsignal that is then transmitted to a personal computer/programmable logiccontrol (PC/PLC) combination (Fig. 1). This controls the gas-delivery system




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

Analysis by Mass Spectrometry

Szabo and Wihelmi discussed mass spectrometric diagnosis of the sur-face nitriding mechanism in a direct-current (dc) glow discharge (Ref 2).Their hypothesis was that it makes no difference whether the nitrogensource is ammonia, or nitrogen and hydrogen. Use of hydrogen, theystated, is an important function in the nitriding process. The hydrogen acts


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

Fig. 1


as a reducing gas to reduce (with heat) surface oxides on the steel and,perhaps more importantly, to influence and regulate the composition ofthe compound zone (white layer) as shown in Fig. 2.

They also investigated the ability of alloying elements to form nitridesin the steel surface, concluding that the following elements will readilyform 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 ana-lyze the gas activities and surface reactions taking place at the steel sur-face. This meant that the solubility limit of nitrogen in iron could be notonly observed, but also controlled.

Difficulties Associated with Gas Analysis

Because the plasma nitriding process is under vacuum conditions, it isdifficult to evaluate accurately the gaseous species activities within the ion-ized gas glow seam and, more importantly, the amount of nitrogen diffusioninto the steel surface in relation to the solubility limits of nitrogen in iron.The Russian method of glow seam observation and analysis of the gaseousactivities seemed to hold the most promise as a method of accurately con-trolling not only the solubility limit of nitrogen in iron, but also the solubilitylimit of carbon in austenite.

Control of the gas activities has eluded scientists and metallurgists inthe field of pulsed plasma ion nitriding. Using sensors to measure gas

Chapter 13: Control of the Process Gas in Plasma Conditions / 141

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

Fig. 2


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

Kinetic Studies

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

Conclusions

At present, control of the quality of the surface metallurgy and the for-mation of the nitride diffusion zone during plasma nitriding requires care-ful 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. Rolinski and G. Sharp

The role of sputtering in plasma nitriding has been a subject of manystudies in the last 35 years (Ref 5–19). A qualitative approach to analysis ofthe 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 lin-ear removal rate of the surface atoms due to sputtering. The first numericalanalyses of these processes were done by Marciniak (Ref 13, 14) and Sunand Bell (Ref 19). The calculated and experimentally verified SRs for sam-ples 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



model developed by Sun and Bell (Ref 19) allowed precise calculations ofthe compound layer growth for a specific steel, the specific nitriding condi-tions, and the assumed SR. The assumed values of SR were between 0.1and 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 bedescribed 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 isSR, c is a coefficient of compound zone growth due to diffusion, and x isnitriding time. The experimental results of Edenhofer (Ref 8) for thekinetics of the compound zone formation on En 9 (AISI 4142) steelplasma nitrided at 450, 530, and 570 °C (840, 985, and 1060 °F) were fit-ted to Eq 1 using the TableCurve 2D software (Ref 20) and are presentedin Fig. 3. The fit of the data for the available 45 h range is very good, sincethe 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), 0.124µm/h for 530 °C (985 °F), and 0.427 µm/h for 570 °C (1060 °F). Thegraph extrapolated to 100 h of nitriding time shows the tendency of thecurves: the maximum, respectively, at approximately 65, 50, and 35 h ofnitriding and the diminishing values thereafter. At the same time, thegrowth due to diffusion (the coefficient c) is larger for higher temperaturesand smaller for lower nitriding temperatures.

Similar graphs based on the experimental data of Marciniak (Ref 14) for36 H3M (0.36C-3Cr-0.6Mo-0.6Mn-0.27Si) steel nitrided at 530 °C (985 °F)


Compound zone thickness versus nitriding time for 42Cr Mo4 (AISI4142 steel) plasma nitrided 3.3 mbar in the atmosphere of 25% nitro-

gen + 75% hydrogen at 570 °C (1060 °F) (upper curve), 530 °C (985 °F) (middlecurve), and 450 °C (840 °F) (bottom curve) based on the experimental data ofEdenhofer (Ref 8, 24). The graph is extrapolated over the original limit of 45 h. Thefit 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.

Fig. 3


are presented in Fig. 4. The curve with a well-distinguished maximum atabout 16 h drawn for nitriding with 50% nitrogen and 50% hydrogen repre-sents 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 com-pound zone and an SR of about 0.036 µm/h. The “diffusion” fraction of thekinetic equation, which represents a nitriding potential, is higher for thesamples nitrided with 50% nitrogen and 50% hydrogen than for samplesnitrided with 15% nitrogen and 85% hydrogen. However, all of the abovepredictions for the range of time exceeding the experimental data range maynot be accurate since extreme caution is advised in relying on polynomialsfor extrapolations and foreasts beyond the range of the dataset (Ref 20).

The diminishing value of the compound zone thickness during a longnitriding time may have important practical meaning; the compound zonecould eventually disappear completely and therefore the nitriding ratewould drop down and/or a denitriding of the steel could take place (Ref 9).Wells and Strydom suggested that the compound zone growth might alsobe affected by redeposition of the sputtered material (Ref 16). This phe-nomenon could be enhanced by oxygen, which may always be present in asmall quantity in industrial systems, and, therefore, the compact portion ofthe compound zone can be significantly reduced by formation of theoxynitride (Ref 16).

Consequently, in our studies, we researched the actual kinetics of com-pound zone formation in long nitriding with an aim to establish the SR insuch a cycle. The 3% Cr-Mo-V steel we investigated is used in the gearindustry to achieve an exceptionally deep case. It was then very important


Compound zone thickness versus nitriding time for 36H3M 3% Cr-Mosteel plasma nitrided at 530 °C (985 °F) in the atmosphere of 50%

nitrogen + 50% hydrogen (upper curve) and 15% nitrogen + 85% hydrogen (bot-tom curve) based on the experimental data of Marciniak (Ref 14). The graph isextrapolated over original limits of 16 and 36 h, respectively. The fit equations arey = 1.275 – 0.425x + 3.295√x, y = 0.5 – 0.036x + 0.479√x, and r2 = 0.994 and0.986, respectively.

Fig. 4


to verify if the thickness of the compound zone would become thinner tothe 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. TheBrinell 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. Allsamples were blasted with 180 grit aluminum oxide before nitriding. Theequipment and processing details are described elsewhere (Ref 21). Thenitriding was carried out at 540 °C (1000 °F) nominal temperature, andthe nitriding times were from 4 to 400 h. The samples were treated in adirect 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 met-allographic 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 asingle measurement was ±0.1 µm. There was a minimum of five samplesused in each run. The x-ray diffraction phase analysis was performedusing Cr Kα radiation.

Results and Discussion

The compound zone thickness changes with the nitriding time, asshown in Fig. 5. A possible maximum value of approximately 13.5 µm is


Compound zone thickness versus nitriding time for 3% Cr-Mo-V steelplasma 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 representnormal distribution and standard error (small interval) at 95%.

Fig. 5


achieved after about 250 h of nitriding, and the final value after 400 h isabout 13 µm. Micrographs of the compound zone are presented in Fig. 6.They do not reveal any significant presence of the vapor-deposited, possi-ble oxygen-contaminated layer, as was suggested by Wells and Strydom(Ref 16). The top portion of the compound zone contains some porosityand probably a layer of the nitrides deposited from plasma; however, itsmain fraction stays very compact.

Optical microscopy of the microsections also revealed a presence of fre-quent conical structures at the surface. Similar features were also observedby others (Ref 22). The intensity of these peaks was higher on samplesnitrided longer. The x-ray diffraction showed that initially after 4 h of


Optical micrographs of surface layers produced on 3% Cr-Mo-V steelby 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

Fig. 6


nitriding two iron nitrides, ε (Fe3N) and γ′ (Fe4N), were formed on the sur-face (Fig. 7). The intensity of the ε patterns is quickly reduced and, after 25h 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 for-mation was promoted by carbon present in the steel. It is known that car-bon stabilizes the ε-carbonitride and that it can diffuse outward from thesteel during nitriding (Ref 9, 17, 23, 24). In the short nitriding processes,carbon atoms were diffused toward the surface and, by reacting with ironand nitrogen, helped in the formation of the ε-phase. During the long expo-sure to the plasma, the surface was decarburized as carbon was sputteredaway and replaced by nitrogen. With a lack of carbon atoms, the nitridingpotential of the plasma was shifted toward γ′; the lower nitrogen phase andthe ε-phase disappeared completely.

The results of the compound zone thickness versus nitriding time stud-ies clearly demonstrate the effect of sputtering: a possible maximum valueat about 250 h and the diminishing values of the compound zone there-after. The calculated SR is 0.0295 µm/h, and the coefficient c of com-pound zone growth due to diffusion of nitrogen is 0.933. The total compound zone thickness a, formed during ramp-up to the final tempera-ture, is about 6 µm. The a value depends on growth due to diffusion ofnitrogen and sputtering. Since sputtering was taken into account onlyfrom a time when the final temperature was reached, the a value wasaffected by an error of not counting the material removed. In fact, thisvalue is only about 0.06 µm if we assume that SR was the same duringramping and the final soaking.


X-ray diffraction patterns of surface layers produced on 3% Cr-Mo-Vsteel 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

Fig. 7


The SR of 0.0295 µm/h from this experiment agrees very well with thevalue of 0.036 µm/h calculated by us for the literature data of nitridingperformed in the atmosphere of 15% nitrogen and 85% hydrogen on simi-lar steel (Ref 14). However, our kinetic studies as well as the studies basedon the results of others (Ref 8, 9, 14, 15) could be affected by not takinginto account the fact that a phase composition of the steel surface maychange 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 rap-idly; in fact, it took many hours to produce a “pure” γ ′-nitride.

It can then be concluded that the sputtering yield depends not only onpressure, temperature, gas composition, and the plasma power density, butmainly on the phase composition of the compound zone that formed onthe surface. Some of these parameters were probably different in ourexperiment than in the experiments carried out by the other researchers(Ref 8, 14, 19, 24, 25); however, as long as we consider γ ′ compound zoneformation, its SR stayed low.

In comparatively short nitriding cycles with a sufficiently high nitro-gen content carried out by others (Ref 8, 14), a mixture of the γ ′ and εwas produced and, consequently, the SR was much higher than when the“pure” γ ′-phase was produced. This can be seen in Fig. 8, which repre-sents the nitriding kinetics for shorter (up to 25 h) cycles. The graph isnot biased by the phase composition change toward a pure γ ′; the SR isabout 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 thecoefficient b equal to zero to see the effect of the remaining (diffusion)fraction of the equation on the kinetic.


Compound zone thickness vs. nitriding time for 3% Cr-Mo-V steelplasma 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 standarderror (small interval) at 95%.

Fig. 8


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 nitrid-ing potential of KN = 0.58 atm–1/2 (Ref 26). The curves for plasma nitridingof 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 finalvalue for the plasma nitriding curve of 36H3M steel is about 10 µm.Regardless of the sputtering effect, the plasma nitriding process in theatmosphere of 30% nitrogen and 70% hydrogen can be considered a lownitriding potential process.

In a plasma process, activation of the cathode due to ion bombardmentfrom the atmosphere containing sufficiently high nitrogen is very effec-tive, and, therefore, a full coverage of the surface with the compoundzone after only a few minutes of nitriding is achieved (Ref 19). This wasalso evidenced in our experiments by the fact that the curve did not startat the beginning of the coordinate, but earlier (Fig. 5). In a low-potentialgas nitriding process, the γ ′-phase nucleates on ferrite extremely slowlyand only after substantial time becomes a continuous, compact layer(Ref 26).

The experiments showed also that the equilibrium between plasma andthe steel surface was not achieved quickly when the atmosphere of 30%nitrogen and 70% hydrogen was used. Instead, an initially formed mix-ture of γ ′- and ε-phases was slowly converted into a single γ ′-phasestructure. The phase composition changes resulted in a reduction of theSR, as well as a reduction in the diffusional growth of the compoundzone. Sputtering of the surface in plasma nitriding has then an additional


Comparison of a hypothetical, “sputtering-free” kinetic of compoundzone 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 gasnitriding 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.

Fig. 9


effect in the process: It effectively lowers an already low nitriding poten-tial and enhances the ability to reduce the thickness of the compoundzone.

Conclusions

The experimental data of the compound zone formation taken from theliterature were analyzed and presented in graphic forms. It was found byusing the TableCurve 2D software that the SR for the γ ′ (0.036 µm/h) wassmaller than for the ε-type compound zone (up to 0.425 µm/h). At thesame time, it was confirmed that the growth of the compound zone due tothe diffusion of nitrogen was also slower for the γ ′-type, which could beexpected (Ref 26).

The graphs extrapolated over the experimental data range showed thatthe kinetic curves may achieve a maximum value and that the sputteringmay cause a complete disappearance of the compound zone after longnitriding. 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 anydisappearance of the compound zone, and its final value after 400 h ofnitriding was still about 13 µm. The SR calculated from our experimentwas about 0.03 µm/h. This agrees well with the value calculated forkinetic data taken from the literature for a similar steel for presumably theγ ′-type nitride (Ref 14). The analysis limited to cycles not exceeding 25 hresulted in a higher rate of sputtering as well as a faster diffusionalgrowth. This fact can be attributed to the presence of the ε-phase in thecompound zone.

Based on our research, it seems to be likely that the γ ′ compound zonecan disappear completely because of sputtering after an extremely longnitriding time. Instead, it is more likely that it will become more porousand the specific surface area will be greater. However, this will still needto be proved by additional experimental work.

ACKNOWLEDGMENTThe Appendix is reprinted (with minor changes) from E. Rolinski and

G. Sharp, The Effect of Sputtering on Kinetics of Compound Zone For-mation in the Plasma Nitriding of 3% Cr-Mo-V Steel, Journal of Materi-als Engineering and Performance, Vol 10 (No. 4), Aug 2001, pages 444 to448 (reproduced by permission of ASM International).

REFERENCES1. D. Pye, A Review of the Gas Species Activity and Control of Pulsed

Plasma Technology during the Nitriding, Carburizing, and Carboni-triding Processes, 1995 Carburizing and Nitriding with Atmos-pheres, ASM International, 1995, p 347–351

2. A. Szabo, Best Surface GmbH, personal communication, June 2001



3. M.A.J. Somers and E.J. Mittemeijer, Oxidschichtbildung und Gle-ichzeitige Gefugeanderung der Verbindungsschicht (Oxide-Layer For-mation and Simultaneous Microstructural Changes in the CompoundLayer), Härt. Tech. Mitt., Vol 47 (No. 3), May-June 1992, p 169–174

4. E. Rolinski and G. Sharp, Effect of Sputtering on Kinetics of Com-pound Zone Formation in the Plasma Nitriding of 3% Cr-Mo-VSteel, 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 14898. B. Edenhofer, Heat Treatment ’76, Proc. 16th International Heat

Treatment Conf., Stratford-upon-Avon, 6–7 May 1976, The MetalsSociety, 1976, p 7

9. B. Edenhofer, Proc. Heat Treatment ’79, Birmingham, 22–24 May1979, TMS/ASM, 1979, p 52

10. G.G. Tibbets, J. Appl. Phys., Vol 44, 1974, p 507211. E. Rolinski, Ph.D. dissertation, Warsaw University of Technology,

1978 (in Polish)12. T. Karpinski and E. Rolinski, Proc. 8th National Conf. Heat Treat-

ment, Bratislava, Slovakia, 1978, Dom Techniky CSVTS, p 27 (inGerman)

13. A. Marciniak, “Processes of the Cathode Heating and Nitridingunder a Glow Discharge Condition,” Ph.D. dissertation, WarsawUniversity of Technology, 1983 (in Polish)

14. A. Marciniak, Surf. Eng., Vol 1, 1985, p 28315. A. Wells and I. Le R. Strydom, Surf. Eng., Vol 2, 1986, p 28316. A. Wells and I. Le R. Strydom, Surf. Eng., Vol 4, 1988, p 5517. T. Lampe, S. Eisenberg, and G. Laudien, Surf. Eng., Vol 9, 1993, p 6918. H. Michel, T. Czerwiec, M. Gantois, D. Ablitzer, and A. Ricard,

Surf. Coating Technol., Vol 72, 1995, p 10319. Y. Sun and T. Bell, Mater. Sci. Eng., Vol A224, 1997, p 3320. TableCurve®2D, Version 4, SPSS Inc., 199821. E. Rolinski, F. LeClaire, D. Clubine, G. Sharp, D. Boyer, and R. Not-

man, J. Mater. Eng. Performance, Vol 9, 2000, p 45722. Y. Sun, N. Lou, and T. Bell, Surf. Eng., Vol 10, 1994, p 27923. J. Zysk, Metaloznastwo I Obrobka Cieplna, No. 6, 1973 (in Polish)24. B. Edenhofer, Ibsen Industries International GmbH, private commu-

nication, Aug 200025. J.G. Conybear and B. Edenhofer, Proc. 6th Int. Conf. Heat Treat-

ment of Materials, Chicago, IL, 28–30 Sept 198826. L. Maldzinski, W. Liliental, G. Tymowski, and J. Tacikowski, “New

Possibilities of Controlling the Gas Nitriding Process by UtilizingSimulation of Growth Kinetics of Nitride Layers,” presented at the18th ASM Conf., Rosemont, IL, Oct 1998




CHAPTER 14Processing with Nitriding

THE NITRIDING PROCESS can be applied to various materials andpart geometries. This chapter focuses on tool steels, pure irons, low-alloysteels, and maraging steels.

