David Pye - Practical Nitriding and Ferritic Nitrocarburizing (ASM)

FM_Nitriding.
qxd 10/6/03 2:04 PM Page i
© 2003 ASM International. All Rights Reserved. www.asminternational.org

Practical Nitriding and Ferritic Nitrocarburizing (#06950G)
Practical
NITRIDING
and Ferritic Nitrocarburizing
David Pye
ASM International
Materials Park, Ohio 44073-0002
www.asminternational.org
FM_Nitriding.qxd 10/6/03 2:04 PM Page ii

Copyright © 2003
by
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 THIS
PUBLICATION. Although this information is believed to be accurate by ASM, ASM cannot guarantee that favorable
results will be obtained from the use of this publication alone. This publication is intended for use by persons having
technical skill, at their sole discretion and risk. Since the conditions of product or material use are outside of ASM’s
control, ASM assumes no liability or obligation in connection with any use of this information. No claim of any kind,
whether as to products or information in this publication, and whether or not based on negligence, shall be greater in
amount than the purchase price of this product or publication in respect of which damages are claimed. THE REM-
EDY HEREBY PROVIDED SHALL BE THE EXCLUSIVE AND SOLE REMEDY OF BUYER, AND IN NO
EVENT SHALL EITHER PARTY BE LIABLE FOR SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES
WHETHER OR NOT CAUSED BY OR RESULTING FROM THE NEGLIGENCE OF SUCH PARTY. As with any
material, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specific
testing under actual conditions is recommended.
Nothing contained in this book shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in
connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters
patent, copyright, or trademark, and nothing contained in this book shall be construed as a defense against any alleged
infringement of letters patent, copyright, or trademark, or as a defense against liability for such infringement.
Comments, criticisms, and suggestions are invited and should be forwarded to ASM International.
Prepared under the direction of the ASM International Technical Book Committee (2002–2003), Charles A. Parker,
Chair
ASM International staff who worked on this project include Charles Moosbrugger, Acquisitions Editor; Bonnie
Sanders, Manager of Production; Nancy Hrivnak, Jill Kinson, and Carol Polakowski, Production Editors; and Scott
Henry, Assistant Director of Reference Publications.
Library of Congress Cataloging-in-Publication Data
Pye, David, 1939–

Practical nitriding and ferritic nitrocarburizing / David Pye
p. cm.
Includes bibliographical references and index.
1. Nitriding. 2. Case hardening. 3. Steel—Heat treatment. I. Title.
TN752.C3P4 2003
671.3'6—dc21
2003056298
ISBN: 0-87170-791-8
SAN: 204-7586
ASM International®
Materials Park, OH 44073-0002
www.asminternational.org
Printed in the United States of America

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

Contents
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
PART 1 Nitriding
CHAPTER 1
An Introduction to Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Metallurgical Considerations and Process Requirements . . . . . . . . . . . 1
The Pioneering Work of Machlet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Parallel Work in Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Developments in the United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Other Early Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Current Status of Nitriding Technology . . . . . . . . . . . . . . . . . . . . . . . . 11
CHAPTER 2
Why Nitride? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Key Process Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
CHAPTER 3
How Does the Nitriding Process Work? . . . . . . . . . . . . . . . . . . . . . 23
The Liberation of Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Dissociation of the Gas at the Selected Nitriding Temperature . . . . . 25
Why Ammonia Is Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Preheat Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
CHAPTER 4
Microstructures of Nitrided Iron and Steel . . . . . . . . . . . . . . . . . . 31
Influence of Carbon on the Compound Zone . . . . . . . . . . . . . . . . . . . 32
Controlling Compound Zone Thickness . . . . . . . . . . . . . . . . . . . . . . . 32
What Happens Below the Compound Zone? . . . . . . . . . . . . . . . . . . . . 35
iii
FM_Nitriding.qxd 10/6/03 2:04 PM Page iv

Can Plain Carbon Steel Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Calculating the Compound Zone Thickness . . . . . . . . . . . . . . . . . . . . 36
Other Factors Affecting Surface Case Formation . . . . . . . . . . . . . . . . 36
CHAPTER 5
Furnace Equipment and Control Systems . . . . . . . . . . . . . . . . . . . 39
Essential Furnace Design Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Types of Nitriding Furnaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Determining Appropriate Furnace Design . . . . . . . . . . . . . . . . . . . . . . 43
Retort Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Retort Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Sealing the Retort to Prevent Ammonia Leaks . . . . . . . . . . . . . . . . . . 44
Safety Precautions When Using Ammonia . . . . . . . . . . . . . . . . . . . . . 46
Furnace Heating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Process Control and Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . 49
CHAPTER 6
Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Salts Used and Process Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Types of Salt Bath Nitriding Processes . . . . . . . . . . . . . . . . . . . . . . . . 54
Salt Bath Nitriding Equipment and Procedure . . . . . . . . . . . . . . . . . . . 55
Using a New Salt Bath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Bath Replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Bath Testing and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Bath Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Operating the Salt Bath . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Safety Precautions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Design Parameters for Furnace Equipment . . . . . . . . . . . . . . . . . . . . . 63
CHAPTER 7
Control of the Compound Zone or White Layer . . . . . . . . . . . . . . . 65
A Test to Determine the Presence of the White Layer . . . . . . . . . . . . . 66
Reduction of the Compound Zone by the Two-Stage Process . . . . . . 66
Other Methods for Controlling Compound Zone Formation . . . . . . . 67
Case Depth of Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
CHAPTER 8
Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
History of Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
How the Ion Nitriding Process Works . . . . . . . . . . . . . . . . . . . . . . . . . 72
Glow Discharge Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Process Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Other Uses for Plasma Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
iv
FM_Nitriding.qxd 10/6/03 2:04 PM Page v

What Happens in the Ion Nitriding Process . . . . . . . . . . . . . . . . . . . . 77

Gas Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Reactions at the Steel Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Surface Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
“Corner Effect” and Nitride Networking . . . . . . . . . . . . . . . . . . . . . . . 80
Degradation of Surface Finish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Control of the Compound Zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Process Gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Process Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Plasma Generation Philosophies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Oxynitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
CHAPTER 9
Ion Nitriding Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Cold-Wall Continuous dc Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . 89

Hot-Wall Pulsed dc Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Work Cooling after Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . 101
Other Considerations for Ion Nitriding Equipment
and Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Summary: Advantages of Plasma Nitriding . . . . . . . . . . . . . . . . . . . . 107
CHAPTER 10
Nitriding in Fluidized Beds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Heating Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Nitriding in the Fluidized-Bed Furnace . . . . . . . . . . . . . . . . . . . . . . . 114
Oxynitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Operating the Fluid Bed for Nitriding . . . . . . . . . . . . . . . . . . . . . . . . 117
Measurement of the Gas Dissociation . . . . . . . . . . . . . . . . . . . . . . . . 117
CHAPTER 11
Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Size Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Shape Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Control of Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Distortion in Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Stock Removal Prior to Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Postmachining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
CHAPTER 12
Steels For Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Steel Selection Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

Requirements for a Nitriding Steel . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
v
FM_Nitriding.qxd 10/6/03 2:04 PM Page vi

Can Stainless Steels Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Plasma Nitride Case Depths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
CHAPTER 13
Control of the Process Gas in Plasma Conditions . . . . . . . . . . . . 139
Analysis by Photo Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Analysis by Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Difficulties Associated with Gas Analysis . . . . . . . . . . . . . . . . . . . . . 141
Kinetic Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Appendix: The Role of Sputtering in Plasma Nitriding . . . . . . . . . . . . . 142
Experimental Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
CHAPTER 14
Processing with Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Hot-Work Tool Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
High-Speed Steel Cutters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Gears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Pure Irons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Low-Alloy Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Maraging Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Higher Alloyed Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
CHAPTER 15
Stop-Off Procedures for Selective Nitriding . . . . . . . . . . . . . . . . 163
Methods for Selective Gas Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . 163
Methods for Selective Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . 164
Methods for Selective Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . 164
CHAPTER 16
Examination of the Nitrided Case . . . . . . . . . . . . . . . . . . . . . . . . 167
Hardness Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Etching of the Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Safety Precautions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Optical Light Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
CHAPTER 17
Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Gas Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Salt Bath Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Ion Nitriding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
vi
FM_Nitriding.qxd 10/6/03 2:04 PM Page vii

PART 2 Ferritic Nitrocarburizing
CHAPTER 18
What Is Meant by Ferritic Nitrocarburizing? . . . . . . . . . . . . . . . . 193
Process Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Early History of Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . 195
Why Ferritic Nitrocarburize? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
CHAPTER 19
Salt Bath Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . 201
Low-Cyanide Salt Bath Ferritic Nitrocarburizing . . . . . . . . . . . . . . . 202
Salt Bath Nitrocarburizing plus Post Treatment . . . . . . . . . . . . . . . . 207
Kolene Nu-Tride Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Other Methods for Salt Bath Nitrocarburizing . . . . . . . . . . . . . . . . . . 217
CHAPTER 20
Gaseous Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . 219
Development of the Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Process Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
Gaseous Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Properties of Gaseous Ferritic Nitrocarburized Components . . . . . . 221
Industrial Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
Safety Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Appendix: Gaseous Nitrocarburizing—A Suitable Alternative
for the Heat Treatment of Automotive Crankshafts . . . . . . . . . . . . 223
Process Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
Typical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
CHAPTER 21
Equipment for Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . 231
Salt Bath Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Atmosphere Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
Plasma-Assisted Furnace Equipment . . . . . . . . . . . . . . . . . . . . . . . . . 233
Ferritic Oxynitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
CHAPTER 22
Preparation for Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . 241
Gas Ferritic Nitrocarburizing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Enhanced Plasma Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
vii
FM_Nitriding.qxd 10/6/03 2:04 PM Page viii

CHAPTER 23
Evaluating the Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Case Depth Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Case Hardness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
What If the Formed Case Has Low Hardness Values? . . . . . . . . . . . 246
Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
viii
FM_Nitriding.qxd 10/6/03 2:04 PM Page ix

Preface
Usually one seeks out a career, and, although I chose my career, I never
realized how that choice was to impact my life. Although I chose heat
treatment as a career, I did not make a conscious choice for nitriding; the
subject of nitriding chose me.
In 1960, I was a final-year apprentice at DeHavilland Propellors,
Lostock, United Kingdom. My project with a colleague was to evaluate
the nitriding process for the DeHavilland Aircraft Group. It was at that
time that the subject of nitriding chose me.
No matter where I have been, in the United Kingdom, South Africa,
and now the United States, the subject of nitriding has followed me. Yet
each time that I have researched the subject, I have found very few
resource materials available. Unlike carburizing, for example, the subject
of nitriding has had very few reference books or “cook books” written on
the subject. A few books include a chapter or two on nitriding, and some
conference papers are available in proceedings volumes. Up to now, how-
ever, there has not been a practical “how to” or “why to” book available
on the subject.
In 1991, Rodney Allwood of the ASM Education Department urged me
to present a one-day class on nitriding and somehow got me to agree.
Despite feeling that I did not know enough to pull it off, to my surprise I
was able to put together the notes for that course. This was the foundation
for a book on nitriding. Mrs. Veronica Flint of the ASM Reference Publi-
cations Department challenged me to write such a book and, with her
tremendous patience and persistence, forced me to find the time to put pen
to paper.
Many books are dedicated to husbands, wives, children, or even dogs
and cats (we have 8!). I can only dedicate this book to the young heat treater
or metallurgist who is coming into the industry and to my colleagues who
have, without exception, given me tremendous encouragement.
I want to remember my colleagues in South Africa, where to a large
extent I learned my trade. Without that 20 years experience in South
Africa, I could not do what I am doing today. I also want to remember my
ix
FM_Nitriding.qxd 10/6/03 2:04 PM Page x

colleagues at DeHavilland Propellors with whom I worked from 1956 to

1963, especially Wally Simms, our foreman (now deceased), who had the
foresight and tenacity to fight to bring apprentices into the world of heat
treatment and the DeHavilland Propellers (now British Aerospace)
apprentice program; Bill Oddey, our heat treatment superintendent; Joe
Aspinall, our metallurgist; and Alex Thexton, our chemist (who sold me
his briefcase, which I still have), who is now somewhere in Australia. We
will always be grateful to Paul Huber at Seco Warwick, who brought my
family and me to the United States, and I thank my other colleagues there.
I would also like to dedicate this book to all those who have gone before
me whose work in the field of nitriding has brought the process to maturity.
I stand humbly before all of them and thank them for this opportunity.
For help and growth in the field of pulsed plasma nitriding, I give
thanks to Dr. Siegfeid Stramke in Germany. I especially would like to 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 great
respect in the field of plasma physics.
I would like to acknowledge both George Totten and Maurice Howes for
being mentors to me and for their very positive encouragement. Their 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 I
was capable of “going on my own.”
It took my wife Lynn to push me to the successful completion of this
and other goals. She said to me, “You have a dream. Don’t keep dreaming;
live the dream.” I could not have accomplished what I have without her
patience, dedication, and support. This book could not have come together
without her. She typed the original manuscript from my almost illegible
handwriting and “chicken scratching.” Thank you for your support of me
and for listening to my heat treatment and furnace stories. You really are a
gift to me.
I also would like to recognize Valerie Sales, my mother-in-law
(deceased) who interpreted my hand drawn sketches and turned them into
illustrations, and Robin Maloney, who also helped with the illustrations.
Thank you for being patient with me.
I would like to thank all the reviewers who gave freely of their time and
efforts to reviewing the manuscript. I would also like to acknowledge the
editorial contribution of Joseph R. Davis, Davis & Associates, who helped
polish the chapters and prepare them for the production process.
Purpose
Nitriding and ferritic nitrocarburizing in some ways can be seen as
Cinderella processes in comparison to the other surface modification
techniques, as it was around but often ignored. Though the begining of
nitriding can be traced to Adolph Machlet who applied for a patent in
x
FM_Nitriding.qxd 10/6/03 2:04 PM Page xi

1908 followed by Dr. Adolph Fry’s patent in 1922, it was considered in

many areas as “a new process” when Drs. Wehnheldt and Berghaus
developed the ion nitride process in 1932. Many fine metallurgists have
contributed to the growth in our knowledge of the subject of nitriding. I
have attempted to bring together all of their work by capturing the accu-
mulated knowledge about the process in a book that will contribute to the
continued growth of the nitriding process. The bell has not struck mid-
night on the subject of nitriding, however. Its technology, the refinement
of 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 for
application to nonferrous metals, such as aluminum and the refractory
metals.
This book is intended to assist all of the practitioners of the technology
in the day-to-day process operation of nitriding and ferritic nitrocarburiz-
ing. The contents are based upon my lifetime of experience and the
knowledge gained from my peers. I hope that you, the reader, will gain
some useful knowledge about the subject of nitriding and its derivative
process techniques.
The chapters in this book address many important questions related to
the nitriding process:
• Of the many nitriding methods, which one is for you? There are many
different and valid reasons for choosing each relevant nitriding
process technique, be it the reduction in thickness of the compound
zone, the elimination of the compound zone, deep case formation,
shallow case formation, high wear resistance, or corrosion resistance.
• Of the many different nitridable steels that can be chosen to 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 not
removed?
• Which furnace should be used?
• How should the process be controlled?
• How should the steel be prepared?
• How should the steel be handled after nitriding?
Up to now, many of the answers have been determined mostly by per-

sonal experience (“what works for you”) and possibly by the age old
method of trial and error.
This book is a practical approach to the subject and not an academic or
scientific work, although the subject is of a scientific nature. As with any
other surface treatment process, such as carburizing or carbonitriding, the
process of nitriding draws on many disciplines such as physics, chemistry,
mechanical engineering, and electrical engineering. The “art” of nitriding
xi
FM_Nitriding.qxd 10/6/03 2:04 PM Page xii

also requires individual ingenuity and dogged determination, tempered

with patience to accomplish the process and produce a metallurgically
sound part. The process cannot be guessed. It can offer many metallurgi-
cal benefits, but it needs to be managed and controlled in order to produce
the desired and acceptable surface metallurgy. While it is a very simple
process, there is little widespread understanding of it. This book is my
attempt to help remedy this situation.
Over the years of heat treating and the years in the furnace industry,
many examples of material selection for varied applications, process tech-
niques, and failure evaluations have been both seen and experienced by
myself and shared by others. My years as a consultant have exposed me to
many industrial problems and process techniques. All of these experiences
contributed significantly to this publication.
There are many, many steels available to the engineer who designs the
part and chooses the steel. The problem that the engineer is faced with is
“How do I source my information on steel and metallurgical processing
techniques?” The simple answer (which may sound glib) is “With great
difficulty.” It is hoped that this publication will provide the reader, who
might be a heat treater, metallurgist, or design engineer, with a clearer
insight into the techniques, material selection, equipment, control, testing,
evaluation, and trouble shooting.
The analogy of an iceberg can be used to consider the factors that influ-
ence process costs related to nitriding. Nine-tenths of an iceberg is below
the surface of the water. The top one-tenth could be likened to represent-
ing both the material cost and the process cost. However, inappropriate
material selection and a lack of understanding of process techniques (and
their results) can greatly inflate these costs. The submerged nine-tenths of
this iceberg is the labor cost, machining costs, equipment costs, and time
that is lost if the part does not function properly and must be scrapped
because of improper heat treatment procedures and material selection.
The combination of steel selection and choice of heat treatment can
therefore “make or break the product.” This book is offered to help you
“make the product” and not “break the product.”
David Pye
Pye Metallurgical Consulting Inc.
Meadville, PA
xii
01_Nitriding.qxd 9/30/03 9:58 AM Page 1
Practical Nitriding and Ferritic Nitrocarburizing Copyright © 2003 ASM International ®

David Pye, p1-12 All rights reserved.
DOI: 10.1361/pnafn2003p001 www.asminternational.org
CHAPTER 1
An Introduction to
Nitriding
THE NITRIDING PROCESS, first developed in the early 1900s, con-

tinues to play an important role in many industrial applications. Along
with the derivative nitrocarburizing process, nitriding often is used in the
manufacture of aircraft, bearings, automotive components, textile 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 phase
change from ferrite to austenite, nor does it require a further change from
austenite to martensite. In other words, the steel remains in the ferrite phase
(or cementite, depending on alloy composition) during the complete proce-
dure. This means that the molecular structure of the ferrite (body-centered
cubic, or bcc, lattice) does not change its configuration or grow into the
face-centered cubic (fcc) lattice characteristic of austenite, as occurs in
more conventional methods such as carburizing. Furthermore, because
only free cooling takes place, rather than rapid cooling or quenching, no
subsequent transformation from austenite to martensite occurs. Again,
there is no molecular size change and, more importantly, no dimensional
change, only slight growth due to the volumetric change of the steel 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 the
process, causing movement in the form of twisting and bending.
Metallurgical Considerations and

Process Requirements
Nitriding is a ferritic thermochemical method of diffusing nascent
nitrogen into the surface of steels and cast irons. This diffusion process is
based on the solubility of nitrogen in iron, as shown in the iron-nitrogen
equilibrium diagram (Fig. 1).
2 / Practical Nitriding and Ferritic Nitrocarburizing
Fig. 1 Iron-nitrogen equilibrium diagram. The δ-phase, not shown on this dia-
gram, exists from 11.0 to 11.35% N at temperatures below approxi-
mately 500 °C (930 °F).
The solubility limit of nitrogen in iron is temperature dependent, and at

450 °C (840 °F) the iron-base alloy will absorb up to 5.7 to 6.1% of N.
Beyond this, the surface phase formation on alloy steels tends to be pre-
dominantly epsilon (ε) phase. This is strongly influenced by the carbon
content of the steel; the greater the carbon content, the more potential for
the ε phase to form. As the temperature is further increased to the gamma
prime (γ′) phase temperature at 490 °C (914 °F), the “window” or limit of
solubility begins to decrease at a temperature of approximately 680 °C
(1256 °F). The equilibrium diagram shows that control of the nitrogen dif-
fusion is critical to process success (Fig. 1).
A number of operating process parameters must be adhered to and
controlled in order to successfully carry out the nitriding process. Most
of these parameters can be controlled with relatively simple instrumenta-
tion and methods. Examples of process parameters for gas nitriding
include:
• Furnace temperature
• Process control (see discussion below)
• Time
• Gas flow
• Gas activity control
• Process chamber maintenance
All these factors help to reduce distortion during the process, with the
exception of induced residual stresses. Another benefit of nitriding is that
it acts as a stabilizing process by providing an additional temper to the
processed steel.
Chapter 1: An Introduction to Nitriding / 3
Control of the process parameters is necessary to ensure formation of

an acceptable metallurgical case. Without control, repeatability of the
metallurgical requirements cannot be guaranteed.
The process control factors are those elements that will ensure a 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. He
recognized 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 was
very soluble in iron. Nitrogen diffusion produced a relatively hard surface
in 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 the
process chamber, while still under the protection of the nitrogen-based
atmosphere, thus reducing the risk of distortion yet still producing a hard,
wear-resistant surface with good corrosion resistance.
Ammonia was decomposed, or “cracked,” by heat to liberate the nas-
cent nitrogen necessary for the process. It was not long before Machlet
realized that he needed to control the decomposition accurately. He did
this by using hydrogen as a dilutant gas to reduce the amount of available
nascent nitrogen, thus controlling to some extent the formed case metal-
lurgy. His reasoning behind the control of the process gas was recognition
of what is now known as the “white layer” or “compound zone.” Figure 2
shows a simple construction of the nitrided case. It should be noted that
this schematic is not to scale.
The first patent for the development of the nitriding process was applied
for in March 1908 in Elizabeth, NJ. The patent was finally approved in
June 1913, some five years after the initial application. Machlet had been
working for a number of years on the process prior to his patent application
Compound zone,
dual phase
Diffusion zone consisting

of formed nitrides
Transition zone from

diffusion zone to core
material
Core material
Fig. 2 Schematic of a typical nitrided case structure
and continued to develop both the new process and his understanding of
the resulting process metallurgy. The patent was for “The Nitrogenization
of Iron and Steel in an Ammonia Gas Atmosphere into which an Excess of
Hydrogen Has Been Introduced” (Ref 1).
Although Machlet’s development and patenting of the new nitriding
procedure was technologically important, his work remained largely
unrecognized and faded into obscurity. Even today, very few nitriding
practitioners know who he was and what he accomplished. Most metallur-
gists who are familiar with the nitriding process know the work of the
German researcher Adolph Fry, who is recognized as the “father of nitrid-
ing.” While Fry’s work was more publicized and his methods were taught
at many fine metallurgical academic institutions, it was Machlet who first
pioneered the nitriding process.
Parallel Work in Europe

Adolph Fry. In Germany, a parallel research program was under way
at the Krupp Steel Works in Essen. This program was headed by Dr.
Adolph Fry in 1906. Like Machlet, Fry recognized that nitrogen was
very soluble in iron at an elevated temperature. He also recognized very
early in his work that alloying elements strongly influenced metallurgi-
cal and performance results. Fry first applied for his patent in 1921, three
years after the First World War ended. His patent was granted in March
1924 (Ref 2).
He used a technique similar to that of Machlet, where the nitrogen
source had to be cracked by heat to liberate nitrogen for reaction and dif-
fusion. Like Machlet, Fry used ammonia as the source gas, but he did not
use hydrogen as a dilutant gas. Thus was developed the single-stage gas
nitriding process as it is known today.
Fry then investigated the effects of alloying elements on surface hard-
ness. He discovered that the nitriding process produced a high surface
hardness only on steels containing chromium, molybdenum, aluminum,
vanadium, and tungsten, all of which form what are known as “stable
nitrides.”
He also discovered the critical nature of process temperature in terms of
case depth and surface metallurgy. Processing the steel at higher tempera-
tures placed the surface at risk to form what is known today as “nitride
networks” (a saturated solution of nitrogen in the immediate surface of the
formed case).
Because steels with higher alloy contents were not readily available for
nitriding, Fry became responsible for developing a group of steels for
Krupp known as the “Nitralloy” group. These steels, specifically designed
as nitriding steels, soon became internationally recognized. Even today
the Nitralloy steels are specified.
British Standard Nitriding Steels. Shortly thereafter in the late 1920s,
a company in Sheffield, England, also began work on developing a group
of nitriding steels under the licensed guidance of Krupp Steels. These
steels were also marketed under the brand name of Nitralloy. The com-
pany was Thomas Firth and John Brown Steelworks, more commonly
known as Firth Brown Steels. The steels from Firth Brown were known as
the “LK” group, designated by British Standard 970 as En 40 A, En 40 B,
En 40 C, En 41 A, and En 41 B. Developed for nitriding applications,
these were chromium-molybdenum steels (see Table 1 for chemical com-
positions). The En 41 series contained aluminum, which produced a much
higher surface hardness after nitriding. Aluminum has a strong affinity for
nitrogen, forming very hard aluminum nitrides that are quite stable in
amounts up to 1.0% Al. Much above 1.0%, aluminum has no effect on the
resultant nitriding hardness.
Differences Between the U.S. and German Processes. The princi-
pal differences between the process developed in the United States and
that developed in Germany were that:
• The U.S. process used hydrogen as a dilutant gas to control the nitrid-
ing potential of both the gas and steel, which in turn controlled the
final surface metallurgy.
• The Germans manipulated the process through alloying and improved
on such aspects as core hardness and tensile strength.
Table 1 British standard nitriding steels

Composition, %
Designation(a) 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 ... 0.90–1.30
En 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 it

was perceived to have little if any commercial value to U.S. industry. By
contrast, the Germans exploited Fry’s process in the early years following
WWI. The German process enjoyed great success throughout Europe in
the aircraft, textile, railroad, automotive, and machine tool industries.
During the mid to late 1920s, information about Fry’s success began to
filter through to American industrialists, prompting the Society of 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 meeting
that Jeffries suggested to Fry that he attend the forthcoming annual SME
conference in Chicago and present a paper on the process techniques and
applications. Fry could not attend, so his friend and colleague Pierre
Aubert made a presentation on his behalf. The presentation helped bring
about commercialization of the process in the United States.
Developments in the United States