Hot-Work Tool Steels

The hot-work group of tool steels is usually considered to be the AISIH-series and includes chromium-base, tungsten-base, and molybdenum-base steels. While all of the hot-work tool steels contain chromium rang-ing from 2 to 12%, they are distinguished by their principal alloying ele-ment. All can be readily processed via gas nitriding, salt bath nitriding, orion nitriding.

Forging DiesSelection of a hot-work steel grade depends on the forge die applica-

tion. For many forging steel applications, the steel of choice is H13, whichis 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 serviceand has a good toughness factor after nitriding—provided that the preheattreatment has been done correctly in terms of core hardness for case sup-port. For the diffused case to perform within its operating environment,the core must be able to support the case when a compressive load isplaced on the steel component (Fig. 1).

Core hardness generally is determined by the hardness at which thesteel can be cut, rather than the best hardness for supporting the case. Theappropriate hardness for case support is around 44 to 47 HRC, which pro-duces a tough, springlike condition. This core hardness will allow someflexibility in the die without taking a permanent set for deformation.

Another advantage of the nitrided case is that it will withstand high-temperature operating conditions with no significant loss of surface hard-ness. Nitrided hot-work tool steels are unlike carburized steels that rely onthe diffusion of carbon and then a phase transformation to martensite,




which necessitates a quench (and results in distortion and the possibilityof mixed phase conditions).

If a mixed phase is present, the nitriding process will decompose it. Itmust be expected that with the decomposition of the mixed phase, somesize change will occur due to transformation of the retained austenite tomartensite.

Other considerations are the surface metallurgy requirements of the die,including case depth, compound layer formation, and temperature. Eachfactor 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 preheattreatment has been conducted. Inflexibility leads to surface cracking, andthus to press downtime for die repair.

Compound Layer Formation. The thickness and phase constructionof the compound layer significantly influence die performance. If a thickcase is produced, there is a strong likelihood that a thick, inflexible com-pound layer will be produced on the die surface. Therefore, the cycle mustbe run with a thin diffused nitrided case with a thin compound surfacelayer. 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 compoundlayer can be expected to be roughly 10% of the total measured case depth.However, these values will change with both controlled nitriding (dilu-tion) and particularly with ion nitriding. The latter two methods offer agreater degree of surface metallurgy control.

Temperature. Process temperature selection plays a significant role inthe thickness of the diffused case and formed compound layer. If higherprocess temperatures are selected for gas or salt bath nitriding, there is aninherent danger of nitride networking on corners. This is a very brittlephase condition, and care should be taken to minimize its potential.Process temperature traditionally has been selected based on the temper-


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


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

Aluminum Extrusion DiesAluminum extrusion dies typically are manufactured using the hot-

work steel H13 described earlier. The main area of concern with extrusiondies is the bearing face area (Fig. 2).

The mechanics of operation are based on the press load factor on thealuminum billet to be extruded. The aluminum billet is usually preheatedto around 425 °C (800 °F). At room temperatures, the surface of the alu-minum billet will oxidize to form aluminum oxide. Aluminum oxide isextremely abrasive and will begin to abrade and wear the bearing surfaceof the extrusion die. The die aperture will become larger and out of toler-ance, resulting in a costly shutdown of the press. Typical extrusion pressconfigurations are shown in Fig. 3.

Nitriding—gas, salt bath, or fluidized bed—enhances the hardness of thedie bearing surface and reduces wear. Again, a deep case is not required; ashallow case will suffice up to a maximum of 0.25 mm (0.010 in.), withformation 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 pushedover the bearing surface. Aluminum oxide formation becomes much moreaggressive due to the frictional forces now being developed as the hot

Chapter 14: Processing with Nitriding / 155

Schematic cross section of an aluminum extrusion die made from H13steel showing the bearing (wear) surface and a core with hardness of

38 to 44 HRC. Source: Ref 1

Fig. 2


aluminum is pushed through the die (Fig. 4). Thus, the wear factor becomeseven greater.

Ideal die surface metallurgy will reduce formation of the compoundlayer on the surface. This can be accomplished by controlling the processgas using the controlled nitride method (dilution) or ion (plasma) nitrid-ing. The two-stage process, which uses a higher process temperature of565 °C (1050 °F), is not recommended because of the possible risk ofnitride networking.

High-Speed Steel Cutters

Enhancing the surface hardness of high-speed steel cutters via thenitriding process offers many advantages. However, once again, a deepcase is not required. A short cycle is necessary only to diffuse a case of nogreater than 0.038 mm (0.0015 in.). The formed case will be in the region


Extrusion presses. (a) Schematic of a horizontal extrusion press show-ing a hydraulically powered ram forcing the heated aluminum billet

through the die. (b) Typical direct-drive hydraulic extrusion press. 1, hydraulicpower unit; 2, tie rods; 3, butt shear; 4, extrusion platen; 5, container shiftingcylinders; 6, swiveling operator’s console; 7, die slide; 8, container; 9, containerhousing; 10, billet loader; 11, press base; 12, billet loader cylinders; 13, pressingstem; 14, crosshead; 15, side cylinders; 16, cylinder platen; 17, main cylinder

Fig. 3


of 1100 HV or harder and will be supported by a substrate material hard-ness in the region of 850 HV. Remember, the nitride processing will givethe high-speed steel a further temper, producing a dimensional stabilizingeffect. There will be a very slight, but insignificant, increase in size due tothe diffusion.

Another procedure is to follow the nitriding with a thin-film depositionusing the plasma-assisted deposition technique. The cutter is nitrided, andthen the nitrided surface is immediately coated with titanium nitride. Twoprocess techniques are accomplished during the same furnace operation.


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

nitrided surface

Fig. 4


Gears

Nitriding is finding greater acceptance as a surface hardening methodfor precision manufactured gears, although it requires careful steel selec-tion. The steel selected must not be a high-carbon grade; otherwise, thereis a risk of forming ε as the dominant phase in a dual-phase compoundzone. This could lead to brittle fracture of the compound layer, depositingfine pieces of fractured steel from the gear surface that will usually lodgebetween the meshing teeth, thus causing further deterioration of the gearpressure face.

Furthermore, the steel must not have a high aluminum content. Other-wise, high hardness will result, leading to premature chipping on tooth cor-ners 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 forma-tion 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, pressuremust be controlled to ensure that the plasma glow seam is uniformly posi-tioned over the entire gear tooth surface. This is necessary for uniformcase depth on the pressure face and tooth root (Fig. 5).

As described below, surface metallurgy—in terms of growth, compoundlayer formation, and case uniformity—is critical to gear performance.

Growth. By selecting a process temperature around 485 to 500 °C (900to 925 °F) and then controlling it, the compound zone will not form as thickly


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

Fig. 5


as at higher temperatures and the risk of nitride networking will be reduced,along with the risk of edge or corner chipping. Remember, the solubilitylimit of nitrogen in iron increases as temperature increases. The solubilitylimit using a process temperature of approximately 485 °C (900 °F) isapproximately 6 to 7%. If compound layer thickness is reduced, growthalso is reduced.

If surface growth is reduced, less stock will have to be removed by lap-ping or grinding operations. In fact, the gear pressure face could bemachined slightly undersize and “grown” into size, thus reducing expen-sive machining time. Selection of a lower process temperature thus offersimportant 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 geartooth will appear dark around the area of the burr.

Compound Layer Formation. Process temperature affects formation ofthe γ′- and ε-phases within the compound zone. A lower process tempera-ture tends to reduce formation of the ε-phase, whereas higher process tem-peratures encourage its formation—particularly if the steel has a highercarbon level. A greater presence of ε-phase means that the immediate gearsurface 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 to1000 °F) and the core hardness is temperature sensitive, the core hard-ness may be reduced by up to 2 to 4 Rockwell points due to the processtime at the nitriding temperature. This could take the gear core hardnessout of specification. Pretreatment of the gear to establish the case sup-port core hardness is extremely important. Emphasis should be placedon producing a core hardness that will ensure a good case support andimprove fatigue bending performance of the gear tooth rather than acore hardness that will ease machining.

Case Uniformity. If a uniform case depth is not maintained, then thegear tooth could fail under service loads. Nonuniform case formation usu-ally results in a shallow case in the gear tooth root. With salt bath nitrid-ing, case uniformity usually is not a problem as the molten salt contacts allgear surfaces. With gas nitriding, it can be a serious problem due to gasstagnation in the root of the tooth. This can be caused by inadequate gascirculation. With ion nitriding, using too low a process pressure or failingto adequately control the process pressure can cause a nonuniform glowseam 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 bathmethods. (Please note that this discussion is restricted to nitriding and notferritic nitrocarburizing, which will be discussed in Chapters 18 to 23.)



The case that forms is strictly of iron nitride. However, the resulting hard-ness value is low—in the region of 35 HRC (345 HV). Resistance to sur-face corrosion and torque improves significantly.

Ion nitriding of pure iron produces higher hardness values in the regionof 60 HRC (700 HV). This is accomplished by manipulating the processgas 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 hardnessvalue will be no greater than 700 HV, but corrosion resistance will signifi-cantly improve. This technique often is used when corrosion resistance ismore important than wear resistance.

Low-Alloy Steels

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

Maraging Steels

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


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

Fig. 6


perature passes through 450 °C (850 °F) on heat up, a phase change fromferrite to austenite occurs—along with a subsequent size change. Thenitrogen diffuses into the maraging steel at that point, and the solubilitylimit of nitrogen in iron is exceeded. After the cycle time is completed, themaraging steel must be cooled quickly in the process chamber under nitro-gen. This will produce a shallow layer on the steel surface due to the rapidcooling of fine nitrogen-austenite products. This product is detrimental interms of surface mechanical properties.

Therefore, the nitriding process temperature for maraging steels shouldbe lower than for conventional steels. The process temperature must belower than the ferrite-to-austenite phase transformation temperature ofapproximately 450 °C (850 °F). A process temperature of between 425and 450 °C (800 and 850 °F) will produce a surface hardness of approxi-mately 67 HRC (900 HV). This lower-than-normal process temperatureresults in an extended case formation time (Ref 3).

The maraging steels can be successfully ion nitrided at lower tempera-tures to avoid the phase change and subsequent growth, with better con-trol of the hydrogen-to-nitrogen process gas ratios. The ratio generallywould be approximately 5 parts hydrogen to 1 part nitrogen or even 6parts hydrogen to 1 part nitrogen. This means that the solubility limit ofnitrogen in iron is not reached, and thus there can be no risk of possiblecorner networking.

Higher Alloyed Steels

Once again, the greater the level of alloying elements, the more difficultit becomes for nitrogen to diffuse into the steel surface and form stablenitrides. This is evident for the whole range of stainless steels and thehigher alloyed tool steels. The net result of nitriding the stainless steels isa very high surface hardness.

REFERENCES1. D. Pye, “Practical Nitriding” course notes, Pye Metallurgical Consult-

ing, 19972. 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




CHAPTER 15Stop-Off Procedures for

Selective Nitriding

STOP-OFF COATINGS prevent nitriding of selected areas on compo-nents. The success of a coating depends on such variables as its densityand thickness, its adhesion to steel, and part surface finish.

Methods for Selective Gas Nitriding

There are essentially two stop-off techniques for gas nitriding: electro-plating and paint-on methods.

Electroplating. One method of selective gas nitriding is to plate theareas to be stopped-off using copper or bronze. Nickel and silver are alsoeffective but costly, restricting their use to special applications. Copperplating is perhaps most widely used, because it is least expensive in rela-tion to other types of plated deposition. However, it is still an expensivemethod—not so much because of the copper plate cost, but because of thelabor-intensive part preparation required. All of the areas to be nitridedmust be masked prior to copper plating.

Thickness and density of the deposit material will determine the effec-tiveness of the plating. The following is a guide to electroplated deposi-tion thickness:

• Bronze and copper: Up to 25 µm (0.001 in.) for ground finishes; up to50 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 causedby the cyanide-base plating solution, which must be neutralized. Nickel isdifficult to strip, and the stripping process could affect part surface quality.




Paint-On Methods. Proprietary brands of stop-off paints generally con-tain a liquid carrier and a metallic compound such as copper or even tin andzinc. For copper paint, the carrier liquid is usually an alcohol-base material.When applying copper-base paint, follow the paint manufacturer’s applica-tion instructions and do not take shortcuts. Use a high-quality paintbrush thatwill not leave bristles on the painted steel surface. When the steel is heated tothe process temperature, loose bristles on the surface will burn off andexpose the steel immediately beneath. Nitriding will take place in the area ofthe burnt-out paint bristles, which can wreak havoc with cutting tools.

Some stop-off paints are water-based and must be correctly cured. Anywater remaining in the stop-off paint will manifest itself during nitriding. Thewater will form a steam pocket under the paint, producing a bubble. Whenthe bubble bursts, a localized hard spot will form on the exposed steel.

Stop-off paint residues can be reduced by brushing or washing, orremoved by light blasting with fine abrasives.

Methods for Selective Salt Bath Nitriding

Copper Plating and Paint. Selective nitriding can be accomplished insalt baths by stopping-off nitrogen penetration with either copper plate orcopper-base paint. Because cyanide-base salts can dissolve copper, saltbaths with relatively low cyanide contents must be used. One successfulsalt formulation contains 8 to 10% sodium cyanide (NaCN) with approxi-mately 45% barium chloride (BaCl) energizer. Noncyanide nitriding saltswill not dissolve copper. Following the stop-off procedure, the depositedcopper should be inspected to ensure it does not contain pinholes.

In addition, if too much plated work is processed for short cyclesthrough the cyanide salt bath, the copper plate will begin to strip off andgo into solution with the cyanide salt. When fresh work is introduced intothe salt bath, the copper in solution in the cyanide will deposit onto theexposed areas, reducing the effectiveness of the nitriding salt bath.

Partial Immersion. Another method for selective nitriding entails par-tial immersion into the salt bath so that only the immersed areas arenitrided. 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 cannotnitride.” The area to be masked should be covered with a piece of shimstock steel attached to the component. The shim stock serves as an effec-tive mechanical barrier to nitrogen diffusion.

Some paints are effective stop-offs in the ion nitriding process; how-ever, extreme caution must be exercised during the sputter cleaning stage.Sputter cleaning should not be aggressive; otherwise, the painted surfacewill begin to transfer the paint onto the furnace wall or the anode. The



sputtered paint will migrate and adhere to the process wall, leading to achange in the electrical response characteristics of the wall. Once a cyclehas been completed, the inner wall should be cleaned of deposits.

Another stop-off consideration is that when a component is in contactwith the cathodic hearth, it will not nitride. Therefore, a support system isneeded. As shown in Fig. 1, the part can be mounted onto steel points. Thepoints will maintain the cathodic contact with the furnace hearth andnitriding can take place. The case depth will be slightly more shallow atthe point of contact but not to a significant degree.

Threaded holes must be protected from nitriding. This is accomplishedsimply by partially inserting a threaded stud into the hole. Screwing threeor four threads deep is sufficient (Fig. 2).

Chapter 15: Stop-Off Procedures for Selective Nitriding / 165

Extrusion dies lifted from the furnace hearth to allow the plasma glowseam to cover the die completely. Note the die supports (steel tubes)

used to maintain cathode potentials.

Fig. 1

Fig. 2 Masking of blind tapped holes on a component for plasma nitriding.



CHAPTER 16Examination of the

Nitrided Case

EXAMINING AND EVALUATING the nitrided case is generallyaccomplished by hardness testing and microscopic examination. Thischapter discusses both characterization methods, as well as samplepreparation.

Hardness Testing

Hardness testing is perhaps the most widely used method for evaluatinga nitrided case. Hardness of a steel is used as a cross-reference to manyother properties, such as tensile strength and impact strength. Generally,hardness testing of a nitrided case is accomplished with a low load ormicrohardness load below 1 kg. Do not use the Rockwell machine with a150 kg load, which will punch through the nitrided case and provide anincorrect reading.

Hardness testing can be categorized as either macrohardness or micro-hardness. 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 loadapplications below 1 kg. It can be carried out by:

• Rockwell superficial testing• Vickers hardness testing• Knoop hardness testing




Testing Hardness Profiles through the CaseThe first method is to step-grind through the case and measure the hard-

ness as the “steps” go deeper into the case. This requires a complex grind-ing method and is somewhat time consuming (Fig. 1).

The second method is to cut a sample at right angles to the case and pre-grind and polish the cut surface. Once this has been accomplished, a hard-ness profile can be constructed through the case using a microhardnesstest method (Fig. 2, 3). This method of hardness testing requires either (a) a test coupon of the same material that is in the same metallurgicalcondition or (b) a sacrificial component.

Sample PreparationThe 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


Step-ground specimen for hardness traverse method of measuringdepth of medium and heavy cases. Arrows show locations of hardness-

indenter impressions.

Fig. 1

Cross section of a round test coupon cut through to expose the nitridedcase in preparation for a microhardness traverse test. The exposed sur-

face must be polished before testing.

Fig. 2


sample is sectioned using an abrasive cutoff wheel while submerged in thecutting fluid. Do not use excessive pressure or localized overheating likelywill occur, adversely affecting the cut surface metallurgical condition andsubsequent test results. Mixing a rust inhibitor into the coolant will reducethe risk of surface oxide formation on both the sectioned sample and theremaining sample portion. Diamond sawing is another method of cold cut-ting. This gives a very cool cut and reduces the amount of surface scratch-ing usually caused by abrasive wheel cutting. This improved surface finishconsiderably reduces the pregrind time (Fig. 4, 5).