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

• Influence of alloying elements in the newly developed Nitralloy steels
• Influence of temperature on growth and distortion
• Influence of time on case depth distortion and growth
• Effects of the ammonia/hydrogen relationship and dilution by hydrogen

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

fan insulation
Furnace door
Furnace
thermocouple
Exhaust ammonia
gas outlet tube
Nitride process
To atmosphere chamber
exhaust
Process delivery
gas (ammonia) Process chamber
thermocouple tube
Ammonia gas
inlet tube
Load preparation
table
Fig. 3 Schematic of a simple ammonia gas nitriding furnace
the nascent nitrogen diffusion if present in significant quantities. More

detailed information on alloying effects can be found in Chapter 12,
“Steels for Nitriding.”
V.O. Homerberg and J.P. Walsted. Professor Homerberg was an asso-
ciate professor of metallurgy at the Massachusetts Institute of Technology
and consulted for the Ludlum Steel Company. Mr. Walsted was an instruc-
tor at the same university at the time that they presented their findings on
the nitriding process (Ref 5). They studied the effects of temperature up to
750 °C (1400 °F), with its resulting increase in case depth but reduction of
surface hardness. In addition, they studied the effects of decarburization
on a nitrided surface and concluded that surfaces must be free of decarbur-
ization prior to nitriding.
They reviewed Fry’s process technique and the decomposition of
ammonia 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 layer
on the nitrided steel surface. The white layer was identified as a multi-
phase compound layer of ε and γ′ phases. Much recognition was given to

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

out on virtually all steels and cast irons as well as refractory metals, alu-
minum (not yet on a commercial basis), and sintered ferrous materials. A
schematic of an ion nitriding furnace layout is shown in Fig. 4.
Ion Nitriding Production in the U.S. After WWII and through the late
1950s, the General Electric Company of Lynn, Massachusetts, operated a
laboratory known as the Electromechanical Engineering and Physics
Unit. Dr. Claude Jones and Dr. Derek Sturges, along with Stuart Martin,
developed the first ion nitriding unit in the United States and applied the
process to a variety of materials and parts. Their ion nitriding units met
all normal nitriding standards and were accepted by the U.S. Navy. Sum-
maries of the properties, applications, and advantages associated with the
process were published in 1964 (Ref 7) and 1973 (Ref 8).
Other Uses of Plasma Technology. Ion nitriding is not the only heat
treatment process that utilizes the glow discharge phenomenon. Simply
put, if one uses the appropriate process gases and the proper furnace, the
glow discharge technique can be applied to plasma-assisted ferritic 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 of
ASM Handbook.
Plasma technology is not new. One has only to observe the Northern
Lights to witness a natural plasma. Lightning is also a natural plasma.
Vacuum process
vessel –
Work piece
Power
source
Process gas
manifold
(nitrogen,
argon) +
Vacuum
pump
Fig. 4 Simple schematic of the layout of an early plasma (ion) nitriding fur-
nace system
Neon signs are a plasma glow, and “lumina storms” found in gift shops are
just a few of the many examples of plasma technology at work.
Current Status of Nitriding Technology

The success of any heat treatment is measured by hardness. However,
hardness is relevant to the materials application and its mechanical
requirements. Nitriding often is applied to low-alloy steels to “harden” the
steel and improve corrosion resistance. In addition to conventional nitrid-
ing, the following processes have been developed:
• Oxynitride process, during which a controlled postoxidation treatment

is carried out to further enhance the surface corrosion resistance
• Ferritic nitrocarburizing (a controlled process using nitrogen and 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 an
almost parallel course since the early part of the 20th century. The process
has found its place in both low- and high-tech applications and is becom-
ing better understood by process technicians, metallurgists, applications
engineers, furnace designers, and academics.
Many developments in process techniques are being driven by environ-
mental concerns and legislation. This has resulted in the introduction of
more effective, efficient, and economical methods and equipment.
Improvements can be seen in the development of gaseous methods, salt
bath methods, fluidized-bed methods, and plasma processing techniques.
REFERENCES
1. A. Machlet, U.S. Patent 1,092,925, 24 June 1913
2. A. Fry, U.S. Patent 1,487,554, 18 March 1924
3. H.W. McQuaid and W.J. Ketcham, Some Practical Aspects of the
Nitriding Process, reprinted from Trans. ASST, Vol 14, 1928, Source
Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American
Society for Metals, 1977, p 1–25
4. R. Sergeson, Investigation in Nitriding, reprinted from American
Society for Steel Treaters (ASST) Nitriding Symposium, 1929, in
Source Book on Nitriding, P.M. Unterweiser and A.G. Gray, Ed.,
American Society for Metals, 1977, p 26–55
5. V.O. Homerberg and J.P. Walsted, A Study of the Nitriding Process—
Part I, reprinted from American Society for Steel Treaters (ASST)
Nitriding Symposium, 1929, in Source Book on Nitriding, P.M. 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, Vol
32, 1944, Source Book on Nitriding, P.M. Unterweiser and A.G. Gray,
Ed., American Society for Metals, 1977, p 144–171
7. C.K. Jones and S.W. Martin, Nitriding, Sintering and Brazing in Glow
Discharge, Met. Prog., Feb 1964, p 94–98
8. C.K. Jones, D.J. Sturges, and S.W. Martin, Glow-Discharge Nitriding
in Production, reprinted from Met. Prog., Dec 1973, Source Book on
Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for
Metals, 1977, p 186–187

CHAPTER 2
Why Nitride?
THE UNIQUE ADVANTAGES of the nitriding process were recognized

by the Germans in the early 1920s. It was used in applications that required:
• High torque
• High wear resistance
• Abrasive wear resistance
• Corrosion resistance
• High surface compressive strength
Early on, the process did not gain much recognition in the United States
because of its moderate hardness values for plain carbon, cast iron, and
low-alloy steels. The very long nitriding process cycle times required to
reach the same case depths achieved by more conventional methods such
as carburizing were considered a disadvantage. For example, to achieve a
case hardening depth of 1.0 mm (0.040 in.), the nitriding process requires
90 h, compared to 4.5 h for carburizing (Fig. 1, 2).
As described in Chapter 1, the patent for gas nitriding was first applied
for by Adolph Machlet and was for the nitrogenization of iron and steel in
Total case depth

required = 1.0 mm (0.040 in.)
4.5 h
Temperature,°C
1700
Temperature,°F
927
1475 800
Quench
Time
Fig. 1 Example of carburizing, followed by quenching, to produce a total

case depth of 1.0 mm (0.040 in.)
Total case depth

required = 1.0 mm (0.040 in.)
Temperature,°C
Temperature,°F
90 h
925 495
Free cool
Time
Fig. 2 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
an ammonia gas atmosphere diluted by hydrogen. The hardness results that

he achieved were not high by the standards of today or even by the German
standards of the day. Hardness results generally measure the success of the
process and are expected to be in the region of approximately 60 to 64 HRC.
Early surface hardness values obtained by Machlet were in the region of 30
to 35 HRC, considered too low in terms of wear properties. Keep in mind,
however, that hardness is relative to the wear characteristics of the steel part
being treated. What was not recognized was the excellent corrosion 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 synonymous
with the nitriding process (and which exhibited considerably higher hard-
ness values than those obtained by Machlet). German industrialists began
to control the steel analysis for nitridable steels, licensing it to various
steelmakers in other countries. The attitude toward the use of the process
in America began to change after the 1927 presentation of Fry’s paper at
the SME conference in Chicago.
Key Process Considerations

Several factors helped nitriding to gain acceptance:
• Compared to other case hardening methods, nitriding is a relatively

low-temperature process.
• Nitriding is relatively easy to control in terms of process parameters.
• It produces enhanced corrosion resistance in low-alloy and low-carbon
steels.
• Core hardness is not significantly affected, due to prehardening and
tempering.
• No quenching is required, thus reducing distortion.
Low-Temperature Process. The nitriding process requires a relatively

low temperature compared to the more widely recognized surface treatment
methods. The temperature employed is in the region of 500 °C (925 °F),
Chapter 2: Why Nitride? / 15
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 above
the tempering temperature, then the core hardness of the pretreated steel
will diminish according to time and temperature. It is most important to
maintain the core hardness of the pretreated steel to provide adequate sup-
port to the diffused case. Without good core support, the case can fail.
Surface treatment processes that require higher temperatures than the
nitriding process requires include:
• Carburizing, which employs a temperature in the region of 970 °C

(1775 °F) (Fig. 3a)
• Carbonitriding, which employs a temperature in the region of 870 °C
(1600 °F) (Fig. 3b)
Figure 4 compares the temperature ranges used by various diffusion sur-

face hardening techniques, along with case depth characteristics. Case
depths accomplished by carburizing are usually considerably deeper than
those accomplished with nitriding. Some carburized case depths can be up
2.5 h
1700
925
Temperature,°C
Temperature,°F
1475 800
Quench
(a) Time
2h
1600
870
Temperature,°C
Temperature,°F
1475
800
(b) Time
Fig. 3 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)
Thermochemical diffusion techniques
Carburize Cabontride Ferritic nitrocarburize Boronize Nitride
Pack Gas Salt Ion Gas Salt Ion Gas Salt Ion Pack Gas Pack Gas Salt Ion
Diffuses carbon Diffuses carbon Diffuses carbon Diffuses boron into Diffuses nitrogen
into the steel and nitrogen into nitrogen, sulfur, the steel surface into the steel
surface the steel surface oxygen Process surface
Process Process (individually temperatures Process
temperatures temperature or combined) into 1400-2000°F temperatures
1600-1950°F 1550-1650°F the steel surface (760-1095°C) 600-1020°F
(870-1065°C) (845-900°C) Process Case depth: (315-550°C)
Case depth: Case depth: temperatures shallow Case depth:
medium shallow 1050-1300°F shallow
(565-705°C)
Case depth:
shallow
Fig. 4 Comparison of various diffusion surface hardening techniques. Source:

Ref 1
to approximately 6.35 mm (0.250 in.) (requiring an extremely long cycle

at the process temperature). The downside of the deep case, of course, is
the occurrence of grain growth, which is inevitable due to the extended
cycle time at elevated temperatures for the carburizing procedure. 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 core
is 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 the
steel 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 of

nitrogen in iron will occur, which can cause the problem known as nitride
networking. This condition will embrittle sharp corners, leading to spalling
or even exfoliation.
No Quench Requirement. With conventional surface diffusion process
techniques such as carburizing and carbonitriding, the steel must be at a
suitable austenitizing temperature (depending on the steel composition)
after carburizing, followed by quenching (or rapid cooling) to transform the
austenite phase (face-centered cubic, or fcc, lattice) into martensite (body-
centered tetragonal, or bct, lattice) (Fig. 6). One should also be cognizant of
1800 3270
1700 3090
1600 2910
1538°C 1495°C Solubility of
1500 Liquid graphite in 2730
(δ-Fe) liquid iron
1400 2550
1394°C
1300 2370
4.26% 1227°C
1200 (γ-Fe) 2.08% 1154°C 2190
Austenite
1100 2.11% 1148°C 4.30% 6.69% 2010
Temperature,°C
Temperature,°F
1000 Austenite Cementite 1830
0.68% (Fe3C)
912°C +
900 cementite 1650
A3
Acm
800 1470
770°C 738°C
700 0.77% 1290
A1(727°C)
600 1110
500 930
(α-Fe)
Ferrite 400 Ferrite 750
+
300 cementite 570
200 390
100 210
0 30
Fe 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
Carbon, wt%
Fig. 5 Iron-carbon equilibrium diagram. The nitriding process is carried out at

temperatures below the A1 line.
Face centered
cubic structure
Additional carbon
dissolves into
structure
Rapid
cooling
Heating to high
temperature
Iron atoms
Carbon atoms
Room temperature
body centered cubic structure Room temperature
body centered tetragonal structure
Fig. 6 Crystal lattice changes that take place during high-temperature heat
treatment processes such as carburizing. Ferrite is bcc structure;
austenite, fcc; martensite, bct. Source: Ref 2
the size of the lattice structure between the two phases of austenite and
martensite. Remember that austenite is an enforced condition in an alloy
carburizing steel, as is martensite. Martensite cannot be created without the
transformation from austenite by rapid cooling. Also, in order to form
martensite it is necessary to have carbon present in sufficient amounts when
the steel is rapidly cooled (Fig. 7). A rapid cooling rate is unnecessary for
nitrided steel after completion of the process cycle. The process chamber
free cools under the ammonia atmosphere down to a suitable temperature of
approximately 200 °C (400 °F) and then is purged with clean, dry nitrogen.
Unlike the ferritic nitrocarburizing, carbonitriding, and carburizing proce-
dures, the nitriding process does not involve a critical cooling rate.
Minimal Distortion. Distortion is one of the most persistent problems
facing heat treaters and engineers. Distortion manifests itself during the
final heat treatment process in the form of either:
• Shape distortion: a change in geometrical form, such as curving,

twisting, or bending
• Size distortion: a change in workpiece volume due to either growth or
shrinkage
Austenite
(fcc)
14 atoms
Temperature
Quench
Martensite Ferrite (bcc) Martensite (bct)

start 9 atoms 9 atoms
Time
Fig. 7 Schematic illustration of the phase transformation taking place when

hardening steel with sufficient carbon present. Crystal lattice: bcc,
body-centered cubic; fcc, face-centered cubic; bct, body-centered tetragonal
Austenite
and
Temperature
phase
eve)
change
(stres up
Quench
s reli
He at
(thermal shock)
Martensite Phase change

start line (stress induced)
Time
Fig. 8 Phase changes in relation to thermally induced stress on quenching