When using an abrasive wheel cutoff unit, the wheel residuals and themetallic fines that result from sectioning must be removed from the recir-culating cooling system. The fines will build up quickly in the drain anddelivery hoses, significantly reducing the pipe diameter and thus restricting

Chapter 16: Examination of the Nitrided Case / 169

Fig. 3 Results of a microhardness traverse test on a nitrided test coupon


the coolant drainage and delivery system. This applies to the machinesump and the recirculating reservoir, both of which must be cleaned fre-quently. In addition, the bearings on which the drive motor is connected tothe wheel load handle must remain well lubricated. If the bearings becometight, then the force required to depress the abrasive wheel could lead to anexcessive load being placed on the sample being sectioned, creating thepotential for localized overheating of the sample surface.

Vapor degreasing is often used to clean the specimen after it has beenmachined prior to mounting. Hot vapors of a chlorinated or fluorinatedsolvent remove oils, greases, and waxes that may be on the specimen.


Fig. 4 Laboratory abrasive cutoff device

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

Fig. 5


Mounting of the sample prior to examination is necessary for good met-allurgical evaluation. Mounting can reduce the risk of “edge rounding,”allowing the formed and diffused case to be accurately observed by bothhardness traverse testing and microexamination after etching. Tables 1 and2 list the properties of commonly used thermosetting and thermoplasticmounting materials. Tables 3 and 4 show typical problems with compres-sion mountings and castable materials.

There are essentially two types of mounting systems: the cold-mountsystem and the hot-mount system. Each has its own advantages and disadvantages.

The two-component cold-mount system comprises a polymer and a cat-alyst that must be accurately measured, mixed, and poured over the speci-men in a mold. An identification piece or small plastic clip should also beplaced with the sample to identify the surface of the nitrided sample to beexamined (Fig. 6).


Table 1 Typical properties of thermosetting molding resinsMolding conditions Heat Coefficient

distortion of thermal Abrasion PolishingTemperature Pressure Time, temperature(a) expansion rate, rate, Chemical

Resin °C °F MPa psi min °C °F in./in.°C µm/min(b) µm/min(c) Transparency resistance

Bakelite 135–170 275–340 17–29 2500–4200 5–12 140 285 3.0–4.5 × 10–5 100 2.9 Opaque Attacked(wood-filled) by strong

acids andalkalies

Diallyl phthalate 140–160 285–320 17–21 2500–3000 6–12 150 300 3.5 × 10–5 190 0.8 Opaque Attacked (asbestos-filled) by strong

acids andalkalies

(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

CoefficientHeating Cooling Heat distortion of thermal Abrasion PolishingTemperature Pressure Time Temperature Pressure Time temperature(a) expansion, rate, rate, Chemical

Resin °C °F MPa psi (min) °C °F MPa psi (min) Transparency °C °F in./in.°C µm/min(b) µm/min(c) resistance

Methyl 140– 285– 17– 2500– 6 75– 165– max max 6–7 Water, white 65 150 5–9 × 10–5 ... 7.5 Not resistant methacrylate 165 330 29 4200 85 185 to clear to strong acids

and some solvents, especially ethanol

Polystyrene 140– 285– 17 2500 5 85 185– max ... 6 ... 65 150 ... ... ... ...165 330 212

Polyvinyl 220 430 27 4000 ... ... ... ... ... ... Light brown, 75 165 6–8 × 18–5 20 1.1 Not resistant formal clear to strong acids

Polyvinyl 120– 250– 0.7 100 nil 60 140 27 4000 ... Opaque 60 140 5–18 × 10–5 45 1.3 Resistant to chloride 160 320 most acids

and alkalies

(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



Table 3 Typical problems of compression mounting materialsProblem Cause Solution

Thermosetting resins

Radial split

Edge shrinkage

Circumferential splits

Burst

Unfused

Thermoplastic resins

Cottonball

Crazing

Too large a section in the given moldarea; sharp cornered specimens

Excessive shrinkage of plastic awayfrom sample

Absorbed moisture; entrapped gasesduring molding

Too short a cure period; insufficientpressure

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

Powdered media did not reach maximum temperature; insufficienttime at maximum temperature

Inherent stresses relieved upon or afterejection

Increase mold size; reduce specimensize.

Decrease molding temperature; coolmold slightly prior to ejection.

Preheat powder or premold; momen-tarily release pressure during fluidstate.

Lengthen cure period; apply sufficientpressure during transition from fluidstate to solid state.

Use proper molding pressure; increasecure time. With powders, quickly sealmold closure and apply pressure toeliminate localized curing.

Increase holding time at maximumtemperature.

Allow cooling to a lower temperatureprior to ejection; temper mounts inboiling water.

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

Fig. 6


The primary disadvantage of the cold-mount system is the difficulty ofobtaining two parallel faces after curing and retrieval from the mold. Itsprimary advantage is its low cost.

The second method is the hot-mount system, also known as compres-sion molding. The mold is usually made of a thermosetting material:

• Bakelite (phenolics)• Epoxy• Diallyl phthalate


Table 4 Typical problems of castable mounting materialsProblem Cause Solution

Acrylics

Bubbles

Polyesters

Cracking

Discoloration

Soft mounts

Tacky tops

Epoxies

Cracking

Bubbles

Discoloration

Soft mounts

Too violent agitation while blendingresin and hardener

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

Resin-to-hardener ratio incorrect;resin has oxidized

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

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

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

Too violent agitation while blendingresin and hardener mixture

Resin-to-hardener ratio incorrect; oxidized hardener

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

Blend mixture gently to avoid airentrapment.

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

Correct resin-to-hardener ratio; keepcontainers tightly sealed.

Correct resin-to-hardener ratio; blendmixture completely.


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

Blend mixture gently to avoid airentrapment.

Correct resin-to-hardener ratio; keepcontainers tightly sealed.



This system requires a pressure mold equipped with a heater system toensure that full curing takes place while the sample and mold material areunder compression. The curing time can be from 8 to 15 min, depending onthe type of thermosetting system. The rise in temperature from the curingheat will not be sufficient to disturb the surface metallurgy of the nitridedsample. The temperature rise is usually approximately 165 °C (300 °F).

Pregrinding. The sample preparation techniques in this section are basedon single-mounted samples and not multisampling. Hardness testingrequires a polished, unscratched surface. This necessitates pregrinding. Thepregrind involves an initial rough grind using 180-grit silicon carbide paperfollowed by intermediate pregrinding steps using 320-, 400-, and 500-gritpapers. Do not spend too much time on coarse-grit grinding: instead, con-centrate on careful sample surface preparation using finer grit sizes of 500and higher. For rough polishing, use an 800- to 1200-grit (maximum) paper.

The surface finish quality will be determined by the pregrind wheelrotational speed. Generally, this is accomplished on a rotary wheel run-ning at approximately 350 rpm (Fig. 7).

The abrasive silicon carbide paper must be kept well flushed withwater. The action of pregrinding will load up the surface of the paper withmetal particles as well as mount material from the sample face; this cancause “rescratching.” Ideally, the used pregrind sample paper should bediscarded. If the paper is reused, a burnishing/glazing effect can occur onthe steel sample surface. This will prevent a true image from being exam-ined after etching. Discarding each paper can become expensive, so thedecision rests with the technician or metallurgist.


Fig. 7 Fine grinder for wet or dry grinding of metallographic specimens


Some silicon carbide papers and polishing cloths are self-adhesive. Theplaten wheel must be thoroughly clean and free of adhesive before placinga new paper or cloth onto the platen. If any adhesive contaminant remains,the platen will not present a completely flat surface for the pregrind/roughpolish step.

The sample can be rough ground—wet or dry—using a rugged beltsander with an initial silicon carbide paper of 180 grit (Fig. 8, 9). If drygrinding is performed, be extremely careful not to cause localized over-heating of the sample surface.


Fig. 8 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 abra-

sive paper

Fig. 9


Another pregrinding method is to use silicon carbide strips of a precutlength, laid out on a smooth, flat surface. Once again, sufficient water flowis important to ensure that the silicon carbide strip is well flushed. Thesecret to reducing the polishing time is careful pregrinding.

Polishing follows the pregrinding operation. Do not let the sample sitafter the pregrind for any length of time. If the exposed sample surface iswet, it will begin to oxidize, which could cause very slight pitting. Thepolishing operation should start immediately after pregrinding to reducethe risk of oxide formation. One of two polishing mediums can be used:

• Diamond paste. While diamond paste (Fig. 10) is perhaps the hardestmaterial and will produce very highly polished surfaces, additionallubrication is necessary to ensure good distribution of the paste overthe polishing cloth to improve polishing efficiency. Recommendedparticle sizes for the final polish are 1 µm or 0.3 µm. Diamond pastesare very expensive.

• Aluminum oxide slurry in suspension. Aluminum oxide slurries forfinal polishing can have particle sizes of 1 µm down to 0.5 µm. Theparticles are suspended in water or in paste or powder form. Selectiondepends on personnel and availability. Such slurries are self-lubricatingand economical.

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


Fig. 10 Application of diamond paste to a nylon cloth for preliminary polishing


rinse with alcohol. Examine the sample for possible surface stainingand/or residual surface grease. Do not touch the surface with your fin-gers; any body oil present will inhibit etching. Great care must be takenduring the entire pregrind and polish procedure to prevent edge rounding,a condition that causes difficulties when trying to align the diamondindentor at the sample edge and when trying to focus on the edge duringmicroscopic examination.

Microhardness TestingCase Depth. After polishing, hardness testing must be performed before

microscopic examination of the surface. To establish the effective casedepth accomplished by nitriding, the transition hardness between case andcore 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 deriv-ative nitriding processes. Be cautious when being told of deep casedepths. The ISO and the European DIN specification writers haveinvested a great deal of time in defining and understanding case depth.However, many processes do not define the case depth measurement, andothers state that “the case depth achieved is x.” The conclusion drawn bythe uninitiated is that the total case depth is hard, but this is not true. In thiswriter’s opinion, the real case depth is the effective case depth and not thetotal case depth. Quoting the total case depth (which while relevant, is notalways well defined) is misleading.

The microhardness test is based on the resistance to indentation prin-ciple. As with any hardness testing procedure, the unit must be tested withthe 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 beingtested. If the immediate sample face is not at an exact right angle to theindentor, a uniform indentation will not be obtained and accurate readingscannot be made in the diffused case.

When setting the x-y traverse, carefully note the starting point from thesurface transversely across the nitrided case. When starting the traversethrough the diffused case, it is important to start as close to the edge (sur-face) as possible to accomplish an almost true surface reading. If theindentation 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 beencompleted. If etched beforehand, the hardness indentations may not beclear enough to read accurately.



Etchants Nital is perhaps the most commonly used etchant for a visual case met-

allurgical examination. Simple-to-prepare, nital consists of 4 to 6 mLHNO3 (nitric acid) and 94 to 96 mL ethyl alcohol (or denatured alcohol).

The common error when using nital involves etching time. The sampleis usually oversoaked in the etchant, resulting in a surface that is too darkfor accurate examination. The specimen should be etched for approxi-mately 3 to 8 s, taking care not to overetch. The exact etching time, ofcourse, depends on the type of steel being examined.

Nital will reveal the compound zone on the immediate surface and sta-ble nitrides in the case directly below. It will not, however, show the twophases of the compound zone. The compound zone will show as the sur-face white layer. In the case of a low-alloy steel or a plain carbon steel, theiron nitride precipitates will be seen as well-defined straight lines, belowwhich the steel core will exhibit iron nitride precipitates.

Picral is another common etching medium used to examine a nitridedsurface. It is not complex, consisting of 4 g picric acid and 100 mLethanol. The addition of 0.5 to 1% zephiran chloride improves etch rateand uniformity. Specimens should be immersed for 3 to 15 s, taking carenot to overetch.

Care should be exercised when using picric acid (refer to the MaterialSafety Data Sheet, MSDS). Proper storage of picric acid powder is criticalto operator and technician safety, and the powder should be kept verymoist or even under water.

Tips for Using Etchants. Here are a few helpful guidelines:

• When mixing etchants, always add reagents to the solvent unless spe-cific instructions indicate otherwise.

• Where water is given as the solvent, distilled water is preferredbecause the purity of tap water varies.

• Methanol is usually available only as absolute methanol. When usingthis alcohol, it is imperative that approximately 5 vol% of water beadded whenever an etchant composition calls for 95% methanol. Gen-erally the alcohol used is denatured alcohol.

• For conversion of small liquid measurements, there are approximately20 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 other-

wise indicated. An economical source of dc current for small-scaleelectrolytic etching is a standard 6 V lantern battery.

• Microscope objectives can be ruined by exposure to hydrofluoric acidfumes from etchant residue inadvertently left on the specimen. This



problem commonly occurs when the specimen or mounting media con-tain porosity and when the mounting material (such as Bakelite) doesnot bond tightly to the specimen, resulting in seepage along the edges.

• In all cases, extreme care should be taken to remove all traces of theetchant by thorough washing and complete drying before placing thespecimen on a microscope stage.

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

Safety Precautions

Here are some practices that will contribute to the safe preparation ofspecimens. The list is not exhaustive. Obviously, care must be taken whenmachining samples.

Preparing Etchant. When preparing an etchant, the operator or techni-cian is using corrosive and violently reactive acids. It is of utmost impor-tance to follow these safety precautions (Ref 1):

• Wear the proper safety clothing, including goggles, face mask, rubbergloves, 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 responsi-ble, 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 chemicalmanufacturer.

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

• Provide adequate ventilation.• Do not breathe the vapor fumes.



• 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 high-

intensity light, which could cause formation of gases such as phos-gene gas, dichloroacetyl chloride gas, and carbon dioxide gas.

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

Optical Light Microscopy

Optical light microscopy is the metallurgist’s most important tool forobserving the structure of the nitrided case. Generally, a sample must beetched before the nitrided case can be seen. The choice of microscopeshould 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?


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

Fig. 11


Microscopes range widely in cost and performance. The availability ofspare units and service from the supplier is another consideration. For qual-ity control and troubleshooting, light microscopy up to 1000 to 1200× willcover most metallurgical requirements. When research of the developmentof the case and surface metallurgy is involved, the user can usually obtainscanning electron microscopy services through a local metallurgical ormaterials science school. Examples of case microstructures are shown inFig. 12 to 18.


AMS 6470 steel with 0.15 to 0.35% Pb added, oil quenched from900 °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 awhite layer of Fe2N and a matrix of tempered martensite. 2% nital. 400×

Fig. 12

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. 13


H13 steel, heated to 1030 °C (1890 °F) in a vacuum, quenched innitrogen gas, triple tempered at 510 °C (950 °F), surface activated in

manganese phosphate, and gas nitrided 24 h at 525 °C (975 °F). White surfacelayer is iron nitride. Grain-boundary networks of nitride are present throughoutthe martensitic case. 2% nital. 300×

Fig. 16

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, andFe4N, and tempered martensite. 2% nital. 400×

Fig. 15

Same steel and processing as Fig. 12, except slack quenched andground heavily before nitriding. Because the surface was not chemi-

cally activated before nitriding, nitrogen diffusion was retarded. 2% nital. 400×

Fig. 14



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 appearblack. Modified Fry’s reagent. 1000×

Fig. 17

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 dif-

fusion zone of nitride containing tempered martensite. Nital. 750×

Fig. 18

REFERENCES1. N.I. Sax, Handbook of Dangerous Materials, Reinhold Publishing, 19512. 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 Corpo-

ration, St. Joseph, MI



CHAPTER 17Troubleshooting

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

• Process problems• Steel problems• Machining problems

Troubleshooting is a process of elimination and plain old detectivework. One must be both observant and systematic during the trouble-shooting procedure.

Gas Nitriding

Surface Cleanliness. A typical problem with gas nitriding originatesfrom part surface cleanliness. If the work surface is contaminated with ahydrocarbon-based substance such as cutting oils, machining oils, lappingcompounds, or fingerprints, the affected area will not successfully nitride,and the area below it will be soft. The remedy is to ensure absolute surfacecleanliness prior to gas nitriding. This can be accomplished by degreasingusing either chemical or ultrasonic methods.

If the part surface is contaminated with a cutting fluid that is chloride-or sulfide-based, then serious surface pitting and loss of hardness mayresult. It can be safely said that any grease or acidic compound on the sur-face of a steel is very likely to cause nitriding problems.

Loss of Gas Dissociation. When gas dissociation cannot be achievedduring the nitriding process, something within the process hardware istaking the gas flow away from the components being nitrided. Check thatthere is no restriction in the feedline from the ammonia gas source to theprocess chamber. If the line is internally oxidized, it will restrict theammonia flow. Internal oxidation of the process gas-delivery system




results from the incorrect pipe material selection. Check also that the inletpipe to the process chamber is clean and clear.

If the support fixture for the work is made from a low-alloy materialsuch as mild steel, then it will pull the dissociated gas away from theworkpiece. This also applies to the process chamber fabrication material.Even if the process chamber is made from the correct material, it canbegin to pull the ammonia necessary for the nitriding process away fromthe workpiece if it is reaching the end of its useful life, or if it has becomesaturated with nitrogen. This can be remedied by shot blasting the insidesurface of the process chamber or by regenerating the chamber by heatingup to a temperature of 785 °C (1450 °F), followed by light shot glass beadblasting or sandblasting.

Surface Discoloration. When the process chamber is opened afternitriding, the parts occasionally may be discolored with an almost rain-bow effect. The discoloration is an oxide formation on the steel surfaceand is in no way detrimental to the part. In fact, it will enhance the surfacecorrosion resistance. This is the principle behind the oxynitride proce-dure, where the steel surface is deliberately oxidized in a controlled man-ner to improve its corrosion resistance. Many engineers have been unnec-essarily concerned regarding surface discoloration after the gas nitridingprocess.