Distortion can be minimized but not eliminated. Stresses are induced into
the steel through rolling and forging as well as machining procedures. These
induced stresses will be manifested at the final heat treatment (Fig. 8).
As steel is heated, the heat-up cycle becomes a stress relieving process,
and the stresses begin to be relieved from the steel. If distortion is to be
kept to a minimum, then it is necessary to introduce an interim stress-
relieving procedure after the preheat treatment and rough machining and
before the final machining and nitriding process. The stress relief process
will act as a stabilizing process on the final nitriding process. Another
method of stabilizing is to cryogenically treat the steel (particularly an
alloy or tool steel). This will transform any retained austenite into untem-
pered martensite that will be tempered by the nitriding process, ensuring
better dimensional stability during and after nitriding (Fig. 9). If the heat
treater exercises careful control over the heating rate during the heat-up
cycle, stress relieving can be minimized.
The reverse side of the coin becomes apparent after the appropriate
soaking period has been completed and it becomes necessary to cool the
steel rapidly, as during hardening of either a through-hardened steel or a
carburized or carbonitrided steel. Thermal stress patterns are then induced
into the steel due to geometric section changes, differential cooling rates,
and phase changes in the steel.
Nitride
Temperature
)
ve
C oo h a s e
p
(no ange)
res e a t u
elie
ch
(No phase change)

l do
p
sr
(Additional temper)
H
wn
(Dimensionally stabilized if
( st
retained austenite is present)
Time
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 added

to create changes in surface chemistry, volume changes will occur in the
form of size change and uniform growth. For example, a ring will grow in
terms of outside and inside diameters, the net effect being that the bore
diameter decreases (Fig. 10).
Nitriding is a relatively low-temperature process and produces a shal-
lower case than does carburizing. Most importantly, no quench is
involved. 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 of
thermal gradients due to sectional changes.
If the engineer can live with extended cycle times, the nitriding process
can offer many distinct advantages:
• Reduced distortion • Improved part cleanliness

• Reduced final machining time • Improved dimensional stability
If one considers the cost of scrap due to distortion and time spent
straightening, nitriding is not an expensive process compared to 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 simply
because there is no opportunity for retained austenite to form, as can occur
in carburizing and quenching. Over time, retained austenite will decom-
pose to untempered martensite. In carburizing, retained austenite will
leave mixed phases in the formed case, which can lead to dimensional
instability. The nitriding process does not promote dimensional instability.
It also acts as a stress relieving procedure.
However, growth most certainly takes place during the nitriding process,
due primarily to the diffusion of nitrogen into the steel surface. The amount
of growth is influenced by:
• Time
• Temperature
• Steel chemistry
• Gas flow
• Gas dissociation
• Steel surface condition prior to the nitriding process
• Surface metallurgy (compound layer thickness)
• Total case depth
Control of the growth aspect of nitrided steel is discussed in Chapter 11,

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

temperature, then the core hardness value will not be affected. A process
temperature approximately 55 °C (100 °F) below the final tempering tem-
perature generally is selected.
Concluding Remarks
The nitriding process was first put to commercial use in the automotive
industry on transmission gears, particularly spiral bevel pinions and spiral
bevel gears because of their natural tendency to try to “straighten” at the
heat treatment process temperature. These types of gears were notorious
for distortion on the gear teeth. Current spiral bevel machining design
practice is to cut distortion into the gear. This technique requires an inti-
mate knowledge of the steel combined with extremely good control of the
heat treatment process.
Nitriding has become more widely accepted because of its ability to
serve many applications that previously were not considered possible or
even worthwhile. Engineers and metallurgists use nitriding creatively,
making it a viable and commercially acceptable process. As our under-
standing of the benefits associated with the process evolves, nitriding will
continue to grow in use and popularity.
REFERENCES
1. D. Pye, Diffusion Surface Treatment Techniques: A Review, Ind.
Heat., March 2001, p 39–44
2. K.G. Budinski, Diffusion Processes, Surface Engineering for Wear
Resistance, Prentice Hall, 1988, p 78–119

CHAPTER 3
How Does the
Nitriding Process Work?
SEVERAL PROCESS PARAMETERS must be considered in order to

ensure nitriding success in terms of metallurgy and distortion:
• Nitrogen source
• Heat
• Time
• Steel composition
In gas nitriding the nitrogen source is almost always derived from the
decomposition (or dissociation) of ammonia gas supplied via an external
bulk storage system or individual bottles connected to a manifold (Fig. 1).
Ammonia gas begins to decompose when heat is applied, usually from
an external source within the furnace. At the usual nitriding temperatures
of 500 to 570 °C (930 to 1060 °F), ammonia is in an unstable thermo-
dynamic state and decomposes in the following manner:
2NH3 ↔ 2N + 3H2 (Eq 1)
A
B C
Fig. 1 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.
Typically, three reactions take place at the steel surface when the steel is
at 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 unstable

and will unite with other like atoms to form molecules as shown in Eq 3
and 4. It is while they are in the atomic state that diffusion takes place.
The released nitrogen diffuses into the steel at the nitriding processing
temperature, but very slowly, to the point where it is not economically
practical or effective. The temperature of 500 °C (930 °F) is considered to
be an “economical” temperature. Nascent nitrogen has an affinity for steel
and iron and will readily diffuse into both materials at elevated tempera-
tures. The higher the temperature, the faster and deeper the nitrogen diffu-
sion. An economical temperature is one that produces an optimum case
depth while not adversely affecting the core properties of the treated steel.
The Liberation of Nitrogen (Ref 1)

Nitrogen decomposes or dissociates in accordance with Eq 1. At the
instant of decomposition, the liberated nitrogen will exist as nascent or
atomic nitrogen and as such can be absorbed by the steel.
Nitrogen has an atomic diameter of 0.142 nm and is dissolved in iron in
interstitial positions in octahedral voids of the cubic lattice that have a
maximum diameter of 0.038 nm in bcc alpha (α) iron and a maximum
diameter of 0.104 nm in fcc gamma (γ) iron. Nitriding of pure iron at 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 room

temperature and 0.115% N at 590 °C (1095 °F)
• Face-centered cubic γ′ nitride, Fe4N, which dissolves 5.7 to 6.1% N
• Hexagonal (epsilon) (ε) nitride, Fe2-3N, which exists in the range of 8
to 11 wt% N
Orthorhombic zeta (ζ), Fe2N, forms at temperatures below 500 °C (930 °F)
and nitrogen contents exceeding 11 wt%—conditions that are not used in
nitriding practice.
The process of the smaller nitrogen atoms passing between the iron-
base crystals as heat is applied up to a suitable process temperature is
known as “interstitial diffusion.” This process is shown schematically in
Fig. 2.
Chapter 3: How Does the Nitriding Process Work? / 25
Fig. 2 Schematic of interstitial diffusion during the nitriding process. The work
is heated to the nitriding temperature with ammonia flowing into the
retort. The ammonia gas dissociates to nitrogen and hydrogen at the part surface.
The nitrogen diffuses into the work in atomic form, and the hydrogen becomes a
part of the furnace atmosphere.
Dissociation of the Gas at the

Selected Nitriding Temperature (Ref 1)
At temperatures around 500 °C (930 °F), the stability of ammonia is
questionable, leading to dissociation rates higher than 98% and thus to
formation of a protective gas without any nitriding effect. Despite the ben-
eficial catalytic effect of the workpiece surfaces and the furnace wall, the
dissociation of ammonia is an extremely slow process. Therefore, ammo-
nia-based nitriding atmospheres for steel treatment rarely contain less than
20% ammonia and frequently up to 50%, and thus their degree of dissoci-
ation is far from equilibrium. The remaining ammonia content is decisive
for the effect of nitriding, during which nitrogen diffuses into the steel
according to the reaction (Ref 2):
NH3 → N(α) + (3/2)H2 (Eq 5)
which occurs at the boundary layer. The nitrogen activity, which is the
driving force in the mass transfer, can be calculated according to (Ref 3):
pNH3
aN = KN (Eq 6)
p3/2
H
where KN is the equilibrium constant of nitrogen, and pNH3 and p3/2H are the
partial pressures of ammonia and hydrogen, respectively.
The nitrogen transfer is comparatively low, and the release of hydrogen
from the ammonia molecules determines the process rate. Therefore, as
mentioned earlier, nitriding times are rather long, up to 120 h.
The nitriding effect of a nitriding atmosphere is defined by its degree

of dissociation, with high degrees of dissociation always indicating a
near-equilibrium state where the nitriding effect is low. In processing, it
is common to use a constant degree of dissociation (e.g., 30%), but 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 actual
degree of dissociation. In ammonia dissociation according to Eq 1, two
molecules of ammonia dissociate to one molecule of nitrogen and three
molecules of hydrogen. This increase in volume dilutes the ammonia
content, as do additional gases, and must be taken into account in deter-
mining the actual degree of dissociation.
The nitriding effect can be determined more easily by means of the
nitriding potential:
pNH3
Np = (Eq 7)
p3/2
H2
The nitriding potential is generally used for describing the nitridability of

an ammonia atmosphere. Its control allows the formation of predictable
case structures and depths for a range of steels, including some tool and
stainless grades (Ref 4). Higher Np values produce higher surface concen-
trations of nitrogen and steeper concentration gradients. Lower potentials
allow the development of nitrided cases with no brittle-compound (white)
layer in high-alloy steels.
Why Ammonia Is Used

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

bcc lattice structure remains unchanged. Growth, however, will be experi-
enced due to an increase in the volume of the immediate steel surface and
by the nitrogen diffusion (Fig. 3).
What this suggests is that the observable growth will be primarily to
surface volume rather than to the core, as no phase changes occur in either
the core or the immediate surface. Nascent nitrogen is being diffused at
the surface and dispersed into the immediate surface, increasing the sur-
face volume.
The atomic nitrogen and hydrogen as described in Eq 1 is considered
unstable and will begin to unite with like atoms to form a molecule as
shown in Eq 2 and 3. During the atomic state is when diffusion into the
steel begins to take place.
Because gaseous molecular nitrogen will not dissociate into nascent
nitrogen when cracked by heat at process temperature, no diffusion of
molecular nitrogen will occur. The steel surface, in conjunction with heat,
will act as a catalyst for the decomposition of ammonia into nascent nitro-
gen and for diffusion into the steel surface.
Provided that all the appropriate conditions are met, nascent nitrogen
will begin to diffuse into the steel surface at a rate determined by the
process temperature. In other words, the process temperature can be as
low as 350 °C (600 °F) and will react with the alloying elements to begin
to form the solid solution of stable nitrides. Conversely, the temperature
can be as high as 600 °C (1050 °F). However, although the reaction of the
nascent nitrogen will still occur, it will also strongly affect the thickness of
the surface compound layer and give rise to the risk of nitride networking.
Distortion
Do not be misled into thinking that the nitriding process causes 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
Fig. 3 Illustration of the growth in volume due to nitriding. The amount of growth
will be determined by the time at the selected process temperature.
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 result
from:
• 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 treated
and is left in, for example, the annealed condition, nitriding will be very
difficult to accomplish. Even if it were possible to nitride an annealed
steel, the workpiece would not perform well within its operating environ-
ment simply because a load on the case could exceed the core strength of
the steel. This is one of the reasons why it is necessary to preheat treat the
steel (harden and temper to produce tempered martensite).
Annealed steel is in the ferritic condition; that is, the atomic structure of
the ferrite phase is in the bcc lattice structure. On heating the steel for the
hardening operation, the phase will change to austenite, which is a 14-atom
fcc lattice configuration (see Fig. 6 and 7 in Chapter 2). On quenching for
the hardening operation, the lattice structure changes to that of martensite,
which has reverted back to a 9-atom structure, but now in the bct 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 to
reduce the core hardness back to the appropriate hardness value that will
best 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 the
nitriding operation. Remember, the nitriding process temperature is
slightly lower than the primary tempering temperature of the steel. The
tempering procedure will decompose the retained austenite into a tem-
pered martensite condition. In addition, the nitriding process temperature
will 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 hardness
will also be determined by the steel chemistry, the austenitizing tempera-
ture, and the final tempering temperature.
REFERENCES
1. J. Grosch, Heat Treatment With Gaseous Atmospheres, Steel Heat
Treatment Handbook, G.E. Totten and M.A.W. Howes, Ed., Marcel
Dekker, 1997, p 663–719
2. C.A. Stickels and C.M. Mack, Overview of Carburizing Processes and
Modeling, Carburizing: Processing and Performance, G. Krauss, Ed.,
ASM International, 1989, p 1–9
3. B. Edenhofer and H. Pfau, Self-Adaptive Carbon Profile Regulation in
Carburizing, Heat Treatment and Surface Engineering: New 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 4
Microstructures of
Nitrided Iron and Steel
FORMATION OF THE NITRIDED CASE begins through a series of

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

of formed nitrides
Transition zone from

diffusion zone to core
material
Core material
Fig. 1 Typical nitrided case

Fig. 2 Typical nitrided case structure showing the white layer (top), the diffu-
sion zone, and the core below the diffusion zone. Source: Ref 1
process progresses and continues until the subsequent nucleation of ε at

the steel surface interface (Fig. 3). Note that the nitrogen diffusion is
much slower in the compound layer than in the steel substrate.
Influence of Carbon on the Compound Zone

Carbon manipulates the amount of γ′ and ε formed within the surface
structure. In a typical nitriding steel with a carbon content of approxi-
mately 0.4 wt%, the formation of γ′ to ε will be roughly equal using the
gas nitriding process (Ref 1). The higher the carbon content of the steel,
the greater the ε-phase in the compound layer. The lower the carbon 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 carbon
in 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 amounts
of each phase. If the compound layer thickness is a critical issue, then
the steel must be selected carefully to accomplish the required surface
metallurgy.
Controlling Compound Zone Thickness

The thickness of the compound layer is controllable to a large extent by
manipulating the process technique. As discussed subsequently, it can be
controlled by dilution, the two-stage Floe process, or by ion nitriding.
Chapter 4: Microstructures of Nitrided Iron and Steel / 33
Fig. 3 Schematic showing the nucleation of γ ′- and ε-nitrides on iron. Source:

Ref 2
Whether the compound zone is detrimental to the formed case will

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

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

(nitride networking) at the corner
What Happens Below the Compound Zone?