Case Exfoliation. When the case exfoliates, or peels off, this is usuallyindicative of surface decarburization present on the immediate surface ofthe steel. Surface decarburization can occur as a result of:

• Insufficient stock removal from the steel surface, leaving a decar-burized layer that will result in defective surface metallurgy. Theremedy is to ensure removal of more than 10% of the surface stockthickness.

• A decarburizing condition during the preheat treatment operation thathas left the surface seriously decarburized

Components that exhibit these surface conditions cannot be salvaged andare scrap.

Should the surface exhibit a failure of the case due to chipping (e.g., onthe pressure face of a gear), it is usually indicative of an aluminum-bearing(Nitralloy) steel. The aluminum-bearing steels have very high surfacehardness values and cannot be subjected to an impact load condition. If agear tooth tip begins to chip off, it generally indicates a supersaturated caseformation of nitrogen, where the solubility limit of nitrogen in iron hasbeen exceeded, resulting in nitride networking and a brittle case. This usu-ally takes place on sharp corners and edges. The only course of action is toinvestigate the gas dissociation control.

If the surface of the case exhibits light, flaky peeling, some type of sur-face contamination may be present. Investigate the cleaning treatments



prior to nitriding. Surface contamination is the most probable cause of anyflaking 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 fur-nace, the source of contamination may likely be within the furnace processretort. Examine the internal faces of the retort for signs of contamination.The interior walls may simply require a cleaning operation or wipe down. Ifnecessary, the wipe down can be accomplished by using an alcohol-basedsolution and hand wipes. (Appropriate ventilation should be used to ensurethat the alcohol fumes do not overcome the operator.) It may be advisable toconsider either a burnout of the process retort or complete replacement.

Orange Peel Effect. If the workpiece surface shows what can only bedescribed as an orange peel effect, with the appearance of dimples, thenonce again it can be traced back to surface decarburization. The remedy isto investigate the surface machining and the preheat treatment conditions.The part is not salvageable unless there is a deep case and the surface dim-pling can be ground off. This, however, is not a practical solution.

Case crushing will occur when the case is too thin and the core hard-ness too low in relation to the designed surface load conditions. This canbe 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 tem-

perature, 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 andensure reasonably repeatable nitriding conditions: process salt chemistryand bath cleanliness.

Process Chemistry. The subject of discussion here is salt bath nitridingand not ferritic nitrocarburizing. There is a great distinction between thetwo processes. It is important to understand the chemical makeup of thenitriding salt. Once the salt chemistry is understood, along with the basicsof salt titration for measuring nitriding activity, the bath should be titratedat least once per day in order to maintain a consistent bath activation. Theusual practice is to titrate for cyanide only, not carbonate.

Chapter 17: Troubleshooting / 187


Bath Cleanliness. Because there are many sources of bath contamina-tion, the bath must be regularly desludged. Desludging should take placeat least once per day, or even once per shift if the bath is operating on ashift basis. This is extremely important. The main contaminant in the bathcomes from iron oxide and the precipitation of products of the carbonatecontent. The sources of iron oxide are fixtures, soft iron support wire forsuspending parts into the bath, and, if the bath is made from mild steel,from the inside of the bath itself. Nitriding baths usually are fabricatedfrom mild steel or a low-carbon or low-alloy steel but in some instancescan be made from a heat-resisting stainless steel.

Postwashing Treatments. If the workpieces have been previously heattreated through salt baths such as neutral salts followed by a quench into amarquenching salt, thorough postwashing is necessary to prevent exfolia-tion or flaking of the case.

Salt Bath Nitriding of Alloy and Tool Steels. When nitriding thehigher alloyed steels and tool steels, the nitrogen diffusion rate will not beas great as for the lower alloy steels. Therefore, do not confuse a shallowcase as being caused by the cyanide level of the bath; it is simply a matterof a slower rate of diffusion. Cycle times are usually short, in the region of5 min up to 180 min, without quench.

Ion Nitriding

With the ion nitriding process, process problems as well as metallurgicalproblems can occur. Process problems generally will affect the processmetallurgy. Care of the furnace is mandatory, even in terms of initial instal-lation of the equipment. Considering the cost of water, the water-coolingsystem usually is a closed-loop system that goes through a water chiller orcooler. The system must be piped from the inlet of the cooler/chiller to thereturn of the chiller/cooler. If the pipework is made from galvanized steeltube, then very aggressive corrosion will begin to occur that over time willrestrict water flow. Use either copper or stainless steel tubing.

O-Ring Seal. Another common problem involves the overuse of vac-uum grease on the O-ring seals before commencing the process cycle. Thegrease can be your friend, but it can also be your enemy in terms of vac-uum accomplishment. Like sticky flypaper, too much grease on the sealwill attract dust. The amount of vacuum grease required is as much as ittakes to make the tip of your finger look wet. It is not necessary to squeezethe grease from the tube into the palm of your hand. Wipe the seal cleanbefore closing the process chamber and then apply the vacuum greasesparingly. Do not use any other type of grease.

Process Vessel Cleanliness. Bell-type plasma nitriding furnacesrequire regular cleaning. Each week, wipe the inside of the process cham-ber walls to remove any carbon particles and sputtered particles, takingcare to keep particle matter from falling onto the O-ring seal. It is impor-tant that the vacuum sealing arrangement be protected.



Backstreaming. Each week, check the vacuum pump oil in terms oflevel and condition. Ensure that the vacuum pipe connecting the vessel tothe pump is fitted with a valve or trap to prevent the possibility of back-streaming of vacuum pump oil when a pressure differential occursbetween the process chamber and the vacuum pump.

Overheating at a workpiece corner or edge can be visually seen throughthe nitride furnace sight port. This likely is due to a problem with high volt-age, high pressure, or a combination of both. Check the vacuum pressurelevel and decrease the pressure. Alternatively, reduce the process voltage.These adjustments, separately or together, will reduce the overheating. Thesame applies when a hollow cathode is observed. If the localized overheat-ing is within 55 °C (100 °F), then nonuniform case metallurgy will occur.Temperature uniformity within the process retort, be it gas or plasmanitriding, is mandatory.

Loss of Nitriding. If the plasma glow seam provides only partial cover-age, the portion of the workpiece that is not covered by the glow will notbe nitrided. Partial coverage is caused by a process chamber pressure thatis too high toward atmospheric pressure. The only way to correct this is toreduce the pressure until the glow seam uniformly covers all the surfacesto be nitrided (Fig. 1).


Complete glow seam coverage, ensuring complete nitriding of all sur-faces. Courtesy of Plateg GmbH

Fig. 1


Another condition that can cause the loss of glow seam coverage iswhen the surface area available for nitriding exceeds the original designcharacteristics of the furnace. This is particularly true for continuousdirect current (dc) systems. The design basis of the continuous dc unit isthe power ratio of current density in relation to surface area to be treated.Therefore, if the surface area exceeds the original design capabilities ofthe system, then there will be a loss of plasma coverage and subsequentpartial nitriding of the component.

Arc discharge is likely to occur if the plasma pulse duration is toolong. The solution is to shorten the pulse duration time to the point whereno 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 theassumption that the operator can see the arc discharge through the furnaceviewing port. If the operator is not in attendance, the arc discharge will notbe seen. It will, however, be evidenced by the appearance of the work-piece after the cycle is complete. The furnace technician must understandthe possible causes of the arc discharge phenomenon and be able to pro-gram the unit so that arc discharging will not occur. For continuous dcplasma furnaces, the technician can only adjust the process voltage and/orthe process pressure (Fig. 2, 3).

Chipping on corners and edges is usually caused by the oversaturationof nitrogen in iron. Remember, the solubility of nitrogen in iron at tradi-tional nitriding temperatures is between 5 and 7 wt% (changing, of course,with changes in alloying elements in the steel). Oversaturation means thatexcess nitrogen is not in solution, but has migrated to grain boundaries atcorners and edges, forming nitrides and thus weakening the area.

It is only possible to observe this phenomenon by light microscopy. Theexcessive nitrides at the grain-boundary areas will be seen as excessive


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 inan avalanche-like increase in power. A big arc will form, destroying the surfaceby melting due to overheating. Courtesy of Plateg GmbH

Fig. 2


white lines in the corner or edge area. The corrective action is to reducethe nitrogen-to-hydrogen ratio or reduce the process temperature. How-ever, once the nitride networking has occurred, it cannot be removed, butonly adjusted for on the next process cycle. It is thus important to place atest coupon in the load (preferably of the same material and condition asthe parts being processed) for metallurgical examination after processing.

Low surface hardness after ion nitriding can be attributable to a num-ber of possible occurrences, even if the case has formed. It usually indi-cates a low nitriding potential, or low nitrogen-to-hydrogen ratio. This canbe determined either optically or by microhardness testing.

If the unit uses pulsed dc, low surface hardness could indicate a prob-lem with pulse duration and repetition. Power that is on for too short atime period and then off too long means a loss of nitriding effect on boththe surface and the case depth.

Surface contamination is another possible cause of low surface hard-ness. Some cutting fluids are silicone based; any residual deposits will actas a barrier to nitrogen diffusion.

Surface flaking is usually caused by some form of surface contamina-tion. The contamination can be due to decarburization from previous heattreatment operations, machining operations using coolants that containsilicones, or inadequate cleaning after salt bath treatment. Another prob-lem source is contaminated grit on glass bead blast or shot blast. Preclean-ing is mandatory for both gas nitriding and plasma nitriding, regardless ofwhat system is used.


The same workpiece as in Fig. 2, but during plasma nitriding withpulsed dc glow discharge. Conditions such as vacuum pressure, gas

mix, and power input remain the same. By using pulsed dc with a repetition fre-quency of about 10 kHz, the formation of micro-arcs is suppressed. Courtesy ofPlateg GmbH

Fig. 3



CHAPTER 18What Is Meant by

Ferritic Nitrocarburizing?

FERRITIC NITROCARBURIZING accomplishes surface treatment ofa 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 carbondiffuse into the steel surface. The process is categorized as a thermochemi-cal 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 carbonatoms into a solid solution of iron, thus entrapping the diffused atoms in theinterstitial lattice spaces in the steel structure (Ref 1).

As with the nitriding procedure, there are many methods and deriva-tives of ferritic nitrocarburizing. These are discussed in the chapters thatfollow.

Process Benefits

Ferritic nitrocarburizing improves the surface characteristics of plaincarbon steels, low-alloy steels, cast irons, and sintered ferrous alloys. Asdescribed in later sections of this chapter, resistance to wear, fatigue, andcorrosion are improved with the introduction of nitrogen and carbon.

Scuffing resistance means the resistance to wear on the metal sur-face. This is accomplished by changing the nature of the surface com-pound layer, which is also known as the white layer. The completedcompound 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 com-position of the compound layer. This means that treated steel has greaterresistance to fatigue failure than an untreated steel (Ref 1).

Corrosion Resistance. After ferritic nitrocarburizing, steel parts canwithstand many hours in a salt spray environment, whereas an untreatedplain carbon steel will fail the corrosion test very rapidly.




Low Distortion. Another major advantage of the ferritic nitrocarburiz-ing process is that the procedure is carried out at a low temperature thatprevents phase changes in the steel (from ferrite to austenite), thus reduc-ing the risk of distortion. Distortion is the result of the release of inducedstresses, the thermal shock of quenching, and the risk of incomplete trans-formation to martensite. No phase change occurs during the ferritic nitro-carburizing treatment.


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

Fig. 1


Early History of Ferritic Nitrocarburizing

Ferritic nitrocarburizing has been a proven process for many years andis now gaining much acceptance by engineers. This increased interest inthe process, the author believes, is due to engineers gaining a better under-standing of materials selection and metallurgists gaining a greater under-standing of process capabilities and restrictions. In addition, many furnacemanufacturers want to serve their clients by developing new and moreefficient process methods and equipment.

The early methods of ferritic nitrocarburizing were accomplished inlow-temperature (550 °C, or 1020 °F) salt baths working on the principleof the decomposition of cyanide to cyanate (in the ferrite region). Imper-ial Chemical Industries in England pioneered the salt bath process,which was called the “Sulfinuz” treatment (Ref 2). The salt also con-tained a sulfur compound in its chemistry. The process was based on theformation of:

• Nitrides: The nitrides were formed as a result of the nitrogen compo-nent contained in the cyanide salt. The nitrogen diffused into the steelto form iron nitrides in low-alloy steels and stable nitrides in higheralloyed steels.

• Carbon: The carbon was supplied from the salt in limited quantitiesand 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 causedslight surface porosity on the treated steel. This allowed the surface poresto become minute reservoirs, retaining lubricant on the immediate surface.

The net result was that the treated component resisted scuffing andexhibited excellent resistance to frictional wear problems. The processwas 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 solu-ble. The treated component required extensive hot water cleaning aftertreatment. Cleaning became a major issue.

Problems associated with salt bath processing led to experimentationwith gaseous methods of ferritic nitrocarburizing. Experiments were con-ducted in the late 1950s with gaseous methods by Cyril Dawes of JosephLucas Ltd. in England. The company successfully applied for a patent onthe process in 1961 (Ref 3).

The gaseous procedure produced a porous layer very similar to the layerproduced with the Sulfinuz process (with the exception of forming surfacesulfides), which claimed to provide good antifrictional properties. Theprocess patent stated that the gaseous atmosphere consisted of ammonia,

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


with a hydrocarbon gas and other small amounts of carbon-containinggases (Ref 3).

An important study that contributed greatly to the scientific understand-ing of gas nitrocarburizing treatments and compound layer structure waspublished by Prenosil in 1965 (Ref 4). As a result of the study, many com-panies developed variations of the original patented process and the pro-cedure was accepted by engineers and metallurgists alike.

Advances in gaseous nitrocarburizing did not stop or hinder the processtechnique of using salt baths for the ferritic nitrocarburizing process. Ifanything, it spurred on the salt manufacturers to develop more environ-mentally friendly salts and cleaner procedures. Degussa of Germanydeveloped the salt bath process of “Tufftride,” a two-component processthat formed both nitrides and carbides in the immediate surface of thesteel (Ref 1). The process will produce only very shallow case depths,approximately 0.05 mm (0.002 in.) deep, but with high surface hardnessvalues, good fatigue properties, and excellent corrosion resistance. Theprocess cycle times are relatively short (in the region of 1.5 h), followedby a quench (Fig. 2).

Once again, the process relies on the decomposition of cyanide tocyanate, which is accelerated by the introduction of a titanium aerationtube. The aeration tube passes air through the molten salt from the bottomof the salt pot. The system requires good operational maintenance in termsof regular bath desludging, salt analysis, and periodic regeneration. Thisrequires raising the bath temperature to 575 °C (1070 °F) and holding forapproximately 2 h, followed by another desludging operation. The pur-pose is to precipitate out of the molten salt any free iron originating fromwork support baskets and fastening wire used to wire the components inplace in the work basket (Fig. 3).


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

Fig. 2


With the advent of pulsed plasma technology in the early 1980s for ionnitriding, it did not take long to realize that another method of ferritic nitro-carburizing had been discovered. This procedure was soon commercial-ized. Advantages include faster process cycle times, less surface cleaningand preparation, deeper case formation, and better control of surface metal-lurgy formation. Equipment is now being built that is capable of performing


Work-holding fixtures and wiring techniques used in liquid nitrocar-burizing. (a) Typical holding basket for small parts, equipped with a

funnel for loading parts into the basket without splashing. Funnel, which is madeof 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 mustbe free from oil, or the parts will stick together. Parts must be dry. (b) Inconel bas-ket of simple design. Upper loop of the handle is for lifting; lower loop accom-modates a rod which supports the basket over the furnace. (c) Simple basket withtrays, intended for small parts. Trays provide a maximum of loading space with-out adversely affecting circulation. Entire fixture is made of Inconel. (d) Nettedfixture, of Inconel, for holding small parts with a head or shoulder. (e) Methods ofwiring small parts. Black annealed steel wire is used for parts weighing less than10 lb; annealed stainless wire is used for heavier parts. (f ) Hooks, made of nickelalloy rod, for holding circular parts. (g) Method for holding large parts in whichtapped handling holes are available or can be provided. Nickel alloys are usedfor such fixtures because of the need for high-temperature strength. Resistance tooxidation is not a factor, as liquid carburizing salts are reducing. (h) Rack forholding six small crankshafts; exploded view shows a crankshaft in position. (i) Special rack for carburizing the outside diameters of bearing races. Holdingplates are made of mild steel; rods, of Inconel.

Fig. 3


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

Why Ferritic Nitrocarburize?

The physical benefits of ferritic nitrocarburizing have been listed. Thechoice to nitrocarburize is an economic one, when compared to othermethods of achieving the same benefits. Figure 4 presents an approximatecost comparison of various surface treatments (Ref 5). Besides the directcost of the equipment, the process selection procedure should consider thetotal investment costs.

Cost of floor space involves direct purchase or rental of space.Remember, floor space also includes storage area for fixtures and fittingsand workload preparation area.

Installation costs are sometimes overlooked. The cost of installationmeans the cost of unloading equipment from the delivery vehicle andpositioning the equipment in place. Will riggers need to be hired? It alsomeans the cost of a new facility if one is built to accommodate the newequipment, including all plumbing, electrical wiring, gas delivery sys-tems, water delivery system, and effluent exhaust system.