The compound layer will always be thicker on plain carbon steel than
on alloy steel for the same cycle time and temperature. However, the met-
allurgy will be completely different in terms of both surface compound
layer and diffusion zone. Beneath the compound zone at nitriding temper-
ature, the nitrogen dissolves into the α-iron and also reacts with certain
alloying 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 nitride
formation below the compound zone is known as the diffusion zone
(Fig. 1, 2). The nascent nitrogen will begin to react with those elements
that will readily form stable nitrides. The nitrides exhibit very high 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 be
nitrided (Fig. 5). However, the compound zone will be much thicker than
for alloyed steel with equal carbon content. This is because the nitrides
formed by the alloying elements will contain more nitrogen than those
formed with iron, or iron in the plain carbon steel. In addition, the hard-
ness values will be much lower. Plain carbon steels that have been pre-
treated to a tempered martensite structure typically will exhibit hardness
values of 400 to 700 HV (49 to 60 HRC) under a 200 gf microhardness
load. Alloy steels can exhibit hardness values from 700 to 1000 HV (60 to
69 HRC) under the same load. Surface hardness values for stainless steels,
particularly when surface chemistry reactions are controlled (as in the
plasma nitriding process), can reach 1500 HV (78 HRC).
Fig. 5 The metallographic appearance of AISI 1015 (UNS G10150) steel after
a 2 h vacuum nitrocarburizing treatment in an ammonia/methane mix-
ture with 1% oxygen addition
Only in a plain carbon steel (one without nitride-forming elements) will

a supersaturated solution of α-iron occur. It is quite possible for the iron
nitrides to segregate on slow cooling from the nitriding temperature or via
an aging process at a low-temperature set point.
Another often overlooked feature of the nitriding process (even on a plain
iron or plain carbon steel) is enhancement of the steel surface corrosion
characteristics. By contrast, the corrosion resistance of stainless steels will
be reduced, due to breakdown of the surface chrome oxide barrier to enable
nitrogen diffusion into the steel. Once the chrome oxide barrier is depleted,
the ability of the stainless steel to resist corrosion is reduced.
Calculating the Compound Zone Thickness

Formation and thickness of the compound zone is influenced by:
• Time • Preheat temperature

• Temperature • Pressure
• Gas composition
A simplified Harris formula can be used as a rough guide for calculating

the compound zone thickness:
Compound zone thickness = √t × f
where t is process time at temperature and f is factor by temperature.

This formula assumes a full ammonia atmosphere with no dilution gas, a
dissociation rate of 30%, and a gas turnover rate of 5 to 1 of the nitriding
chamber volume. If the nitrogen is added to the ammonia, thus potentially
increasing the nitriding potential of the ammonia, or if hydrogen is added
to the ammonia, then the nitriding potential is reduced and the gas turnover
rate of 5 to 1 will increase according to the dilution volume increase.
Other Factors Affecting Surface Case Formation

The steel being processed must be completely free of surface contami-
nants. This means that the work surfaces should be “clinically” clean—a
condition commonly achieved by vapor degreasing the surface after hard-
ening and tempering and prior to nitriding. Surface contaminants, which
can 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 not
decompose and leave a carbon residue. The carbon will form a barrier at
its point of deposition on the surface, thus resisting the nitriding effect and
resulting in a nonuniform nitrided case.
Fingerprints are formed by body oils, which are hydrocarbon based.
Once again, when heat initially is applied to the steel, the oil will decom-
pose, leaving behind a carbon residue on the steel surface. The material
handler should wear lint-free cotton gloves, or other gloves that will not
react or leave a deposit on the steel surface.
Paint residue or even marking ink will set up a nitride-resistant barrier
on the steel surface. Once again, absolute surface cleanliness is mandatory.
Decarburization is a surface problem. The presence of a decarburized
layer will inhibit the uniform formation of a good nitrided surface. If
decarburization has occurred from mill rolling and insufficient machine
cleaning, or if the preheat treatment operation was carried out in a 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, or
the 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 rolled
plate, it is imperative that the machining be uniform. Nonuniform machin-
ing will only contribute to the possibility of a partial decarburized layer on
one side and will also induce a residual stress pattern that will relieve
itself during the nitriding procedure.
REFERENCES
1. D. Pye, Nitriding Techniques and Methods, Steel Heat Treatment
Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker,
1997, p 721–764
2. M.A.J. Somers and E.J. Mittemeijer, Härt.-Tech. Mitt., 46: 375, 1991
3. T. Bell, B.J. Birch, V. Korotchenko, and S.P. Evans, “Controlled
Nitriding in Ammonia-Hydrogen Mixtures,” presented as Heat 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, ASM
International, 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 5
Furnace Equipment and
Control Systems
THE NITRIDING FURNACE is simple in design. Other than the mate-

rials used, little has changed in furnace construction. Principal design
changes involve the control of process parameters, with developments
made in the use of computerized systems employing programmable logic
controllers (PLC) for time, temperature, and gas flow control (Fig. 1).
Good gas circulation is necessary to prevent process gas stagnation within
the process chamber (also called the retort).
Fig. 1 Typical PC/PLC screen configuration for nitriding furnace control.

Courtesy of Plateg USA, Meadville, PA
Essential Furnace Design Criteria

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

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

Good gas circulation is necessary to prevent gas stagnation and to
maintain an approximate 5:1 to 6:1 gas exchange ratio. This, of course,
will depend on the gas dissociation required of the process.
Two methods of gas circulation can be employed:
• Overpressure circulation
• Internal fan circulation
(a) (b)
Fig. 2 Micrographs of white nitride layers developed on vacuum-melted AMS

6470 steel. (a) White layer 0.033 mm (0.0013 in.) thick formed after
single-stage nitriding at 525 °C (975 °F) for 60 h with 28% dissociation. Buildup
of white layer at corner was 0.084 mm (0.0033 in.). (b) White layer 0.020 mm
(0.0008 in.) thick formed after double-stage nitriding at 525 °C (975 °F) for 9.5 h
with 25 to 28% dissociation, then at 550 °C (1025 °F) for 50.5 h and 80 to 84%
dissociation. Buildup of white layer at corner was 0.050 mm (0.00020 in.). 2%
nital etch; 150×. Source: Ref 1
Chapter 5: Furnace Equipment and Control Systems / 41
Overpressure circulation simply means ammonia gas flows in and

out due to the pressure difference. This process relies on gas delivery pres-
sure and, most importantly, on part geometry. Complex geometries can
cause 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 stagnation
occurs and nascent nitrogen is not replenished within the hole or cavity,
then a shallow nitrided case may form. The overpressure method of gas
circulation requires careful loading of components.
Internal fan circulation is unquestionably the best method for pre-
venting gas stagnation (Fig. 3). Gas circulation by a fan nearly guarantees
complete accessibility of the process gas throughout the process chamber.
Another benefit of the internal fan system is good temperature 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 stress
pattern in the nitrided component. Internal fan circulation is not necessar-
ily restricted to the bell-type furnace shown in Fig. 3; a horizontal furnace
or even a pit-type furnace design can be used.
Heating bell
Retort
Cooling bell
Work basket Exhaust fan
Heating elements
Work support
Ammonia supply
Exhaust
Circulating fan
Oil seal
Bell-type furnace with heating bell 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 that
it has the integrity to maintain an atmosphere without leaking and meet
the required process control conditions. This means that even if a furnace
is not gastight, it can still be used with an inner-sealed retort or container.
Nitriding furnace designs include:
• Box temper furnace

• Vertical air-circulating furnace
• Lift-off bell furnace
• Fluidized-bed furnace
• Salt bath furnace
The only restriction is that the furnace should have a method of isolat-
ing the process gas from the shop environment. This is accomplished by
processing the work in a sealed (gastight) furnace or in a sealed process
retort or chamber.
Insulation
A furnace that is repeatedly heated up and down through a temperature
range both uses and loses energy. As much of the energy (heat) as possible
needs to be contained within the process chamber, and external heat losses
need to be kept to a safe working minimum temperature. Therefore, the
furnace must be insulated.
Insulation for the reduction of furnace heat losses can take the form of:
• Insulating refractory brick

• Low thermal mass ceramic fiber insulating material
• Vacuum insulation
Some of the major advantages of a nitriding furnace with regard to

insulation and heating elements are discussed below. There is no ideal
insulation material. The type of insulation depends on whether a process
retort is used and how the retort goes into the furnace. Poor mechanical
handling may damage the heating system and the insulation materials.
Relatively Low Temperatures. Because nitriding furnaces usually are
not subjected to high temperature levels, wear and tear on the insulation
and furnace elements is very low. The heating elements are more often
than not of a higher watt density than is necessary, thus ensuring long life.
If the furnace is gas fired, the radiant tubes are usually overdesigned for
the application.
Long Useful Life. The furnace is not subjected to adverse temperature
fluctuations, thus prolonging the life of the heating system and insulation
material. The author personally knows of a used vertical air-circulating
furnace that in 1963 was converted into a nitriding furnace, with new
refractory brick and heating elements added. The furnace is still in opera-
tion 40 years later, with no further element or refractory replacement. If
the refractory lining and heating elements are subjected to mechanical
damage, of course, then their life expectancy will be reduced. However,
unless a nitriding furnace is subjected to poor mechanical handling, it
requires low maintenance.
Energy Costs. The quality and thickness of the insulation will deter-
mine the energy cost of the cycle. The insulation design should take into
account heat losses to the outer furnace casing and the insulation thickness
and quality. Higher quality insulation may have a higher initial cost, but
that cost will be recovered in lower operating costs.
Determining Appropriate Furnace Design