Cost

CVD TiC

PVD TiN

Cobalt-base

Vapor deposited:

Surface weld:

Iron-base

Ni-Cr-B + WC and fuse

Ni-Cr-B and fuse

13% Cr wire

Combustion gun sprayed:Al2O3

Electrochemical:

Thermochemical:

Electroless:

Plasma sprayed:

WC-Co

Nickel

Cobalt + Cr3C2

Chromium

Nitrocarburizing

Nitriding

Carburizing

Fig. 4 Approximate relative costs of various surface treatments. Source: Ref 5


Cost of Insurance and Freight. What type of crating or container willbe used? After loading is complete, a visual inspection should be carriedout; if possible, digital photographs should be taken of the load in its posi-tion in the event of possible insurance claims when the contents are lateruncrated. 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, incom-

plete or improper paperwork can cause serious delays at either the portof 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 fullyinsured. Before installation, check the suitability of the intended sitewith the insurance carrier. Does the room or building have the neces-sary fire protection? Would the existing fire protection system damagethe new equipment?

• Road transport from the port of arrival. Road transportation permitsmay be necessary if the vehicle load is considered a wide load.

• Access. Be sure that before the equipment arrives, doors and wallapertures are large enough to allow easy access of the equipment intothe facility. It can be embarrassing if the furnace will not fit throughthe door. Preplanning models can sometimes be used to navigate largeequipment through plants.

Operating costs include materials, energy, disposal of spent chemicals,labor (including training), rejected materials, and time. All these costs mustbe 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 hopefullystarts the first time), at least two primary discipline people—the operatingperson and the maintenance person—should visit the manufacturing sitewhen the furnace is being assembled. They also should be present after thehot trials to see how the furnace is dismantled. Photographically docu-ment 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 under-standing the process principles, the method of nitrogen diffusion, andthe expected results in relation to the steel being treated.

• An understanding of the equipment performance: This means under-standing the operation, functions, and capabilities of the equipment, aswell as reactions of the process in relation to part geometry.



While many more considerations may arise when preparing to ship afurnace from one place to another, the previously mentioned ones willserve to stimulate thinking between the team responsible for deliveringthe equipment and the client. In order for a furnace project to be success-ful for both the purchaser and the seller, there must be clear lines of com-munication regarding each party’s responsibilities. This must includeexpectations of performance from both the furnace manufacturer and theclient.

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 1,011,5804. 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 transla-tion 4720)

5. J.R. Davis, Ed., Surface Engineering for Corrosion and Wear Resis-tance, ASM International, 2001, p 191



CHAPTER 19Salt Bath FerriticNitrocarburizing

FERRITIC NITROCARBURIZING is a diffusion process that is amodified form of nitriding, not a form of carburizing. In the process, nitro-gen and carbon are simultaneously introduced into the steel while it is inthe ferritic condition, that is, at a temperature below which austenitebegins to form during cooling. As shown in Fig. 1, methods of accom-plishing the process include salt bath procedures, gaseous methods, andion (plasma) procedures, as well as the process trade names. There areslight variations in the processes identified by the different trade names, aswell as slight variations in the resulting surface metallurgy, but overallthey all constitute ferritic nitrocarburizing.

The process is carried out at a higher temperature than the nitriding pro-cedure, generally in the region of 540 to 625 °C (1000 to 1155 °F). Casedepth, once again, depends on the residence time at the selected processtemperature.

Early work was reported by Professor Tom Bell on the Sulfinuz saltbath process, which was based on the diffusion of cyanide-based saltswith sulfur diffusion (Ref 1). He further reported on the Degussa Tufftrideprocess, which involves the decomposition of cyanide to cyanate (Ref 2).A gaseous process that is based on a mixed-gas composition of ammoniaand propane in equal proportions further demonstrated that the compoundlayer would be predominantly ε-phase in the immediate surface of thetreated steel (Ref 3). The same investigator, Prenosil, found that the intro-duction of oxygen at the end of the process cycle oxidized the immediatesurface of the steel due to the higher dew point levels in the furnaceprocess chamber. The oxide layer was found to have a higher resistance tosurface corrosion than if the steel was left untreated.

The key elements of the process are:

• Temperature • Suitable carbon source• Time • Sealed and controlled environment• Suitable nitrogen source




It is now more than 40 years since the development of the low-temperaturecyanide-based salt bath nitriding process. Salt bath nitriding will form notonly the compound layer at the steel surface, but will allow nitrogen aswell as carbon to diffuse into the surface to improve fatigue strength, tor-sional strength, and tensile strength. In addition, the process will furtherimprove wear resistance, galling resistance, and corrosion resistance.Another advantage of both ferritic nitrocarburizing and nitriding is thatthey are low-temperature processes. This means that distortion will bekept to an absolute minimum. However, to say these processes are distor-tion free is misleading.

Low-Cyanide Salt Bath Ferritic Nitrocarburizing

Salt bath nitrocarburizing was first established in the late 1940s whenhigh-cyanide nitrocarburizing salt baths were introduced. Environmentalconsiderations and the increased cost of detoxification of cyanide-containingeffluents have led to the development of low-cyanide nontoxic salt bathnitrocarburizing treatments. This section summarizes the development ofone commercially successful low-cyanide treatment known as the Meloniteprocess developed by Houghton Durferrit (Ref 4).


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

Fig. 1


Melonite ProcessThe Melonite process, also known as Meli 1, uses a molten salt bath of a

special composition. It provides a wear- and scuff-resistant surface onsteels, sintered irons, cast irons, and similar materials. Treated parts exhibitexcellent wear and corrosion resistance and good sliding characteristics.

The process salt bath is mainly an alkali cyanate plus an alkali carbon-ate melted in a steel pot and fitted with an aeration system (as in the origi-nal Tufftride process). The cyanate component thermally reacts with thecomponent surface to form alkali carbonate. Either continuously or inter-mittently, the bath is treated to regenerate the carbonate product back to acyanate.

The process forms a complete multilayer surface case that comprises acompound layer and a diffusion layer. The surface compound layer, consist-ing of different compounds of iron, nitrogen, and oxygen, resists abrasioncorrosion and scuffing and is fairly stable at elevated operating tempera-tures. Surface hardness depends on the steel that is being treated, andresearchers at Durferrit claim surface hardness values ranging from 800 to1500 HV. The diffusion layer consists of nitrides of the appropriate alloyingelements and formed carbides.

Like the surface hardness, the case depth will vary according to the typeof steel being treated. Simply stated: The lower the alloy content of thesteel, 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 result-ing hardness value. The case depth, however, is shallower (Fig. 2).

Compound Layer. During salt bath nitrocarburizing by the Meloniteprocess, a nitrocarburized layer is formed consisting of the outer com-pound layer (ε-iron nitride) and the underlying diffusion layer. The basematerial determines the formation, microstructure, and properties of thecompound layer.

The compound layer consists of compounds of iron, nitrogen, carbon,and oxygen. Due to its microstructure, the compound layer does not possess

Chapter 19: Salt Bath Ferritic Nitrocarburizing / 203

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

Fig. 2


metallic properties. Particularly resistant to wear, seizure, and corrosion, italso is stable almost to the temperature at which it was formed. Comparedwith plasma or gas nitrocarburizing, compound layers with the highestnitrogen content can be obtained by the Melonite process. Layers with highnitrogen 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 hard-ness of about 800 to 1500 HV. Figure 3 compares surface layer hardnessesproduced by various processes.

In the metallographic analysis of salt bath nitrocarburized components,the compound layer is defined clearly from the underlying diffusion layeras a slightly etched zone. During the diffusion of atomic nitrogen, thecompound layer forms. The growing level of nitrogen results in the solu-bility limit in the surface zone being exceeded, which causes the nitridesto precipitate and form a closed compound layer.

In addition to the treatment parameters (temperature, duration, and bathcomposition), the levels of carbon and alloying elements in the materialsto be treated influence the obtainable layer thickness. As stated previously,the higher the alloy content of the steel, the higher the resulting surfacehardness value. The case depth, however, will be shallower.

The profile shown in Fig. 4 is for Meli 1 bath at 580 °C (1075 °F) withthe usual treating duration of 60 to 120 min. The compound layer obtainedwas 0.1 to 0.2 mm (0.0004 to 0.0008 in.) thick on most materials. Depend-ing 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. Herewe are considering only the first Q cycle.

Diffusion Layer. The material largely determines the depth and hard-ness of the diffusion layer. The higher the alloying content of the steel, thelower the nitrogen penetration depth at equal treating duration. On theother hand, hardness increases with higher alloying content.


480°C(895°F) 1.5 h

480°C(895°F) 3.0 h

480°C(895°F) 6.0 h

580°C(1075°F) 6.0 h

1200

1400

1000

800

600

400

2004 8 12 16 20 24 28 32 36 40

Distance from surface, in. × 10–4

Vic

kers

har

dess

HV

0.1

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

Fig. 3


In the case of unalloyed steels, the crystalline structure of the diffusionlayer is influenced by the cooling rate after nitrocarburizing. After rapidcooling in water, the diffused nitrogen remains in solution. If cooling isdone slowly, or if a subsequent tempering is carried out, some of the nitro-gen could precipitate into iron nitride needles in the outer region of thediffusion layer. This precipitation improves the ductility of nitrocarbur-ized components. Unlike unalloyed steels, part of the diffusion layer ofhighly alloyed materials can be better identified metallographically fromthe core structure due to improved etchability. However, the actual nitro-gen penetration is also considerably deeper than the darker etched areasvisible metallographically.

Cooling does not influence diffusion layer formation to any noteworthyextent. Figure 5 shows the compound layer thickness for various materialsin relation to treatment time.


Air350–400 °C(660–750 °F)

580 °C (1075 °F)

350–400 °C(660–750 °F)

350–400 °C(660–750 °F)

Pre

heat

ing

Nitr

ocar

buriz

ing

Oxi

dizi

ng +

coo

ling

Pol

ishi

ng

Pos

toxi

dizi

ngTem

pera

ture

Q P Q

Time

Time versus temperature profile for the QPQ nitrocarburizing treat-ment cycle. Source: Ref. 6

Fig. 4

Carbon steelse.g., 1015, 1045Low-alloy steelse.g., 4140

High-alloy steelse.g., D3, H11, 304Cast iron

18

10

6

4

2

00.5 1 2 3 4

Treating time in the MELI 1 bath, h

Com

poun

d la

yer

in te

n th

ousa

nds

Compound layer thicknessMelonite MELI 1 at580°C (1075°F)

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

Fig. 5


Surface Hardness and Tensile Strength. The surface hardness obtain-able by the Melonite treatment is influenced primarily by the compositionof the material. The higher the content of nitride-forming alloying ele-ments (Cr, Mo, Al, V, Mn, Ti, and W), the greater the surface hardness.Table 1 gives average tensile strengths and surface hardnesses for salt bathnitrocarburized steels.

Wear Resistance and Running-In Properties. Due to the intermetalliccomposition of the compound layer, friction and the tendency to weld with ametallic counterpart are reduced. Melonite treated components exhibit excel-lent sliding and running-in properties, as well as greater wear resistance.

Wear tests and practical applications repeatedly confirm the superiorwear resistance of salt bath nitrocarburized parts over traditional case-hardened, induction-hardened, or hard-chrome-plated surfaces. In manycases the wear resistance of the compound layer is further improved by anoxidative post-treatment. For example, components such as transmissionshafts, plug gages, and hydraulic aggregates have a longer service lifeafter Melonite treatment than after hard chrome plating.

What about the wear resistance of the diffusion layer? Figure 6 com-pares the wear behavior of rocker arms treated by two different heat treat-ment processes. It shows the wear on the surface face of the rocker arm thatran against a salt bath nitrocarburized camshaft made from chilled castiron. Although the surface hardness of the case-hardened rocker arm wasslightly reduced by nitrocarburizing, the much improved wear resistancedue to the presence of the compound layer to approximately 80 h runningtime is clearly visible. After 70 to 80 h, the wear profile runs parallel to thatof 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 Approximate surface hardness afterhardening and tempering at 90 min at 580 °C (1075 °F)

Steel designation 600 °C (1110 °F), MPa using the Melonite process

Germany U.S. 2 h 6 h HV 1 HV 10 HV 30

Ck15 1015 600 550 350 300 200C45W3 1045 750–850 700–800 450 350 250Ck60 1060 750–900 700–800 450 350 25020MnCr5 5120 800–950 800–900 600 450 40053MnSi4 ... 850–950 800–900 450 400 35090MnV8 O2 1000–1200 900–1100 550 450 40042CrMo4 4140 900–1200 900–1100 650 500 450X19NiCrMo4 ... 900–1100 900–1000 600 500 45055NiCrMoV6 L6 1200–1400 1150–1300 650 550 50050NiCr13 ... 1200–1350 1100–1200 600 500 450X20Cr13 420 1000–1200 1000–1200 >900 600 450X35CrMo17 ... 1000–1200 1000–1200 >900 700 550X210Cr12 D3 1500–1700 1400–1600 >800 600 450X210CrW12 ... 1500–1800 1400–1650 >800 600 500X165CrMoV12 ... 1400–1900 1400–1700 >800 650 50045CrMoW58 ... 1500–1800 1400–1700 800 700 600X32CrMoV33 H10 1700–1800 1600–1750 >900 850 700X38CrMoV51 H11 1700–1900 1500–1700 >900 850 700X37CrMoW51 H12 1700–1900 1600–1800 >900 800 700X30WCrV93 H21 1500–1800 1500–1700 >900 850 800

Source: Ref 7


afforded by the diffusion layer. A spontaneous increase in wear after theloss of the compound layer was not observed (Ref 7).

This test again showed that a high surface hardness does not automati-cally mean that the protection against wear is also very high. Assessment ofa 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 ismuch lower than that of other surface layers. Figure 7 shows the results.

Salt Bath Nitrocarburizing plus Post Treatment (Ref 6)

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


Case hardened

Case hardened andMelonite treated

Rocker arm: CrMo alloyedcase hardening steelCamshaft: chilled cast iron,Melonite treated

Test conditions: 1000 rpm

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

0.20

0.25

0.15

0.10

0.05

40200 60 80 100Running time, h

Loss

of w

eigh

t of t

he r

unni

ng s

urfa

ce, g

Influence of surface treatments on rocker arm wear. Curves on left arefor case-hardened components. Curves at right are for case-hardened

and Melonite-treated components. Source: Ref 4

Fig. 6

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

Fig. 7


treatment are carried out on the nitrocarburized surface. The quench-polish-quench (QPQ) process is based on a sequence of process steps that occursdirectly following the nitrocarburizing cycle. The process follows the sametime-temperature profile shown in Fig. 4. The process begins with the nitro-carburizing segment—that is, preheat—salt bath nitrocarburize, and saltbath quench, which produces a compound layer of ε iron nitride. The nextstep is a mechanical polish of the nitride layer by vibratory polishing, lap-ping, centerless grinding, or similar means. Finally, to optimize corrosionresistance, the component is reimmersed in the salt quench bath for 20 to 30min, rinsed, and oil dipped.

The level of corrosion protection provided by the QPQ variant isshown in Fig. 8. The results demonstrate that the QPQ process providesmaximum corrosion resistance compared with chromium plating, nickelplating, and conventional salt bath nitrocarburizing. Another comparativeevaluation of corrosion resistance based on the ASTM B 117 salt spraytest is shown in Fig. 9. These results also show the superior protectionprovided by the QPQ treatment, even after 336 h exposure to the salt spraytesting environment. The QPQ treatment also improves antifrictionalcharacteristics and fatigue properties of steel parts.

Kolene Nu-Tride Process (Ref 8)

The Kolene Corporation offers a proprietary salt known as Nu-Tride forsurface treatment. The procedure, which will be referred to genericallyand subsequently as salt bath nitrocarburizing (SBN), is controlled by theuse of cyanates and carbonate salts in a molten condition as the source fornitrogen and carbon. A typical process reaction is:

8CNO → 2CO3 + 4CN + CO2 + [C]Fe + 4 [N]Fe


10

5

0

0.016

0.008

0Chromium

20 µmNickel20 µm

SBN QPQ

(7.2)

(2.3)

(7.1)

(0.3)

0.0112

0.004

0.011

0.0005W

eigh

t los

s, g

/in.2

Wei

ght l

oss,

g/m

m2

Corrosion resistance of various surface treatments on steel based onfield immersion tests. Test conditions: Full immersion for 24 h in 3%

sodium chloride plus 3 g/L hydrogen peroxide. Salt bath nitrocarburized with nopost-treatment. Source: Ref 6

Fig. 8


As the ferrous component is immersed into the molten salt, a catalytic sur-face reaction takes place, causing the cyanate to decompose and releasethe nitrogen and carbon for the process.

The decomposition of the salt to supply the nitrogen component of thesurface metallurgy satisfies the solubility limit of nitrogen in iron of 6 to9%. The surface carbon content will be in the region of 1% in a solid solu-tion in the ferrous component surface. In addition (as seen in the traditionalnitriding process), a surface compound zone begins to nucleate in whichthe ε-nitride is the dominant phase in partnership with the γ′-phase.

Dimensional StabilityThe SBN process is conducted at subcritical temperatures of 580 °C

(1075 °F); therefore, thermally induced microstructural phase changesand accompanying volume changes do not occur. This permits the treatedcomponent to be cooled from the SBN bath without drastic quenching,thus minimizing dimensional changes and distortion. Accordingly, com-ponents may be finish machined and/or ground prior to SBN. Any dimen-sional 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 of595 °C (1100 °F). Stampings and other parts with thin cross sections arewell suited for SBN because of the capability for dimensional stability andno distortion.

The SBN Processing SequenceReferring 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


100

50

75

25

0Hard

chromeElectroless

nickelQPQ

88 h

72 h

88 h

72 h

88 h336 h

Sur

face

are

a co

rrod

ed, %

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

Ref 6

Fig. 9


of components that has been uniformly preheated will reduce thermalshock 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 com-pound layer. From the salt bath, the parts are quenched into an oxidizingsalt bath (KQ-500) at a lower temperature, typically 400 °C (750 °F), andheld from 5 to 20 min. This intermediate quench cycle is a key ecologicalfeature of the SBN process since during this exposure to the oxidants,cyanates and any cyanides generated within the nitriding bath and con-tained as part of the dragged-out salt are effectively destroyed. Intermedi-ate quenching also retards the cooling rate, thus minimizing thermallyinduced distortion.

After the oxidizing quench, parts are cooled to room temperature,rinsed, and, if required, subjected to post treatment. This may includemechanical polishing, if surface finish is of concern, or the Kolene QPQtreatment to develop maximum corrosion protection and/or enhance thecosmetic appearance. The QPQ treatment provides a lustrous dark finish.

Process ControlAn important factor in producing the desired monophase ε iron nitride

when nitrocarburizing is control of the nitrogen and carbon activities ofthe processing environment. For SBN, this is accomplished by monitoringand regulating the cyanate (CNO–) concentration within the operating saltbath to 34 to 38%. As more parts, and thus more iron, are put through thebath, cyanate content decreases until regeneration becomes necessary.Based on results of standard titration analyses, an appropriate amount ofan organic polymer regenerant is added. This reacts with the carbonate inthe bath and converts it to cyanate.


Fig. 10 Kolene QPQ process cycle. Source: Ref 9


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

Fig. 11

Metallurgical ResultsThe SBN process can be performed on a number of ferrous-based materi-

als, including carbon steels, low-alloy steels, tool steels, and stainless steels.Special nitriding grades respond to SBN, but the presence of nitride-formingalloying elements is not required to produce a compound zone. Formation ofnitride 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 toSBN as measured by depths of compound zones and diffusion zones formedand by the associated hardnesses of both zones is directly related to the mate-rial composition. In general, with an increase in carbon content and/or in theamount of alloying elements, the rate of diffusion decreases, thus resulting inshallower depths of nitrogen penetration (Fig. 12a and b). Diffusion rates areparticularly sensitive to materials containing nitride-forming elements suchas chromium and molybdenum.

Carbon and Low-Alloy Steels. The SBN process time cycles are mostoften limited to 2 h since diffusion rates decrease with time and long cyclesbecome increasingly nonproductive. Typically, a compound zone depth of7.5 to 20 µm (0.0003 in. to 0.0008 in.) would be produced in 90 min forcarbon or low-alloy steels with diffusion zone depths ranging from 0.4 to0.75 mm (0.015 to 0.030 in.), depending on the material (Fig. 13a and b).

Tool Steels. Compound zone depth requirements and associated timecycles for tool steels depend on the application. High-speed cutting toolsrequire a very thin compound layer and thus are treated for only 10 to 20 minat a reduced temperature (540 °C, or 1000 °F) to maintain base-materialhardness. This provides resistance to chip welding at the cutting edge.

The SBN treatment also benefits die casting and forging dies by reducingheat checking, soldering, diffusion, and wear from molten metals (Fig. 14).Cycle times for hot-work die steels may require 2 h to achieve the desiredproperties. The performance of stamping dies is also increased by reducinggalling and seizing (Fig. 15).



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

Fig. 13

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 specificalloys. Source: Ref 5

Fig. 12


Gross heat checking in a low-alloy tool steel forging die due toexcessive temperature. Heat checking occurred after an undeter-

mined number of 225 kg (500 lb) nickel-base alloy preforms had been forgedfrom an average temperature of 1095 °C (2000 °F). Source: Ref 11

Fig. 14

0 10 20 30 40 50 60 700

8

7

6

5

4

3

2

1

Cycles × 103

SBN

Wea

r, in

. × 1

0–4

Case hardenedto HRC 54

Taber abraser testAISI 1117 Steel

Untreated

0 10 20 30 40 50 60 700

8

7

6

5

4

3

2

1

Cycles × 103

SBN

Wea

r, in

. × 1

0–4

Hardened andtempered toHRC 28

Taber abraser testAISI 1045 Steel

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

Fig. 15


Stainless steels are SBN treated primarily to introduce wear resistanceand antigalling properties. While developing these properties, there couldbe significant degradation of the stainless (corrosion resistance) qualities ofthe material, which should be considered when specifying any nitrocarbur-izing process. Stainless steels respond quickly to diffused nitrogen, devel-oping 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 diffu-sion layer depth of only 75 µm (0.003 in.) (Fig. 17).

Cast irons also respond to SBN but at a somewhat reduced rate. Cycletimes of 90 to 120 min are normally specified to generate compound zonedepths 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 generallytreated to improve wear resistance and increase fatigue strength.


0 20

35

SBN time, h

Com

poun

d la

yer

dept

h, in

. × 1

0–4

Diffusion Curves304 & 316 SS, Annealed

SBN in Nu-Tride*Modified Nu-Tride

400°C (750°F)*

455°C (850°F)*

510°C (950°F)*

580°C (1075°F)630°C

(1165°F)+

30

25

20

15

10

5

4 6 8 10

Compound layer in type 316 stainless steel consisting entirely of S-phase. SBN, 455 °C (850 °F) for 5 h. Marble’s reagent, 1000×.

Source: Ref 12

Fig. 16

Diffusion response of annealed types 304 and 316 austenitic stain-less steel to SBN in Nu-Tride at 400 to 625 °C (750 to 1160 °F).

Source: Ref 13

Fig. 17


Cast irons treated in SBN. (a) Gray iron. SBN, 580 °C (1075 °F) for 2 h, followed by salt bathquenching. 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

Fig. 18


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

Engineering PropertiesThe end result of SBN is the presence of two distinct layers of a nitrogen-

bearing microstructure, the outermost identified as the compound zoneand the subjacent layer called the diffusion zone. Each of the zones con-tributes to improving performance by enhancing specific engineeringproperties such as wear resistance, lubricity, corrosion resistance, andfatigue strength. From these, other benefits in performance are realized,including excellent running-in properties, antigalling, and antiseizingcharacteristics, and reduced tendency for fretting corrosion.

Wear resistance is perhaps the most significant property resultingfrom SBN. The ability of the compound zone to resist wear depends onwhether the wear is adhesive or abrasive. Adhesive wear occurs when twocomponents are in relative motion in an essentially abrasive-free environ-ment. Under these conditions, the intrinsic physical characteristics of thecompound zone (i.e., hardness and lubricity) notably improve the slidingand running-in behavior, and consequently increase the resistance to adhe-sive wear (Fig. 20). The phase composition of the compound zone thatdemonstrates the best wear resistance consists predominantly of ε-nitridephase (monophase preferred) with a very small amount of γ ′-phase.

Resistance to abrasive wear depends on the relative hardnesses of theabrading 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, isto increase the content of nitride-forming elements of the base material.


Progressive increases in chromium from 0% (carbon steel) to 17% (stain-less steel) result in a corresponding increase in hardness up to HRC 70(minimum).

Lubricity is the other engineering property that influences resistance toadhesive wear. The compound zone has an inherent low coefficient of fric-tion (see Fig. 20) and functions as a solid-film lubricant by providing anonmetallic interface between mating surfaces. Applications that are thusbenefited include forming tools, extrusion dies, and sliding and rotatingsystems that rely on good running-in properties.


Microstructures of various ferritic materials that have undergone saltbath nitrocarburizing. All etched in 3% nital. All 500×. Courtesy of

Kolene Corp. (a) Ferritic nodular iron; 90 min at 580 °C (1075 °F), oxidizing moltensalt quench. (b) Low-carbon steel; 90 min at 580 °C (1075 °F), oxidizing molten saltquench. (c) Type 304 stainless steel; 90 min at 580 °C (1075 °F), oxidizing moltensalt quench. (d) AISI D2 tool steel; 90 min at 580 °C (1075 °F), oxidizing molten saltquench. (e) H13 medium-carbon hot-work tool steel; 120 min at 580 °C (1075 °F),oxidizing molten salt quench

Fig. 19


Corrosion resistance increases via SBN, most notably with the Kol-ene QPQ process. Figures 8 and 9 show how QPQ treatments improvecorrosion properties.

Fatigue strength under bending and torsional loading develops as aresult of nitrogen present in the diffusion zone and particularly the amountin solid solution. Improvement in fatigue strength produced through SBNoccurs in a range of materials and in varying degrees (Table 2). Low-carbon steels with relatively low strength display the greatest increase infatigue strength (from 80 to 100%), whereas the higher alloy steels andcast irons exhibit from 20 to 80% improvement in strength.

Other Methods for Salt Bath Nitrocarburizing

Sursulf, a ferritic nitrocarburizing procedure developed in France, alsois based on the principle of the decomposition of a low-percentagecyanide to cyanate that is mixed with a low-temperature carbonate salt forthe activated carbon in the salt. The salt also has a sulfur compound in thesalt analysis to assist with the formation of surface sulfides as well assome surface porosity. The porosity will form small reservoirs to hold sur-face lubricants. The lubricants will also work with the sulfur to form analmost self-lubricating surface with high wear resistance and improvedcorrosion resistance.

Tenoplus is a two-stage high-temperature nitrocarburizing process. Thefirst 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). Sub-sequent cooldown takes place in an oxidizing bath to a lower temperatureof 350 to 400 °C (660 to 750 °F). After polishing, the workpiece under-goes a controlled oxidizing process in a special salt bath (Ref 6).


0.2

0.3

0.4

0.1

0.0Chrome plated,

30 µm(0.0012 in.)

Mating disks same treatment

No lubrication

Lubricated withSAE 30 oil

Casehardened

SBN 90 min580°C

(1075°F)plus SBQ

Coe

ffici

ent o

f fric

tion,

µ

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

Fig. 20


REFERENCES1. T. Bell, Ferritic Nitrocarburizing, Met. Eng. Q., May 1976, reprinted


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 (availableas translation BLL RTS 9520)

4. R. Willing-Lepenies and C. Faulkner, “New Developments in SaltBath Nitrocarburizing,” product literature, Houghton Durferrit, Val-ley Forge, PA

5. G. Wahl and S. Alwart, Improvement of Tribological Propertiesthrough Nitrocarburizing

6. Nitrocarburizing, Surface Hardening of Steels: Understanding theBasics, 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,” productliterature, 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 litera-ture, Kolene Corporation, Detroit

11. Tool Materials, ASM Specialty Handbook, ASM International, 1995,p 228

12. J.R. Easterday, Expanding the Temperature Range for Salt BathNitrocarburizing, Ind. Heat., Vol 70 (No. 1), Jan 2003, p 34–38

13. J.R. Easterday, “Influence of SBN on Corrosion Resistance of Stain-less Steels,” product literature, Kolene Corporation, Detroit


Table 2 Improvement in fatigue strengthAlloy treated Fatigue strength improvement, %

Low-carbon steels 80–100Medium-carbon steels 60–80Stainless steels 25–35Chrome/manganese (low-carbon) steels 25–35Chrome alloy (medium-carbon) steels 20–30Cast irons 20–80

Source: Ref 10


CHAPTER 20Gaseous FerriticNitrocarburizing

GASEOUS FERRITIC NITROCARBURIZING, like salt bath nitro-carburizing, involves the introduction of carbon and nitrogen into a steelin order to produce a thin layer of iron carbonitride and nitrides, the“white layer” or compound layer, with an underlying diffusion zone con-taining dissolved nitrogen and iron (or alloy) nitrides. The white layerenhances surface resistance to galling and wear. The diffusion zone signif-icantly increases the fatigue endurance limit, especially in carbon andlow-alloy steel.

The compound-diffusion layer may contain varying amounts of gammaprime (γ ′) and epsilon (ε) phase, cementite, and various alloy carbides andnitrides. The exact composition is a function of the nitride-forming ele-ments in the steel and the composition of the atmosphere.

Following thorough cleaning (vapor degreasing is adequate for mostapplications), parts are nitrocarburized near 570 °C (1060 °F), a tempera-ture just below the austenite range for the iron-nitrogen system. Treatmenttimes generally range from 1 to 4 h. Although there are a number of pro-prietary gas mixtures, most contain ammonia (NH3) and an endothermicgas. Batch furnaces with integral oil quenches are ideally suited forgaseous nitrocarburizing.

Development of the Process

The gaseous nitrocarburizing process was first developed in 1961 atJoseph Lucas (Industries) Ltd. in England. This treatment produced onmild steel a porous layer that was claimed to have good antifrictional prop-erties. The published patent revealed that the gaseous atmosphere consistedof ammonia and hydrocarbon or other carbon-containing gases of unspeci-fied proportions and that the treatment was undertaken in the temperaturerange of 450 to 590 °C (840 to 1095 °F). During the 1960s, further researchled to the development of a large range of gas nitrocarburizing processes




throughout the world. Some of these processes, most of which are propri-etary, 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 con-stant supply of nitrogen (from ammonia), carbon from a suitable gaseoussource, and oxygen. Ammonia is considered to be the most readily andactively available source of nitrogen for the process and is blended with theother process gas supplies of carbon (from a hydrocarbon gas) and oxygenfrom other sources. As in gas nitriding, the cracked nascent ammonia gaswill dissociate at the steel surface and react with the hydrocarbon gas toform both nascent nitrogen and free carbon. The gases will allow carbondioxide 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 theprocess, a drop in the partial pressure of hydrogen will occur. This in turnwill 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 by-product of the process. The hydrogen cyanide will contribute both nitro-gen and carbon to the process reaction, thus improving the nitridingpotential of the total process gas.

The metallurgical results of the process are very similar to the classicalnitriding 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



The rate of growth of the compound layer is approximately the square rootof the process time at the chosen process temperature.

The primary objective of the ferritic nitrocarburizing process is to formnitrides and, most importantly, surface carbon, which will encourage theε-nitride phase in the surface of the steel. The surface carbon comes fromthree sources:

• Decomposition of the process gas• Steel chemistry• Hydrocarbon additive gas

Gaseous Supply

As stated earlier, gas generally is supplied by using an endothermicallygenerated gas with an addition of ammonia. The percentages are approxi-mately 50% for each gas, depending, of course, on the surface metallurgyrequired.

Some proprietary procedures use an exothermically generated gas asthe primary process gas, followed by the addition of ammonia. Theexothermic gas decomposes to provide a small percentage of carbonmonoxide, along with hydrogen and nitrogen from the ammonia, whichwill decompose in a catalytic reaction at the steel surface.

In some proprietary procedures, methane is added to promote ε-nitrideformation in the steel surface. The procedure can then be followed with adeliberate and controlled surface oxidation by adding a source of oxygento 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 oxideswith a low degree of surface porosity.

Properties of Gaseous Ferritic Nitrocarburized Components

The properties obtained as a result of the gaseous ferritic nitrocarburiz-ing process are as varied as those obtained with other process techniques.Surface properties, however, can be modified simply by modifying theprocess parameters.

The compound zone properties formed by the ferritic nitrocarburizingprocess are generally very similar. There will be a significant difference ifthe immediate surface has been oxidized after completion of the ferriticnitrocarburizing treatment. The nonoxidized surface will show a signifi-cant increase in:

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

Chapter 20: Gaseous Ferritic Nitrocarburizing / 221


These properties are determined by traditional methods of property meas-urement.

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-alloysteel. However, the process can also be applied to medium- and high-carbonsteels with various metallurgical results. Typical components that undergogaseous 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 nitrocarbur-izing, he or she should consult with the heat treater or metallurgist to dis-cuss process advantages and limitations as well as the metallurgicalbehavior of the part after the process treatment. Some important consider-ations for three important application areas are discussed below.

Die Casting. The die casting industry needs surface treatments thatwill improve wear resistance, improve release properties, and resist thebuildup or welding of aluminum, magnesium, and zinc on mold and coresurfaces. This has led tool and die makers and mold makers to review theproperties and performance of ferritic nitrocarburized treatments in termsof applications and performance on tool steels.



Gear Heat Treatment. Most gear manufacturing depends on controltolerance and precision treating in terms of distortion. This means that aprocess that can offer a significant reduction in distortion and an improve-ment in gear performance will be closely scrutinized. Ferritic nitrocarbur-izing is performed at a low temperature (in the ferrite region) and intro-duces both carbon and nitrogen into the surface of the steel, followed by aquench to improve fatigue resistance with a minimal amount of distortion.Gear manufacturers are finding that the process can be applied to gear cut-ting hobs, cutting tools, and broaching tools. The process also improvesantiscuffing properties and compression stress at the gear surface.

Sintered Steels. Extensive research work has been done on sinteredpowder metallurgy (P/M) ferrous alloys, and ferritic nitrocarburizing isnow a fully commercial process for the P/M industry. The degree of suc-cess depends on the level of compaction of the sintered part. Both gaseousand plasma nitrocarburizing processes have found commercial use.

Safety Considerations

Because the gaseous ferritic nitrocarburizing process involves a com-bustible atmosphere that is explosive when operated below the self-ignitiontemperature, safety precautions must be rigorously enforced. It is mandatorythat all safety purge systems are in place and fully operable and that the cor-rect purge sequencing has occurred before combustible gases are admitted tothe 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 gasis not admitted according to the furnace manufacturer’s written procedures.

Appendix: Gaseous Nitrocarburizing—A Suitable Alternative for the Heat

Treatment of Automotive CrankshaftsK. Bennett, BOC Ltd.

and Q. Weir and J. Williamson, Leyland Vehicles Ltd.