Furnace configuration and design is influenced by:
• Part design: Is the part a tooling item, dies for pressure die casting, or
production work?
• Part loading: How large are the parts to be treated, and what is the
maximum load that is to be charged into the process chamber? This is
important, as it will determine the heat/energy requirements for the
furnace, chamber, and load.
• Frequency of operation: Will the furnace be operated on a jobbing basis
or will there be guaranteed loads with regular frequency of loading?
• Furnace availability: Will the furnace be used for other operations
such as tempering, subcritical annealing, or stress relieving?
Retort Construction
Steels. Heat-resistant type 309 or 310 stainless steel is quite suitable as
a retort material. The most satisfactory material is Inconel; however, it is
expensive. The primary requirement is that the material must not catalyze
ammonia dissociation at the internal surfaces of the process chamber.
Materials that fall into this category are mild or plain carbon steel, and
types 302 and 304 stainless steel.
If used for the process chamber, mild steels and plain carbon steels will
draw away the nitriding reaction from the workpiece surface to the point
that minimal nitriding, if any, will take place. They also lack the necessary
mechanical strength at the process operating temperature, quickly distort-
ing and failing prematurely. Types 302 and 304 stainless steel also lack
mechanical strength and will distort, as well as deteriorate rapidly on the
inside due to corrosion resulting from the breakdown of surface chrome
oxide caused by the nitriding reaction.
Temperature-resistant glass, in theory, is probably the best retort con-
struction material because of its nonreactive nature with ammonia gas.
However, its fragility would result in mechanical damage, making glass

impractical for use in a production nitriding environment.
Insulating refractory firebrick can be used to line the process cham-
ber, provided that it is dense and heat resistant, with low porosity and low
iron content. The mortared joints must be very thin and contain minimal
porosity; otherwise, the ammonia process gas will begin to leak. If the
iron content of the brick is high, the iron will react with the ammonia to
form iron nitrides and begin to erode from the joint, particularly if there is
a fan circulation system in the process chamber.
Enameling. The container can be enameled but, like temperature-
resistant glass, is subject to chipping by mechanical damage and is not a
practical solution for the nitriding process chamber.
Deoxidized copper can be used, but lacks strength at high tempera-
tures. The copper will not take part in the process reaction, being com-
pletely impervious to nascent nitrogen or anhydrous ammonia. Copper is
used as a masking medium to prevent nitriding from taking place on
selected portions of steel parts.
Retort Maintenance
Even if stainless steel or Inconel is used, the retort will not be mainte-
nance free. No matter what the material of construction, the internal retort
walls require a periodic program of regeneration.
Regeneration of the retort is necessary when gas dissociation becomes
difficult to control and when workpieces exhibit inconsistent metallurgical
results. This indicates that the free nitrogen is reacting with the internal
walls of the process chamber, due to deterioration of their chrome oxide
surfaces. Regneration is simply accomplished by heating the empty
process chamber to about 900 °C (1650 °F) and holding at that tempera-
ture for a few hours. This procedure will “burn” out the nitrogen and cause
the chrome to oxidize, thus forming chrome oxide. After the process
chamber cools, the surfaces are cleaned by either shot-blasting or sand-
blasting. The frequency of regeneration correlates with retort usage and
will vary from plant to plant. Once every 3 to 5 years is typical.
Trays and fixtures require the same considerations as the process retort
in terms of construction materials and maintenance. Each time the nitride
operation is run, the trays and fixtures are exposed to the atmosphere and
the process temperature. Therefore, a planned maintenance procedure
must be in place in terms of crack repair to support webs on the trays.
Sealing the Retort to Prevent Ammonia Leaks

Figure 4 shows a nitriding installation that uses a simple seal that fits on
the retort flange. Mild steel or stainless steel studs, nuts, and washers
secure the sealed top cover. If mild steel fasteners are used, dip the thread
Dissociator
Dissociation
Pipette
Seal
Seal
Ammonia
Tank
Work
Basket
Heating Elements Exhaust
Nitriding Retort
Fan
Gas Inlet
Gas Outlet
Fan Motor
Nitriding Furnace
Fig. 4 Nitriding process retort with a seal that fits on the retort flange
into a mixture of oil and flake graphite or molybdenum disulfide to ensure

easy removal after processing. Otherwise, the threads of both the stud and
nut will seize solid by the end of the process cycle.
Other sealing methods include oil seals (commonly used on the lift-off
bell furnace configuration), clamp arrangements, and vacuum seals. Great
care must be taken to ensure that the process chamber is gastight.
Checking and Identifying Seal Leaks. The new seal will usually leak
slightly on the first cycle without serious consequences. After that, the seal
has “bedded” in.
Ammonia leaks can be identified by using a lighted sulfur pipe cleaner
(such as those used by a pipe smoker) that has been dipped into molten
sulfur. Once lit and manually moved around the flange perimeter, a cloud
of dense white ammonium sulfide smoke will appear at any leak source.
Tighten the appropriate studs or ensure that the clamps are secure, and
check the integrity of the seal once again.
Safety Precautions When Using Ammonia

Ammonia is an extremely pungent gas, which can make breathing
impossible if large volumes of ammonia escape. Even if small volumes of
ammonia gas leak from the process chamber, people near the leak source
may experience loss of breath and extremely painful eye irritation. It is
imperative to read the Material Safety Data Sheet or International Chemi-
cal Safety Card when using ammonia gas.
Ammonia gas is also very flammable. If small leaks occur in the seal of
an internal process chamber, the ammonia can ignite and burn vigorously
especially in the presence of oxygen. If part of the sealing method is made
from aluminum, then there is a strong possibility that the aluminum portion
of the seal will melt, thus increasing the fire risk. Large concentrations of
anhydrous ammonia can cause serious injury, even death, due to violent
explosions if the leaking ammonia is exposed to a naked flame. Great care
must be taken to eliminate the presence of oxygen and an ignition source.
Other important safety precautions include (Ref 2):
• Do not allow ammonia to come into contact with mercury; under

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

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

intended only to alert the reader to the dangers of ammonia. It does not
claim to address all possible safety concerns. The user, in conjunction
with equipment and material suppliers and appropriate government 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 of
heating 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 that
holds the process retort.
If the furnace is heated by natural gas, the burners usually are located in
a remote heater box and the heated combustion air is distributed to the
process retort. Alternatively, the combustion products can heat a series of
external radiant tubes that surround the process retort.
Calculation of the heat input requirement uses this formula (Ref 3):
Steel specific heat × Gross load mass × Temperature rise

kW·h =
3412
where the specific heat of steel is usually considered to be 0.15 Btu/lb·°F;

the load mass (in pounds) equals the total load, including retort fixturing;
the temperature rise equals the required temperature minus the ambient
temperature (in °F); and 3412 is the conversion factor from Btu’s (British
thermal units) to kW·h.
Note that the calculated figure is defined as “that energy required to
heat the complete load up from ambient temperature to the process operat-
ing temperature in an hour.” The calculated figure can then be further
divided by the number of hours in which it is both practical and economi-
cal to raise the furnace temperature. Remember, the calculation must be
for the gross load going into the furnace.
Normally the furnace designer will calculate the heat input calculation
based on the maximum furnace operating temperature. This will account
for the total weight to be heated, including:
• Workload weight
• Total weight of the process chamber
• Total weight of the fixtures and work support materials
Bear in mind that the calculation is for the energy input only and does
not account in any way for heat losses through the furnace insulation or
through apertures such as seals, thermocouple holes, fan-drive shaft holes,
or any other aperture.
The calculation for gas heating energy requirements for gross
workload is (Ref 4):
Btu’s = Specific heat × Load mass × Temperature
Once again, the calculated value refers only to the energy required to heat
the furnace in 1 h. To heat in 1 h may be both costly and impractical, mak-
ing it necessary to divide the resulting value by whatever time factor is

convenient. The calculated value does not account for furnace heat losses
through refractories or apertures cut in the furnace. The “add on” kW or
Btu’s will depend on the refractory value of the furnace insulation 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 control
and a dual PC/PLC system (see Fig. 1). This system arrangement must
conform to ISO 9000 and its derivatives for the requirement of:
• Data acquisition, logging, and storage

• Process control, including time, temperature, gas flow, and dilution
gas if applicable
• Historical trends, maintenance, and so on
It is interesting to note that the hardware for the PC/PLC combination is

usually not as expensive as for more conventional control methods—
except, of course, for the cost of programming. The level of programming
depends on the particular set of process considerations.
Temperature Control
Temperature control is best achieved with a thermocouple attached to
the workpiece, or placed as close to the workpiece as possible. After deter-
mination of the number of control zones, the furnace will require at least
two thermocouples per zone: a set-point temperature thermocouple and an
over-temperature thermocouple. The over-temperature thermocouple is
necessary to prevent temperature overshoots and furnace temperature
“runaways.” If the over-temperature thermocouple is removable, always
remember to reload it.
Fitting a Load Thermocouple into the Retort. A hole is cut into the
retort and a pipe with a single gastight sealed end is welded into place in
the box. The thermocouple is inserted into the open end of the tube. The
tube should extend as far as necessary into the retort to show the best tem-
perature control point during the process cycle. The thermocouple should
be calibrated for the control instrument electromagnetic field (EMF)
requirement.
Monitoring Gas Dissociation

Gas dissociation is simply the amount of decomposition of the ammo-
nia gas (Ref 5). During the nitriding process, ammonia gas is continually
decomposing due to the catalytic effect of the steel and the retort. The gas
breaks down into its elemental form of hydrogen and nitrogen, and thus
the exhaust gases consist of ammonia and hydrogen:
(2NH3 + H2 + N2)
Of the above three gases, only ammonia is soluble in water, to the point that
water will absorb as much as 70 times its own volume of ammonia (Ref 6).
The simple method of measuring the dissociation is to use a pipette
(burette) as shown in Fig. 5. To make a measurement, the ammonia gas in
the nitriding box is first admitted into A by opening taps C and D. After the
air has been expelled, taps C and D are closed. During the measurement,
tap E is opened and the water immediately absorbs the undissociated
ammonia. The water takes up precisely the volume previously occupied
by the ammonia, but the remaining N2-H2 gas (dissociated ammonia) does
not dissove in water.
This is a simple and effective method of measuring the gas dissociation.
Other methods include using an oxygen-measuring unit and recalibrating
to feed in percent dissociation of ammonia or hydrogen.
The rate of dissociation at which the process should operate is usually
around 30%. This, of course, will depend on whether a single-stage or
Water B
0 0 0
E N2 + H 2
25 N2 + H 2 25 25
C N2 + H 2
NH3 + N2 + H2
50 50 50
H2 O
75 H2 O 75 75
H 2O
A 100 100 100
25% 50% 75%
D Height of water column in graduated chamber (A) for 25%,

50%, and 75% degree of dissociation
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 not
always) around 15%.
Oxygen Probe Control

Many advances have been made in the use of the oxygen probe control
in the nitriding process. L. Sproge of Sweden and S. Midea of the United
States, both of the Aga Gas Company, reported their computer control
model in a paper that discusses the equilibrium of hydrogen and oxygen
with water vapor (Ref 7). In order for the control system to be effective,
the flow of the process gas components must be accurately metered using
mass flow controllers. These units will meter very precise gas flows to the
process.
Nitriding Sensors
Development of the nitriding sensor has significantly contributed to
nitriding process control (Ref 8, 9). This system has been designed using
the principle of a solid-state electrolyte, measuring variations in the
magnetic and electrical properties of a steel as temperature increases and
surface chemistry changes. The sensor is placed in the furnace process-
ing chamber and then calibrated according to the desired nitriding result,
the parameters of the parts being treated, and the surface area of the
workpieces.
The control system allows the technician to control the thickness of the
compound layer and case depth. If water vapor or oxygen is present within
the nitriding atmosphere, the oxygen partial pressure can be used as an
indirect measure of the nitriding potential of the process atmosphere. The
resulting generated signal from the sensor indicates the partial pressure of
the oxygen and the degree of dissociation occurring. Initial part clean-
liness is essential, because any surface contamination will affect the sen-
sor results.
REFERENCES
1. Gas Nitriding, Heat Treating, Cleaning, and Finishing, Vol 2, 8th ed.,
Metals Handbook, American Society for Metals, 1964, p 149–163
2. “Safety Requirements for the Storage and Handling of Anhydrous
Ammonia,” ANSI K61.1, American National Standards Institute
3. G. Totten, G.R. Garsombke, D. Pye, and R.W. Reynoldson, Heat
Treatment Equipment, Steel Heat Treatment Handbook, G.E. Totten
and M.A.H. Howes, Ed., Marcel Dekker, 1997, p 293–482
4. D. Pye, “Understanding Nitriding and Ferritic Nitro Carburizing,”
course notes, Pye Metallurgical Consulting
5. J. Grosch, Heat Treatment with Gaseous Atmospheres, Steel Heat
Treatment Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel
Dekker, 1997, p 663–720
6. V.O. Homerberg and J.O. Walstead, A Study of the Nitriding Process:

Part 1, ASST Nitriding Symposium, American Society for Steel
Treaters, 1929
7. L. Sproge and S. Midea, Analysis and Control of Nitriding and 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 Heat
Dekker, 1997, p 483–526
9. Marathon Sensors Inc., Cincinnati, OH, www.marathonsensors.com

CHAPTER 6
Salt Bath Nitriding
AN ALTERNATIVE to the gas nitriding process was sought in the mid-

1930s that would produce a more uniform and better metallurgically
formed case. Notable chemical companies conducted many experiments
to find an appropriate alternative method. A liquid, researchers thought,
would fulfill the uniformity requirement through surface contact of the
liquid to the steel. The depth and quality of case would be determined by
the chemical composition of the liquid. A heat source would be necessary
to drive the nitrogen into the steel surface.
A cyanide-based salt, it was discovered, fulfilled the requirements of
the process, and salt bath nitriding began. Salt bath nitriding is in essence
the same process as gas nitriding; only the medium is different. Salt bath
nitriding offers extremely good case depth uniformity.
Salt bath nitriding utilizes the melting of a salt containing a rich nitro-
gen source. When heat is applied from either an internal or external
source, the salt melts and liberates nitrogen to the steel for diffusion.
When a steel workpiece is introduced into the salt and heated up to a tem-
perature in the molten salt, similar reactions begin to occur as in gas
nitriding; that is, controlled amounts of nitrogen are released to diffuse
into the steel surface.
Salts Used and Process Advantages