The remarkable properties conferred on low-alloy steels by nitrocarbur-izing have been highlighted comprehensively in the literature in the pastfew years (Ref 4–6). In addition to enhanced fatigue resistance, theincrease in wear resistance, attributed to the formation of a thin compoundlayer (Fig. 1) composed of the single-phase ε-nitride or carbonitride, hasbeen shown to be dramatic. Typically, a compound layer of some 12.5 to20 µm has been reported as a result of a conventional 3 h nitrocarburizingtreatment at a temperature of 570 °C (1060 °F) (Ref 4).



It has been suggested recently (Ref 7) that the properties obtained bythe 60/80 h gaseous nitriding of classical nitriding steels can be matchedby a 12/20 h nitrocarburizing treatment (Fig. 2). In this context, an addedadvantage of nitrocarburizing would be the replacement of the brittleduplex white layer, characteristic of gaseous nitriding, by the more ductilesingle-phase ε compound layer. This would eliminate the frequent neces-sity for postheat treatment machining.

Six-cylinder engine crankshafts, manufactured in B.S. 970: 1970708A42 steel (En 19C), are normally nitrided for a process period of 60 h.The service conditions in respect of the crankshaft journals are such thatwhile good wear resistance is required, a high indentation resistance is notessential. A prime requirement exists for a good fatigue resistance.

Following consultation with BOC Ltd., Leyland Vehicles Ltd. decidedto initiate a program of work aimed at evaluating the properties to beobtained by short-cycle nitrocarburizing (3 h duration in this instance) ofsuch automotive crankshafts. If it were possible to produce crank-shaftswith 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 hard-ness tests, microhardness traverses, metallographic examination, andfatigue tests. Such tests would be reinforced by both engine testing androad trials on the actual components.

Process Technique

The geometry of the components is such that it was considered essentialto treat vertically in order to control distortion (Fig. 3). The only suitableequipment available at Leyland for this purpose was a Wild Barfield pit fur-


Fig. 1 Typical compound layer on nitrocarburized mild steel


nace which was normally used for gas carburizing. This is shown schemati-cally in Fig. 4. Some reservations were expressed as the normal techniqueof venting ammonia-bearing exhaust gases to atmosphere could not beaccommodated. The location of the furnace and the height of the heat treat-ment shop presented additional problems in this respect. However, it was


Comparative hardness profiles produced by 80 h nitriding and 16 hnitrocarburizing

Fig. 2

Fig. 3 Furnace load of crankshafts


found in practice that the exhaust burnt quite readily during processing atthe nitrocarburizing temperature. The furnace was modified by the inclusionof a pilot flame at the exhaust pipe. It was decided that the furnace atmos-phere would be the standard nitrogen/ammonia/carbon dioxide nitrocarbur-izing system (Ref 5–7). A further furnace modification was carried out tofacilitate the separate delivery of carbon dioxide to the furnace chamber. (Itis an accepted fact that, should ammonia and carbon dioxide be deliveredthrough the same manifold, then a reaction will occur between the twogases. The manifold will quickly become blocked with heavy deposits ofammonium carbamate.)

With a furnace volume of 2 m3 (70 ft3), the standard purge atmosphereflow employed equated to six volume changes per hour. During process-ing, the standard atmosphere flow was four volume changes per hour. Thefull 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.


Schematic diagram of the Wild Barfield pit carburizing furnace used inthe nitrocarburizing trials

Fig. 4


• Raise furnace temperature to 570 °C (1060 °F) under nitrogen atmo-sphere only.

• When furnace attains a temperature of 570 °C (1060 °F), adjust atmo-sphere makeup to effect nitrocarburizing, while maintaining a processflowrate of four volume changes per hour. A nominal ingoing atmo-sphere 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 furnacetemperature.

• 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 temperaturebelow 400 °C (750 °F) in order to control dimensional movement and toavoid possible cracking in thin-wall areas. The cooling rate below 400 °C(750 °F) would not appear to affect the metallurgical properties of thetreated components. It was evident, however, that the slower-cooled com-ponents 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 hnitrocarburizing were as discussed below.

Hardness. Surface hardness of the order of 580 to 650 HV5 wereachieved. Hardness traverses on sectioned samples indicated microhard-nesses in the compound layer in excess of 700 HV0.01, with the highestvalues being recorded close to the interface of the compound layer and thesubstrate. 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 unbrokencompound layer (Fig. 6). The depth of the compound layer was of theorder of 25 to 37.5 µm, which is considerably deeper than that expected ofa conventional nitrocarburizing treatment of the same duration on thesame material.

Fatigue Resistance. Fatigue testing of treated crankshafts was carried outat 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 treatmentwere run in a six-cylinder turbocharged engine on an experimental testbed 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. 6 Compound layer on a crankshaft after 3 h nitrocarburizing

Fig. 7 Fatigue tests results

Fig. 5 Typical hardness profile on a crankshaft after 3 h nitrocarburizing


Conclusions

The results of the trials demonstrated that:

• The wear resistance conferred on a 708A42 crankshaft by a 3 h nitro-carburizing treatment in a nitrogen-based atmosphere is substantialand comparable with that produced by conventional 60 h gas nitridingof the same material. In view of the fact that the effective case depthof the nitrocarburized component is approximately 50% of that of anitrided component, the wear performance can only be attributed tothe properties of the compound layer.

• The fatigue resistance obtained by 3 h nitrocarburizing is within speci-fication for the automotive crankshaft.

• The adoption of a short-cycle nitrocarburizing treatment in thisinstance would yield a saving in heat treatment costs of some 85%.

In addition it has been shown that the nitrocarburized surface facilitatesthe use of a cheaper bearing than that required for the more friable surfaceof the nitrided component.

ACKNOWLEDGMENTThis appendix was reprinted with minor changes from K. Bennett,

Q. Weir, and J. Williamson, Gaseous Nitrocarburising—A Suitable Alter-native for the Heat Treatment of Automotive Crankshafts, Heat Treatmentof Metals, Vol 8 (No. 4), 1981, p 79–81. Reproduced by permission ofWolfson Heat Treatment Centre.

REFERENCES1. T. Bell, Gaseous and Plasma Nitrocarburizing, Heat Treating, Vol 4,

ASM Handbook, ASM International, 1991, p 425–4362. T. Bell, Ferritic Nitrocarburizing, Met. Eng. Q., May 1976, reprinted


3. J. Grosch, Heat Treatment with Gaseous Atmospheres, Steel HeatTreatment Handbook, G.E. Totten and M.A.H. Howes, Ed., MarcelDekker, 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 Nitrocarburis-ing and Nitriding When Applied to Design and Manufacture of Non-Alloy 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



6. W.I. James, Practical Experience with Nitrogen-Based Nitrocarburis-ing, Heat Treatment of Metals, Vol 6, 1979, p 13–15

7. K. Bennett, Advances in Nitrogen-Based Nitrocarburising, Proceed-ings of the 18th International Conference on Heat Treatment ofMaterials (Detroit, 6–8 May 1980), p 146–160



CHAPTER 21Equipment 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 cleanlinessand effluent disposal. Good housekeeping of the salt bath work area isessential not only in terms of cleanliness but also personal safety.

Effluent disposal is a subject addressed by all of the process salt manu-facturers, who now produce process salts with very low cyanide contents,to less than 4 wt%. Although the salt still contains cyanide, decompositionand subsequent neutralization of the cyanide effluent are much easier.

Salt bath equipment used for nitrocarburizing is essentially similar indesign to salt bath furnaces used for other processes. Although batchinstallations are most common, semicontinuous and continuous opera-tions 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 isnonreactive with the process salt. The pot can be made entirely of titaniumor can be titanium lined to reduce the capital cost of the process bathequipment.





Internally heated salt bath furnace with immersed alloy electrodes andceramic tile lining

Fig. 1

Internally heated salt bath furnace with submerged graphite electrodesand a modified brick lining for use with carburizing salts

Fig. 2


Atmosphere Furnace Equipment

Atmosphere nitrocarburizing requires a furnace that is both gastight andairtight to ensure safety of the operator and the equipment. Furnace designis of paramount importance to prevent an ingress of unwanted oxygen.

The furnace can be either the cold-wall vacuum design for partial pres-sure processing or the integral quench atmosphere design. It should benoted that because ferritic nitrocarburizing takes place below the atmos-phere design limits of an integral quench furnace, the furnace manufacturershould be consulted. Integral quench furnace doors are typically safetyinterlocked. On a normal integral quench furnace, the external door cannotbe opened at a chamber temperature below 760 °C (1400 °F). This featureis to eliminate the risk of an explosion. Consult the furnace manufacturerregarding low-temperature operations such as ferritic nitrocarburizing.

Plasma-Assisted Furnace Equipment

Plasma-assisted ferritic nitrocarburizing systems (Fig. 3) often differfrom plasma nitriding systems in terms of process vessel materials of con-struction. This is because the equipment operates at process temperaturesup 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 processconducted in a partial-pressure condition. A major difference is that theplasma-assisted ferritic nitrocarburizing furnace usually will have an

Chapter 21: Equipment for Ferritic Nitrocarburizing / 233

Typical plasma nitrocarburizing furnace and associated control system.Courtesy of Plateg GmbH

Fig. 3


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 MetallurgyBecause most steels can be treated by plasma-assisted ferritic nitrocar-

burizing, it is necessary to thoroughly understand the steel beingprocessed in relation to process capabilities. In the case of low-alloysteels, there are insufficient alloying elements to accomplish any real sur-face strength. Because all steels will readily absorb nitrogen at tempera-tures above 205 °C (400 °F) but do not always form stable nitrides (otherthan iron nitrides), iron nitride formation can be encouraged by supplying


Electric fan motor treated by the Nitrotec process (right) and zinc-plated (left). Both were subjected to a 250 h neutral salt spray test.

Fig. 4

Towing ball hitch treated by the oxynitride process (right) anduntreated (left). Both subjected to salt spray test. Courtesy of Plateg

GmbH

Fig. 5


additional nitrogen. This is accomplished by increasing the nitrogen-to-hydrogen ratio at the flowmeters (or mass flowmeters). The classical gasnitriding 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 temper-

ature of around 500 °C (925 °F) but with a resulting hardness of onlyabout 35 HRC. Corrosion resistance will improve dramatically but notsurface hardness. A higher process temperature of approximately 580 °C(1075 °F) should be considered, and temperatures can reach as high as620 °C (1150 °F). Determination of the process temperature depends onthe steel analysis and the surface metallurgical requirements.

A plain low-carbon steel also requires a higher nitriding potential. Thisis accomplished by simply increasing the process gas nitrogen-to-hydrogenratio by volume to a ratio of approximately 5:1 and up to 7:1. This willresult in surface hardness values of up to 700 HV (59 HRC). As previouslystated, the carbon source can come from the steel itself or by the addition ofa hydrocarbon gas to the process gas flow. In the case of a plain low-carbonsteel, there is insufficient carbon in the steel to promote formation of thedominant ε-nitride phase; therefore, hydrocarbon gas (methane or propane)must be added. The usual recommendation is approximately 2% by volumeof the total gas flow of nitrogen and hydrogen. This will be sufficient to ini-tiate and complete formation of the ε-phase in the steel surface.


Fig. 6 Typical plasma nitriding PLC screen display. Courtesy of Plateg GmbH


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 fromprocess monitoring points on the furnace. The computer will then respondto any set point signal deviation and take whatever corrective action isnecessary (Fig. 6). To accurately control the gas flow, mass flow con-trollers are used. A mass flow controller can control flow rates rangingfrom centiliters per hour to liters per hour far more accurately than a con-ventional 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 duringthe process time. A general estimate is approximately 1.5 L/min/m3 of thevolume 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 perU.S. gallon:

1.5 × 60 min =90 L/h

= 22.5 gal by volume of process gas4

This makes plasma-assisted ferritic nitrocarburizing cost competitivewhen compared with the conventional gaseous procedure. The reason forthe lower process gas consumption is that gas is being consumed strictlyfor plasma generation and is not being used as a “sweep gas” to preventgas stagnation within the process chamber. Gaseous nitriding and ferriticnitrocarburizing processes often require an overpressure condition toensure gas uniformity throughout the process chamber, particularly whenconsidering blind holes.

In addition, shorter process cycles greatly improve equipment productiv-ity. If there is a continual flow of workload for the plasma nitride furnace,turnaround time is faster than for gaseous nitrocarburizing. Faster floor-to-floor cycle times, the elimination of postcleaning operations, and the poten-tial 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 carburizingother than by using high temperatures of around 1040 °C (1900 °F).



The rate of solid-state diffusion of any element into the surface of anysteel 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 ofphysics will allow it to go. Many claims are made of very deep casedepths being accomplished with very short cycle times at process temper-atures in the region of 580 °C (1075 °F). The problem here is that “casedepth” usually is not defined. Is the claim being made for total case depthor effective case depth? What is meant by core hardness? Is it the coreplus 5 HRC points, or is it the actual core hardness itself? This is signifi-cant 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 themetallurgist than has been previously thought. It would make great senseto dispense with the carburize process and go with the ferritic nitrocarbur-ize 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 timeand temperature for carburizing can serve as an approximate guide forplasma-assisted ferritic nitrocarburizing (Ref 1). The guide is based on aplain low-carbon steel, using the formula:

Case depth = K √t

where the case depth is in inches (for case depth in millimeters divide by25.4), 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 thealloying content of a steel is increased by the addition of chromium,


Table 1 Temperature factor for estimating case depthTemperature

°C °F Temperature factor, K

495 925 0.00046510 950 0.00056525 975 0.00068540 1000 0.00081550 1025 0.00097565 1050 0.00116580 1075 0.00136585 1085 0.00146595 1100 0.00160605 1125 0.00187620 1150 0.00218635 1175 0.00252650 1200 0.00291660 1225 0.00334675 1250 0.00382


molybdenum, aluminum, tungsten, vanadium, manganese, and especiallynickel. It is difficult to say by how much the diffusion will be retardedbecause of the many available alloying element variations. For a steelcontaining all of these elements (not accounting for percentage varia-tions), 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 nitrocar-burizing procedure. Oxynitrocarburizing involves introducing oxygen—in the form of steam, oxygen, or nitrous oxide—in a controlled mannerinto the process chamber.

The process gas must be carefully selected. Steam may cause problemswith electrical equipment such as power feedthroughs and control sys-tems. Nitrous oxide or oxygen is preferred; nitrous oxide tends to be moreuser friendly to valves and control systems than oxygen. The thickness ofthe oxygen-bearing compound zone after treatment will be determined bythe time at temperature and the cooldown time. A typical part that hasbeen ferritic oxynitrocarburized is shown in Fig. 7.

The primary reason for oxygen treatment after ferritic nitrocarburizing isto enhance surface corrosion resistance. The procedure is comparable to


Fig. 7 Milling cutter after ferritic oxynitrocarburizing. Courtesy of Plateg GmbH


the black oxide type of treatment and will enhance the cosmetic surfaceappearance of the steel component (Fig. 8). Just as with salt bath treat-ments, the surface finish of the component will depend on the surface finishof the component prior to treatment. The higher the polish of the compo-nent prior to ferritic nitrocarburizing, the better the finish after the oxidiz-ing procedure.

The surfaces of oxynitrocarburized parts have been subjected to saltspray corrosion tests, exceeding 200 h by a considerable margin. Resis-tance to the salt spray test will be determined by the resulting oxide layerthickness.

The oxidization treatment following ferritic nitrocarburizing has almostno cost attached to it, other than a portion of the amortization of the equip-ment. Generally, there is little or no power consumption. The only otherassociated expenses would be the cost of the oxidation process gas, fol-lowed by furnace occupancy.

REFERENCES1. Heat-Treating, Cleaning, and Finishing, Vol 2, Metals Handbook, 8th

ed., p 98


Shafts treated by the oxynitride process (right) and untreated (left). Bothsubjected to salt spray test. Courtesy of Plateg GmbH

Fig. 8



CHAPTER 23Evaluating the Process

PROCESS EVALUATION begins just as for the traditional nitridingmethods. The same precautions must be taken so as to not disturb the sam-ple face being investigated for case depth, case hardness profile, andvisual 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 thecustomer, and whoever processes the work must be very clear as to whatis required in terms of case depth.

The most effective and accurate method of measuring case depth is by amicrohardness traverse from the surface through the case of a sectioned,preground, and polished sample. Optical examination via light micro-scopy will show the case metallurgy, from which an assessment of thecase depth can be made, but this method is not completely accurate.

The load mass for the traverse will be determined by the accomplishedcase depth to be measured. Shallow case depths should be measured witha light load, up to 200 g. The load selection can go as low as 10 g; how-ever, as the impression becomes smaller it will be become difficult to seeand measure with an older microscope eyepiece.

Case Hardness

It is strongly recommended that the case hardness not be measureddirectly on the steel surface with a heavy load (for example, at 150 kg as




used with the Rockwell C scale). The indenter is likely to penetratethrough the formed case and into the core, thus giving a false hardnessreading. 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 upto 10 kg (Fig. 1).

The sample surface should be clean and free from decarburization andmust be presented squarely 90° to the central axis of the penetrator. Anydeviation from this will result in a false reading.

The microhardness tester is usually used for a case traverse at rightangles to the case (Fig. 2, 3). Microhardness testing requires a highly pol-ished specimen surface in order to clearly see the diamond indenterimpression in the case (Fig. 4). The sample must be held firmly in anappropriate 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 theprocess control. First, poor precleaning of the material surface can reducenitrogen diffusion into the steel. The steel surface must be clean and freefrom contaminants such as chlorides, sulfides, silicon products, cuttingoils, and lapping compounds.