Nitriding Salts. Early on, a cyanide-based salt, such as NaCN, was used
with other salts. A typical nitriding salt mixture would be (Ref 1):
• NaCN, 30% • 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 the

early 1950s because it required a lower capital investment than gas
nitriding. Operating costs also were considerably lower, making it a cost-
effective method of nitriding steel. An added benefit was that salt bath
nitriding 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 equipment
were:
• Relatively low operating cost

• Low maintenance
• Easy operation, requiring a lower skill level
• Small batch-type furnace generally used, occupying less floor space
• Easy startup, easy shutdown
• Slightly faster diffusion cycle times
Types of Salt Bath Nitriding Processes

Since the early development of the salt bath nitriding process, many
derivatives have appeared (Ref 1). The two major European companies that
pioneered the salt bath process were the Cassel Division of Imperial Chem-
ical Industries (ICI) in England and the Durfferit Division of Degussa in
Germany.
The early process was developed by ICI, which was eventually
acquired by Degussa in the late 1970s. The salt, known by ICI as N.S.450,
relied on the decomposition of cyanide to cyanate. In Germany, Degussa
developed the two-component process of forming both nitrides and some
carbides in the steel surface (i.e., nitrocarburizing). The German process
was introduced as the Tufftride process. Tufftride became popular in
Europe in the early 1950s and was introduced to the United States in the
early 1960s. It was quickly accepted as a reliable procedure and identified
by the Society of Automotive Engineers (SAE) designation of AMS 2755.
The process simply involves:
• Preheating
• Immersion into the molten salt bath
• Cooling in still air or a suitable quench medium
• Postwashing
• Optional polishing or oiling
The surface finish produced by the process is usually matte black, which
is corrosion resistant.
Because of emerging environmental concerns, Degussa pioneered the
use of a low-cyanide salt. Kolene Corporation developed a low-cyanide
Chapter 6: Salt Bath Nitriding / 55
salt process, now known as the Melonite process (Ref 2), which offered an
alternative to the Tufftride process. The process uses a cyanide-free salt,
the composition of which is specified under the SAE designation AMS
2753. Another derivative of the Melonite process by Degussa is known as
the QPQ (“Quench-Polish-Quench”) process (QPQ is a trademark of the
Kolene Corp.).
Both the Melonite process and the QPQ process require the part to be
quenched into a molten oxidizing bath to neutralize any residual cyanide
that might be present. After the quench procedure, the part is mechanically
polished, followed by a further resurface oxidation process.
Salt Bath Nitriding Equipment and Procedure

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

A newly made-up nitriding salt bath should not be used immediately
on completion of the melt procedure. The bath first must be “aged” or
“conditioned.”
If the newly molten salt bath is used as soon as the salt is molten, the
salt will immediately begin a vigorous surface reaction on the steel,
attacking the surface and causing a form of pitting known as orange peel.
This means that the bath is vigorously decarburizing the surface of the
steel.
Fig. 1 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
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.) deep
will take approximately 12 to 16 h to age. It is important to note that no
work should be processed through the bath during the aging process.
Fig. 2 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. 3 SAE 5115 (UNS G51150), chromium-manganese low-carbon steel,

salt bath nitrided 90 min at 580 °C (1075 °F), oxidizing molten salt
quenched. 500×, nital etch. Courtesy of Kolene Corp.
As with gas nitriding, workpieces must be cleaned prior to salt bath

nitriding to prevent any oil, grease, paint, and so forth from contaminating
the bath. In addition, the steel surfaces must be free of oxides or other
contaminants.
A cyanate decomposition percentage up to 25% normally will produce
acceptable surface composition and hardness results. If the carbonate con-
tent is high, the temperature of the bath should be lowered to around 440
to 455 °C (825 to 850 °F). This will cause precipitation of the carbonate,
which can then be scooped out as sludge from the bottom of the bath.
The salts do not last indefinitely. If used on a daily basis, the bath
should be discarded once every 3 to 31/2 months and replaced with a new
bath. Again, care must be taken to age the new bath.
Bath Replacement
The bath is ready for replacement if it is becoming difficult to maintain
a cyanate level around 25% or if signs of corrosion are evident on the steel
surface. Be careful not to confuse the corrosion with that caused by post-
washing of the steel.
The cyanide to cyanate control level is not the same for all steels
processed through the nitriding bath and will vary for various steels. Sodium
cyanide (NaCN) levels are normally maintained as follows:
• High-speed steels: 15% minimum NaCN

• Hot-work tool steel: 20% minimum NaCN
• Alloy steels: Approximately 25% NaCN
Bath Testing and Analysis

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

The following is a list of materials and equipment required for titration
testing of the nitriding salt bath:
• One burette calibrated to read directly in percentage of NaCN. Each

graduation on the burette is equal to 0.5 mL (0.5 cm3)
• One sample spoon designed to hold approximately 1 g of powdered
sample
• One mortar and pestle for crushing the sample
• One bottle of chemical reagent (0.2 N AgNO3)
• Two 250 mL glass beakers and two glass stirring rods
• One wide-mouth 500 mL bottle
• One once-dropping bottle of indicating solution (10% KI)
• One vial of lead carbonate powder (2PbCO3·Pb (OH)2)
It is advisable to maintain a shift logbook of titrations of the nitriding bath

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

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

Sodium Carbonate. The determination of sodium carbonate in the
bath consists of the following steps:
• Accurately weigh a 0.5 g sample of the bath, transfer to a 300 mL

Erlenmeyer flask, and add 100 mL of distilled water.
• After the sample has dissolved, add 25 mL of a neutral 19% barium
chloride solution. Stop flask tightly, and let stand for 2 h.
• Filter and wash the precipitate with neutral 1% barium chloride until
the washings give no color with phenolphthalein.
• Place the filter paper and precipitate in the original flask. Add 50 mL
of distilled water and a measured excess of standard N/10 (0.1 N)
hydrochloric acid (HCl). Boil the solution for 5 min and allow to cool.
• Add 4 to 5 drops methyl red indicator and titrate with standard N/10
(0.1 N) sodium hydroxide (NaOH).
The following formula can then be used to determine the sodium carbon-
ate percentage:
(VHCl · normality HCl – VNaOH · normality NaOH) · 5.3

%NA2 CO3 =
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 cyanate
in the bath consists of the following steps:
• Accurately weigh an approximate 1 to 2 g sample of the bath and

transfer to a 650 mL Kjeldahl flask. Use a 1 g sample when checking a
nitriding bath and a 2 g sample when checking a carburizing bath.
• Add 100 mL ammonia-free distilled water, 2 or 3 selenized Hengar
granules, and 5 mL of concentrated sulfuric acid. Digest under the
hood or in a Kjeldahl digesting unit until the volume reaches approxi-
mately 25 mL. Allow to cool. This digestion drives off all the cyanide
nitrogen 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 include
a Kjeldahl trap before entering the condenser. Place the tip of the deliv-
ery tube into a 500 mL Erlenmeyer flask containing 50 mL of saturated
boric acid solution with methyl red indicator added. The tip of the
delivery tube must be below the surface of the boric acid solution.
• Distill off 150 to 200 mL of liquid, disconnect, and wash down the
condenser with ammonia-free distilled water.
• Titrate the distillate with standard N/10 (0.1 N ) hydrochloric acid.
The following formula can then be used to determine the sodium cyanate
percentage:
VHCl · normality HCl · 6.5

% NaCNO =
sample weight
where VHCl is the volume in mL and sample weight is measured in grams.

The normality of a solution is the number of gram equivalent weights of
solute per liter. Note: To avoid running appreciable blanks, reagent-grade
chemicals and ammonia-free water should be used for all solutions.
Bath Maintenance
Maintenance of the salt bath and related equipment is critical and quite
simple. The procedures can be broken down into daily, weekly, and monthly
activities (Ref 1):
Daily maintenance consists of:
• Analyzing the bath for cyanide and cyanate content at the start of each
shift
• Checking that the temperature control instrument is in proper working
order
• Checking the cleanliness of the bath by using a perforated scoop and
desludging the bath before the start of the shift
Weekly maintenance consists of:
• Checking the cleanliness of the bath around the top (that is, the part
above salt level)
• Removing the salt pot and checking the integrity of the external sur-
faces, especially if the salt bath is gas fired. Watch for signs of excess
scale and “ballooning” at the bottom of the pot, which means that the
pot wall is thinning and the salt is too heavy for the pot at that wall
thickness.
Monthly maintenance consists of:
• Checking the gas burner train if the nitride bath is gas fired. Make sure
that the burner train linkages are secure and that modulating valves (if
fitted) are operational.
• Determining that the burner ignition system is operational and that the
flame rod is clean
• Checking that no traces of nitriding salt are present on the cleanout
port
• Checking that the elements are fully immersed if the furnace is 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. Deterioration
of 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 potential
breakages.
Operating the Salt Bath

As discussed, furnace operating personnel must ensure that the bath is
fully aged, regenerated, and of the correct salt analysis. Once these param-
eters have been established, the bath is ready for operation. As with any
salt bath procedure, the steel workpiece must be thoroughly preheated and
free of any surface moisture—either on the steel, the support fixture, or
the fastening wire—before immersing it into the molten salt. On immer-
sion, the workpiece will freeze the immediate salt surrounding it and
cause 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 excessive
heat. This will reduce the risk of distortion through thermal shock.
Due to its nature of being molten when under heat, the salt will slowly
oxidize at the salt/air interface and cause a breakdown of the cyanide to
carbonate. This will slowly begin to affect the internal surfaces of the pot,
thus reducing pot life.
Safety Precautions
Safety precautions are as important with low-temperature salts as with
high-temperature salts (Ref 3). Some of the more important considera-
tions are:
• Always preheat to both reduce thermal shock and remove moisture

from the steel part surface.
• Do not mix nitrate salts with cyanide-based salts, as this poses a 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 gas
nitriding. However, the salt bath furnace tends to be a more compact unit
that occupies less floor space than its gas nitriding counterpart.
In addition to batch-type equipment, continuous systems are available
for salt bath nitriding. Most systems are encapsulated, with a protective
front and an observation panel.
REFERENCES
1. Liquid Nitriding, Heat Treating, Vol 4, ASM Handbook, ASM Interna-
tional, 1991, p 410–419
2. J. Easterday, Salt Bath Nitriding Proceedings (Detroit), Kolene Corp.,
Oct 1995
3. Cassel Manual of Heat Treatment and Case Hardening, 7th ed., Impe-
rial Chemical Industries, Ltd., Jan 1964

CHAPTER 7
Control of the
Compound Zone or
White Layer
THE COMPOUND ZONE is more commonly known as the white

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

ing, Inc.
Fig. 2 Formation of the compound zone. Source: Ref 1
the necessary surface metallurgical phase should be determined by the

part application.
A Test to Determine the

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

Zone by the Two-Stage Process
The first method of reducing the compound zone was developed by Dr.
Carl Floe of the Massachusetts Institute of Technology (see Chapter 1 for
historical background). He developed what was originally known as the
Floe process, now more commonly called the two-stage or sometimes the
double-stage process. The purpose of the two-stage process is to reduce
the thickness of the compound zone on the immediate surface of the steel
by reducing the nitriding potential at an elevated temperature. The first
stage of the process ensures rapid formation of the white layer, and the
second stage arrests the formation of the white layer without allowing the
diffusion zone to be denitrided.
The two-stage nitriding process is a relatively simple procedure (Fig. 3).
Nitriding is carried out using ammonia gas with a dissociation of approxi-
mately 30% at 495 °C (925 °F). This is followed by raising the temperature
during the last third of the cycle to 565 °C (1050 °F) with a dissociation
rate of 75 to 80%. In some applications, formation of the surface com-
pound zone is desirable. If the layer is not required, it can be removed
mechanically (by grinding), or chemically (using a cyanide solution devel-
oped by Bell Helicopters) (Ref 4).
Chapter 7: Control of the Compound Zone or White Layer / 67
Fig. 3 Typical two-stage process
Other Methods for

Controlling Compound Zone Formation
Nitriding Potential. Control of the compound zone can be achieved by
controlling the nitriding potential of the process gas. In a 1973 paper, B.J.
Lightfoot and D.H. Jack of the University of Leeds (Ref 5) suggested that
at nitriding potentials that completely avoid white layer formation, only
steels that contain aluminum will nitride satisfactorily. Steels that depend
on chromium as the primary nitride-forming element will nitride very
slowly.
The white layer constituents must be stable to provide a fine dispersion
of chromium nitrides in ferrite. Strict control of the nitriding potential
restricts white layer growth.
Building on Adolph Machlet’s early work with dilution techniques (see
the later discussion in this chapter), Thomas Bell of the University of Liv-
erpool began work in the early 1970s on dilution of the ammonia gas by
hydrogen (Ref 2). His objective was to establish the mechanism of com-
pound layer formation and to determine how the layer thickness can be
managed. By monitoring the nitriding atmosphere (an ammonia-hydrogen
mixture), and knowing the input gas (whether ammonia to enrich the
nitrogen content or hydrogen to dilute the nitrogen content), he discovered
that the nitriding potential can be controlled to produce components with a
white layer thickness of no more than 0.004 mm (0.00016 in.).
The dilution process was first developed by Machlet when he patented
his process in 1913 (see Chapter 1 for further details). The patent was
granted for the nitrogenization of low-alloy steels and cast irons, using
ammonia gas diluted with hydrogen. Machlet had observed compound
layer formation without realizing its significance, yet recognized that the
layer could be controlled by dilutant gases such as hydrogen. The sole
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 this
method, a Canadian company (Nitrex Metal, Inc.) manufactures computer-
controlled furnace equipment to control compound layer formation. The
equipment seems to be enjoying good success (Ref 6).
Ion Nitriding. By ionizing molecular gases and preparing the steel 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 as
required 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 diffusion
of nitrogen into the steel surface is a contentious, often-discussed subject.
Formation of both the compound layer and the diffusion layer is based on:
• Time
• Temperature
• Gas composition
• Steel analysis
• Steel surface condition
The nitriding procedure is based on diffusion at a selected temperature in

relation to:
• Potential for nitride networking

• Growth
• Distortion
• Corrosion resistance
The preceding considerations are factors in determining temperature.