Fig. 1 Vickers hardness testing machine


Chapter 23: Evaluating the Process / 247

Microhardness testing system with features including cameras andcomputer imaging. Courtesy of NewAge Industries

Fig. 3

Fig. 2 Microhardness testing machine


The low hardness generally indicates that the case has not beenprocessed to the correct depth specification. In other words, the case depthis too shallow and the indenter punches through it. The remedy is to eitherincrease the case depth requirement or, if the case depth is correct, select alighter load for the hardness test.

The second reason is an imbalance of the furnace atmosphere. If thegaseous method of ferritic nitrocarburizing is being used, check the gasflow rates and relative volumes. Next, check the gas decomposition withinthe furnace. This can be accomplished by a gas analyzer that measures thecarbon monoxide, carbon dioxide, and hydrogen contents and adjusts thegas composition and flow rates accordingly.


Typical fixtures used for holding and clamping workpieces for micro-hardness testing

Fig. 5

Fig. 4 Comparison of identations made by Knoop and Vickers indenters


If using the salt bath method of ferritic nitrocarburizing, analyze the saltcomposition for cyanide and cyanate composition. If the resulting analysisshows an excess of carbonate or cyanate, regenerate the bath. This is accom-plished simply by raising the bath temperature to 595 °C (1100 °F) and vig-orously aerating the bath for approximately 90 min, followed by desledging.

If the plasma-assisted technique of ferritic nitrocarburizing produces ashallow case, then:

• The process cycle time is too short to accomplish the required casedepth (assuming that the gas flow rates are correct).

• The gas flow rates could be incorrect. Increase the nitrogen flow rateto increase the nitriding potential of the process.

In the fluidized-bed method of ferritic nitrocarburizing, a shallow casecan occur due to low nitrogen volumes. Set the process gas flows accu-rately at the outset.

Corrosion

If surface corrosion develops after ferritic nitrocarburizing, the initiat-ing source must be established. For example, is the corrosion originatingfrom an external environmental source? If so, check the operating atmos-phere or liquid for acidic content and adjust the process parametersaccordingly. The corrosion may also be generated by an uncontrolled saltbath, gas mixture, or plasma condition. If this is the case, check that thesalt bath composition is within specified operating limits and adjustaccordingly. Check also that the gas dissociation and gas ratios arewithin the required operating limits and that the process temperature iscorrect. Lastly, check that the plasma conditions are correctly set in termsof power density, voltage, amperage, pressure, gas ratios, and processtemperature.

Material selection may also play a role in initiating corrosion. Does theselected material have the necessary corrosion resistance for the givenenvironment? Even with improved corrosion resistance imparted by thenitrocarburizing treatment, a more corrosion-resistant base material mayhave to be selected.

Distortion

Distortion is always a contentious subject. Despite the low nitrocarbur-izing process temperature, there is no absolute guarantee that distortionwill not occur. Distortion can be classified as either shape distortion(warpage) or size distortion (Ref 1).

Shape distortion originates from induced residual stresses from suchoperations as forging, rolling, and machining and will result in twisting,

Chapter 23: Evaluating the Process / 249


bending, and ovality. The degree of distortion that occurs will be deter-mined by the amount of residual stress present in the steel. The residualstress can be in the steel as a result of mixed phases due to incompletetransformation at the preharden and temper operation. The only way toreduce the induced mechanical stress is to stress relieve prior to finalmachining and prior to nitriding. To deal with the possible mixed phaseproblem, resulting from pre-heat-treatment, it may be necessary to cryo-genically treat the steel, followed by tempering to deal with any trans-formed retained austenite to martensite. This will stabilize the steel priorto the ferritic nitrocarburizing procedure if retained austenite was present.

Size distortion results from thermochemical treatments such as ferriticnitrocarburizing, and from variations in the steel chemistry from heat toheat. If the steel has been prehardened and tempered, the prior austenitiz-ing temperature, as well as time at the austenitizing temperature, willinfluence grain size (Ref 2).

The flow diagram in Fig. 6 presents factors that influence distortionduring final heat treatment. Remember, both ferritic nitrocarburizing andnitriding will act as a stress-relieving process. If residual induced stresshas not been dealt with beforehand, then the procedure will act as a stressrelieve operation and distortion will occur. Steel will respond to heat fromany source.

REFERENCES1. G.E. Totten, C.E. Bates, and N.A. Clinton, Handbook of Quenchants

and Quenching Technology, ASM International, 1993, p 4432. G.E. Totten and M.A.H. Howes, Chapter 5, in Steel Heat Treatment

Handbook, Marcel Dekker, 1997, p 251–292


Fig. 6 Flow diagram showing factors contributing to distortion

Meltingchemistryvariations

Rollingor forging

temperature

Annealingtemperature

Normalizetemperature

Rough machine

Preheattreatment

Stress relieve

Final machine Ferriticnitrocarburize

Distortion


Index

A

Alloy steelsAISI 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, 161salt 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. 3Ammonia. See also Gas nitriding

decomposition of 3, 23–27, 50–51, 77–78,77(F), 117–118, 185–186, 220

leaks in retorts 44–46safety precautions 46–47

Applications 153–161Aubert, Pierre 6Automotive crankshafts, gaseous ferritic

nitrocarburizing of 223–229

B

Bell, Thomas 67, 201Berghaus, Bernhard 9, 72British standard nitriding steels 5, 5(T)

C

Carbon, and compound zone 32Carbon steels

AISI 1015, nitrocarburized microstructure35(F)

AMS 6470, nitride networking 40(F)AMS 6470, nitrided case 181(F), 182(F)hardness 35, 203(F)

nitriding 35–36for retorts 43SAE 5115, salt bath nitrided 57(F)salt bath ferritic nitrocarburizing 211, 211(F),

212(F), 216(F)Case depth. See also Nitrided case; Surface

hardnessautomotive crankshafts 227carburizing vs. nitriding 13(F), 14(F), 15–16,

15(F), 16(F)determination of 68–69, 69(T), 136–137extrusion dies 155ferritic nitrocarburizing 245fluidized–bed nitriding 117(F), 118(T)forging dies 154gears 159, 160(F)ion ferritic nitrocarburizing 236–237, 237(T)ion nitriding 136–137microhardness testing 177, 245–248, 246(F),

247(F), 248(F)temperature factor 69(T), 137(T), 237(T)

Case formation 31–37. See also Compoundzone (layer)

Central Alloy Steel Corp. 7Compound zone (layer)

in automotive crankshafts 223, 224(F)carbon, influence of 32control of 65–69, 82–83dual phase formation 4(F), 31(F), 32(F),

33(F), 65(F), 66(F), 141(F)early studies 7in extrusion dies 83(F), 155–156and Floe process 8–9, 34, 66, 67(F)in forging dies 154in gears 159kinetic studies 142–150and Melonite process 203–204, 204(F)and sputtering 142–150tests 66thickness 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)

Index_Nitriding.qxd 10/1/03 3:37 PM Page 251



Contaminants. See Surface contaminantsCopper

deoxidized, for retorts 44for selective nitriding 163–164

Core hardnessautomotive crankshafts 227extrusion dies 155(F)forging dies 153, 154(F)gears 159high-speed steel cutters 157and preheat treatment 28–29

Corner effect. See Nitride networkingCosts, various surface treatments 198(F)Crankshafts, gaseous ferritic nitrocarburiz-

ing of 223–229Cutters, high-speed steel 156–157Cutting fluids. See Surface contaminants

DDawes, Cyril 195Decarburization. See Surface contaminantsDegussa 54, 55, 196, 201Deposition, by ion nitriding 76–77, 157Die casting, and gaseous ferritic

nitrocarburizing 222Dies

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, 67–68Distortion

control of 18–20, 120and ferritic nitrocarburizing 194, 249, 250(F)in gears 158–159growth 19–20, 19(F), 27–28, 27(F), 120,

121–122, 158–159postmachining 122–123shape 18, 120, 121, 249size 18, 119–120, 249stock removal 122

Double-stage process. See Floe process; Two-stage process

EElectroplating 163, 164Enameling, for retorts 44Equilibrium 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 177–179Examination methods 167–183

FFerritic nitrocarburizing. See also Ferritic

oxynitrocarburizing; Gaseous ferriticnitrocarburizing; Ion ferritic nitrocar-burizing; Salt bath ferritic nitrocarburizing

advantages 193–194, 198–199corrosion 248–249costs 198–199, 198(F)distortion 249, 250(F)early history 195–198equipment 231–239hardness testing 245–248process evaluation 245–250surface preparation 241–243training 199–200

Ferritic oxynitrocarburizing 234, 234(F),238–239, 238(F), 239(F)

Fingerprints. See Surface contaminantsFirth Brown Steels 5Fixtures. See Trays and fixturesFloe, Carl F. 9, 34, 66Floe process 8–9, 26. See also Two-stage

processFluidized-bed nitriding

furnaces 111–118, 112(F), 113(F), 114(F),115(F)

gas dissociation 117–118oxynitriding 116

Fry, Adolph 4–5, 6, 7, 8, 14, 20, 127Furnaces

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)

GGas ionization. See Ion nitridingGas nitriding. See also Nitriding

case crushing 187discoloration 186exfoliation 186–187furnace design 8(F), 23(F), 39–51, 39(F),

41(F), 45(F)furnace heating 47–49





gas circulation 40–41, 41(F)gas dissociation 50–51, 185–186insulation 42–43leak prevention 44–46orange peel effect 187oxygen probe control 51process control and instrumentation 49–51retort construction and maintenance 43–44safety precautions 46–47sensors 51stop-off procedures 163–164surface contaminants 185, 186–187temperature control 40, 49–50troubleshooting 185–187

Gaseous ferritic nitrocarburizingapplications 222–223of automotive crankshafts 223–229early history 195–196furnaces 224–227, 226(F), 233gaseous supply 221process development 219–220process principles 220–221safety considerations 223surface cleanliness 241–242surface 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, 72Glass, for retorts 43–44Glow discharge process. See Ion nitridingGrowth. See Distortion

H

Hardness. See Core hardness; Hardness testing; Surface hardness

Hardness testingof automotive crankshafts 227, 228(F)hardness profiles 168, 168(F), 169(T), 228(F)macrohardness 167microhardness 167, 177, 245–248, 246(F),

247(F), 248(F)sample preparation 168–177

Harris, F.E. 69Heat input requirement 48–49High-speed steels

compositions 126(T)cutters 156–157

Homerberg, V.O. 8Houghton Durferrit 202, 203

I

Imperial Chemical Industries 54, 195Insulation, for gas furnaces 42–43Interstitial diffusion 24, 25(F)

Ion ferritic nitrocarburizingcase depth 236–237, 237(T)corrosion resistance 234, 234(F), 238–239,

239(F)early history 197–198furnaces 233–237, 233(F)gas consumption 236oxynitrocarburizing 234, 234(F), 238–239,

238(F), 239(F)process control 234–235surface cleanliness 243trade names 202(T)

Ion nitridingadvantages 85–86, 107–108of blind holes and cavities 75, 102–103,

103(F), 165(F)case depths 136–137cathode and anode 93–94cold-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 76–77diffusion techniques 76early developments 9–11, 71–72equipment 89–109and ferritic nitrocarburizing 197–198furnace loading 105furnaces 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), 103–104hollow 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 161masking 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, 98–101process control 75–76, 84, 90, 105, 139–142process gas flow 92, 139–142process gas ratios 77–78process gases 83process principles 72–73, 73(F), 77–80of pure irons 160sputter cleaning 104–105, 104(F), 164–165sputtering 142–150of stainless steels 129–136stop-off procedures 105–106, 164–165, 165(F)surface degradation 81surface reactions 78–80surface stability 80thermocouples 91–92troubleshooting 188–191vacuum pumps 92–93, 93(F), 94(F)work cooling 101–102

Index / 253




Ironscast, 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. 67Jeffries, Zay 6Jones, Claude 10, 72Joseph Lucas Ltd. 195, 219

K

Ketcham, W.J. 6–7Kinetic studies, of compound zone formation

142–150Klockner Ionen GmbH 9, 72Kolene Corp. 55, 207, 208Krupp Steel Works 4, 5, 6, 14, 127

L

Leyland Vehicles Ltd. 224Light microscopy. See Microscopic

examinationLightfoot, B.J. 67Ludlum Steel Co. 8

M

Machlet, Adolph 3–4, 6, 13, 14, 67Maraging steels 160–161, 183(F)Martin, Stuart 10, 72Masking. See Selective nitridingMass spectrometry, for process gas control

analysis 140–141Massachusetts Institute of Technology 8, 9, 66McQuaid, H.W. 6–7Melonite process 55, 202, 203–207Metallographic examination. See

Microscopic examinationMicrohardness. See Hardness testingMicroscopic examination

of automotive crankshafts 227, 228(F)etching 177–180microscope selection 180–181nitrided 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. 68Nitride networking 5, 16, 34, 34(F), 40(F),

80–81, 80(F), 81(F), 154–155Nitrided case. See also Case depth; Surface

hardnessalloying elements, effect of 128(F)examination methods 167–183on 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 153–161current status of technology 11examination methods 167–183furnace equipment and control systems

39–51historical background 3–11, 13–14metallurgical considerations 1–2microstructures 31–37, 32(F), 35(F), 181(F),

182(F), 183(F)process principles 23–29process requirements 2–3, 14–21steels for 125–137, 153–161troubleshooting 185–191U.S. vs. German processes 5–6

Nitriding potential 26, 67Nitrocarburizing. See Ferritic

nitrocarburizingNitrogen diffusion 24–27, 33–34, 139–142

See also Solubility limitNu-Tride process 207, 208–217

O

Oils. See Surface contaminantsOptical light microscopy. See Microscopic

examinationOxidation, resistance to 21Oxygen probes 51Oxynitriding 86–87, 87(F), 116. See also

Ferritic oxynitrocarburizing

P

Paint residue. See Surface contaminantsPaints, stop-off 164–165Paschen curves 74, 74(F), 96, 96(F)Phase transformations 17(F), 18, 18(F), 24,

26–27, 28–29Photo spectrometry, for process gas control

analysis 139–140, 140(F)





Piston rods, oxynitrided 87(F)Plasma nitriding. See Ion nitridingPlasma-assisted chemical vapor deposition

76–77, 157Plasma-assisted nitrocarburizing. See Ion

ferritic nitrocarburizingPreheat treatment 28–29Process chambers. See RetortsProgrammable 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 16–18Refractory firebrick, for retorts 44Resins, molding 171(T), 172(T)Retorts

construction 43–44maintenance 44sealing 44–46, 45(F)

Rocker arms, wear resistance 206–207,207(F)

Ryzhov, N. 139

S

Safety precautionsfor ammonia use 46–47etchants 179, 180(F)gaseous ferritic nitrocarburizing 223salt bath nitriding 62–63vapor degreasing 179–180

Salt bath ferritic nitrocarburizingcorrosion resistance 208, 208(F), 209(F), 217dimensional stability 209early history 195–196, 196(F), 197(F)engineering properties 215–217fatigue strength 217, 218(T)furnaces 231, 232(F)low-cyanide 202–207lubricity 216, 217(F)Melonite process 55, 202, 203–207metallurgical results 211–215, 216(F)Nu-Tride process 207, 208–217post treatment 207–208process control 210process parameters 201–202, 209–210QPQ process 205(F), 207–208, 208(F),

209–210, 209(F), 210(F), 211(F)surface cleanliness 241–242surface hardness 206, 206(T)Sursulf process 217Tenoplus process 217tensile strength 206, 206(T)trade names 202(T)

wear resistance 206–207, 207(F), 215–216,217(F)

Salt bath nitridingadvantages 54bath aging 55–58bath maintenance 61–62, 188bath replacement 58bath testing and analysis 58–61early developments 9, 54equipment 55, 56(F), 58, 63furnace types 56(F)procedure 55–63process types 54–55safety precautions 62–63salts 53–54, 58, 60–61, 187stop-off procedures 164surface contaminants 57, 59troubleshooting 187–188

Sample preparationcutting 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, 163–165Sensors 51Sergeson, Robert 7–8Sintered steels 223Society of Manufacturing Engineers (SME)

6, 14Solubility limit, nitrogen in iron 2, 139–142Sputter cleaning 104–105, 105(F), 243Sputtering, and ion nitriding 142–150Stainless steels

corrosion resistance 36, 129, 134(F)hardness 35, 129nitridability 129–137for retorts 43salt 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 131–134type 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; High-speed steels; Maraging steels; Nitralloysteels; 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 31–37selection 125–126

Index / 255




Stop-off procedures 163–165Stress relieving. See DistortionSturges, Derek 10, 72Sulfinuz process 195, 201Surface contaminants

and case formation 36–37in ferritic nitrocarburizing 241–243in gas nitriding 185, 186–187in 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 227diffusion techniques compared 16(F)early results 14high-speed steel cutters 156–157maraging steels 161Melonite process 203–204, 203(F), 204(F),

206, 206(T)pure irons 160stainless steels 161values 20–21

Sursulf process 217

T

Temperaturecontrol 49–50, 154–155factors 69(T), 137(T), 237(T)requirements 14–16

Tenoplus process 217Thermocouples 49–50, 91–92Thin-film deposition 76–77, 157Timken Detroit Axle Co. 6Tool 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 44for salt bath ferritic nitrocarburizing 197(F)

Troubleshooting 185–191Tufftride process 54, 55, 196, 201, 203Two-stage process 34, 66, 67(F). See also Floe

process

U–Z

University of Aachen 72Vapor degreasing 36, 170, 179–180Walsted, J.P. 8Wehnheldt, Dr. 9, 71, 72White 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]

In Europe

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

In Japan Neutrino Inc. 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 - D. Pye (ASM, 2003) WW

Documents

tablecurve

japan takahashi

spiral bevel

ceramic tile

synthetic

worldwide

blend mixture

containers