But what of the required time at temperature to establish the desired total
Chapter 7: Control of the Compound Zone or White Layer / 69
case depth? In the late 1930s and early 1940s, F.E. Harris determined a
simple formula for the effect of diffusion at temperature (Ref 7). The 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 0.00150
470 875 0.00155
475 885 0.00172
480 900 0.00195
500 930 0.00210
510 950 0.00217
515 960 0.00230
525 975 0.00243
540 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 what
might be considered abnormally high alloying contents (such as might be
found in stainless steels). The factors also do not account for surface con-
dition and variations in the particular cast/melt. The factor values are
approximate and should be used only as a guide. Similar guides for esti-
mating case depth for plasma nitriding and ferritic nitrocarburizing can be
found in Chapters 12 and 21, respectively.
Remember, the higher the alloying content of the steel is, the slower the
diffusion rate of nitrogen into the steel surface. The lower the alloying
content is, the faster the diffusion rate.
Temperature uniformity within the process retort is critical to uniform
case depth as well as uniform case metallurgy. If temperature variations
occur within the process retort, both the compound layer and the case
depth also can vary.
REFERENCES
1. M.A.J. Somers and E.J. Mittemeijer, Härterei-Technische Mitteilun-
gen, Vol 47 (No. 5), 1992
2. T. Bell et al., Controlled Nitriding in Ammonia-Hydrogen Mixtures,
Heat Treatment ’73, The Metals Society, Dec 1973, reprinted in Source
Book on Nitriding, American Society for Metals, 1977, p 259–265
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 Society
for Metals, 1977, p 242–247
4. D.A. Dashfield, Nitriding Problems and Their Solutions, Met. Prog.,
Feb 1964, p 88–93
5. B.J. Lightfoot and D.H. Jack, Kinetics of Nitriding With and Without
White-Layer Formation, Heat Treatment ’73, The Metals Society, Dec
1973, reprinted in Source Book on Nitriding, American Society for
Metals, 1977, p 248–254
6. W.K. Liliental, G.J. Tymowski, and C.D. Morawski, Typical Nitriding
Faults and Their Prevention Through the Controlled Gas Nitriding
Process, Ind. Heat., Jan 1995, p 39–44
7. F.E. Harris, Case Depth, Met. Prog., Aug 1943, p 265–272

CHAPTER 8
Ion Nitriding
GAS IONIZATION is a method of causing a gas to develop an electrical

imbalance. At a particular pressure and electrical potential, the gas will
glow—not unlike the gas used in neon light tubes. The phenomenon of gas
ionization was first investigated in the 1800s, when studies made in the upper
reaches of the Arctic Circle in Norway showed the aurora borealis to be low-
pressure ionized air that exhibits a characteristic spectrum of the rarer gases
present in the upper atmosphere. Observable in an indefinite region of the
night sky around the magnetic pole, the ionization is influenced by the 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 the
nitriding process that utilizes it is known as ion nitriding. Ion nitriding is
also called:
• Plasma nitriding
• Glow discharge nitriding
• Plasma ion nitriding
Derivatives of the three processes listed previously include:
• Continuous direct current (dc) plasma nitriding

• Pulsed dc plasma nitriding
• Hot-wall furnace system
• Cold-wall furnace system
History of Ion Nitriding

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

entrepreneur Dr. Bernhard Berghaus (Ref 1). Together they were able to
develop, control, and market the process as a gas nitriding alternative that
offered acceptable control of the compound zone. Development of the
plasma relies on:
• Low (less than atmospheric) pressure

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

The process is based on the phenomenon of current flowing between
two electrodes placed in a sealed gaseous environment. The gas within the
tube acts as an electrical conductor and carries the current from one side to
the other as it would if it were a wire conductor (Fig. 1). The gaseous
atoms become excited and are propelled along a very short “mean free
path” and collide with one another. When this occurs, energy is released
and a “glow” occurs—hence the name “glow discharge nitriding.” The
color of the glow is determined by the type of gas used.
At normal atmospheric temperature and pressure, the resulting energy
is too low to have any significant use as an energy source in terms of 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 free
path that each gas molecule has to move to find another molecule to
impact. Thus the plasma can then generate heat, but in an amount insuffi-
cient to heat the workpiece surface for plasma processing. In other words,
Chapter 8: Ion Nitriding / 73
Fig. 1 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 not
release much usable energy. (b) At low pressure, for example, (0.013 Pa, or 2 ×
10–6 psi), very few gas molecules exist and collide infrequently, releasing low
energy. The glow appears almost like a fog. (c) At higher pressure, say 1.3 to 13 Pa
(2 × 10–4 to 2 × 10–3 psi), the gas molecules move freely and impact frequently,
releasing usable plasma energy that glows brightly and crisply.
because of the infrequency of molecular impact, the resulting energy still

cannot be used as a heating medium.
There is no ideal pressure value, but there is a range in which the pres-
sure can be adjusted to suit the operating parameters of the particular
material and geometry. That pressure range is given to be 50 to 500 Pa
(0.007 to 0.07 psi). Control of the process chamber pressure will 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 part
will become intermittent. Where there is no glow on the steel surface,
there will be no nitride formation. This means that the voltage has failed to
support the plasma ignition at that particular pressure. In addition, high-
temperature areas can occur at sharp corners, often resulting in localized
overheating and even burning. If the pressure is too low (i.e., toward high
vacuum), then the glow seam around the part will be very “foggy,” 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 occur
much like a lightning strike. This occurrence is known as the arc discharge
region. If the voltage is too low and the pressure too high, the glow will
disappear and nothing will happen.
Glow Discharge Characteristics

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

current limiting resistance is reduced, thus causing a voltage drop between
the cathodic workpiece and the anodic internal vessel wall. Within this
area, voltage stability cannot be maintained.
Subnormal Glow Discharge. In this region the glow discharge is
beginning to ignite and a very fuzzy glow will be seen.
Normal Glow Region. It is at this point on the Paschen curve that a
glow seam will completely cover the work. Its thickness will be deter-
mined by the chamber vacuum pressure and the process voltage.
Abnormal Glow Region. It is in this region that the glow seam will
completely cover the steel work surface uniformly, following the geomet-
ric profile of the workpiece. One must now adjust the process pressure to
ensure penetration of blind holes and cavities. However, the ratio of hole
diameter to hole depth must be considered:
1× hole diameter : 4× hole depth
The blind hole depth can equal four diameters. This ratio will follow for
small holes down to a diameter of approximately 4 mm (1/8 in.). However,
the ratio will double if the hole is a straight-through hole to:
1× hole diameter : 8× hole depth
A through-hole depth may equal eight diameters. The ratio is valid for
small hole penetrations, but for large holes from 50 mm (2 in.), the rule
will not apply; the glow seam can be “forced” down the hole by the
adjustment of pressure. It is in this region of abnormal glow that plasma
nitriding takes place and where ideal conditions exist.
Arc Discharge Region. As the current density increases, a noticeable
increase in the voltage drop will occur, causing an appreciable increase in
the power density at the work surface. As power density increases, the
temperature of the steel work surface rises to the point of serious over-
heating, resulting in metallurgical problems such as grain growth, local-
ized melting, and pitting. As the power intensity builds, the potential for
an arc will occur. This arc, visible through the process chamber sight
glass, looks like a lightning strike.
Process Control
Process parameters requiring good control are:
• Current density
• Power
• Process chamber pressure
• Gas composition
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 a
process chamber and a constant voltage is applied to the chamber with the
appropriate cathode potential and anode, then the gas in the partial pres-
sure will ionize. If a steel part is placed in contact with the cathode so that
it is at the same electrical potential, then not only will the partial pressure
gas “glow,” it also will create heat in the steel part at the cathode potential
due to the kinetic energy generated by ionic bombardment.
If the process gas used in the process chamber is diffusible in steel
when heat is applied, then the diffusion processes can be accomplished.
The diffusible gases are nitrogen, hydrogen, carbon monoxide, boron,
sulfur, and methane.
Elements. In the cold-wall or continuous dc type of system, there are
no heating elements within the furnace, simply because the thermal
energy is created by the ionic bombardment of the workpiece by the free
gas ions. The workpiece is heated by kinetic energy from ionic 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 the
workpieces 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 used
for other surface treatments such as diffusion and deposition techniques.
Diffusion techniques include (Ref 3):
• Plasma-assisted nitriding
• Plasma-assisted ferritic nitrocarburizing
• Plasma-assisted carbonitriding
• Plasma-assisted carburizing
Deposition techniques, more commonly known as “thin film”

processes, can be accomplished through plasma generation. The thin-film
deposition technique can be divided into two groups: (a) tribological
(wear and corrosion resistance) and (b) decorative coatings.
Thin-film deposition techniques deposit a thin metallic film onto a care-
fully prepared steel surface, resulting in hardness and corrosion resistance.
The thin-film coating is usually (but not always) preceded by a plasma
nitriding procedure, which produces a hard substrate to which the 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 of

nitrogen and hydrogen. The reverse will happen on cooling the decom-
posed gas. In ion nitriding, unlike gas nitriding where ammonia gas is
used, the gases are brought in as separate gases:
N2 + H2
Because the gases are not in a combined form, the metallurgist is able to
vary the nitriding potential by varying the proportions of the individual
gases. By varying the hydrogen-to-nitrogen ratio of the elemental gases, the
compound zone (white layer) formation can be controlled. When the 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 into
the 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 nitrogen
molecule, 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 surface
metallurgy. The ratio can be a low nitrogen to hydrogen ratio, or a high
nitrogen 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 the
material 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 surface
of the work by an action known as sputtering. The impact of the nitrogen
ions bombarding the work surface dislodges the contaminants, which are
removed by the vacuum pumping system. Contaminant removal can be
loosely described as atomic cleaning and allows nitrogen to diffuse into
the work surface:
N+ → Work surface = Sputtered Fe and sputtered contamination
Reaction 3. As a result of the impact of the sputtered iron atoms, case

formation begins at the work surface of iron nitrides:
Sputtered Fe + N = FeN
At this point, intensive surface cleaning of the workpiece occurs due to

the sputtering action on the surface. This can be likened to atomic shot
blasting where the carrier medium is the air blast and the cleaning medium
is the steel shot. In this instance the gas ion is like the steel shot, and the
electrical imbalance is like the air blast. The surface becomes atomically
cleaned.
Reaction 4. At the work surface the breakdown of FeN begins under
the influence of continual plasma bombardment. The plasma causes insta-
bility of the FeN, which begins to break down into the ε-phase, followed
by the γ′-phase and an iron/nitrogen compound zone (Fig. 4):
FeN → Fe2N + N
Fe2N → Fe3N + N (ε-phase)
Fe3N → Fe4N + N (γ′-phase)
Fe4N → Fe + N (iron/nitrogen compound zone)
Fig. 4 Glow discharge ion nitriding mechanisms (Koelbel’s models). Note the
voltage profile on top. The potential drop is greatest near the workpiece
so this is where the ions have the most kinetic energy and this is where the
plasma will glow brightest.
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 a
conventional heat source. The plasma energy is the heat source, and the
process temperature can be adjusted to suit the steel by manipulation of
plasma voltage and pressure. However, use of a lower process 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, and
carburizing (Fig. 5). Because nitrogen is diffusing from all angles of the
corner, a normal reaction of nitrogen saturation occurs at the corner (par-
ticularly on gear teeth). If allowed to proceed unchecked, a supersaturated
solution of nitrogen can form at the corners, leading to nitride networking
throughout the corner region. The net result is that the corner can become
very brittle and prematurely fail by chipping or spalling (Fig. 6, 7). If the
Fig. 5 Illustration of the corner effect due to the multidirection of nitrogen

into the steel surface. Nitride networking also is shown.
Fig. 6 Chipping at the pressure point of a gear tooth. Courtesy of Pye Metal-
lurgical Consulting, Inc.
Fig. 7 Extrusion die with surface exfoliation. Courtesy of Pye Metallurgical

Consulting, Inc.
nitrogen is controlled to reduce the nitriding potential to suit the part

application, then the risk of nitride networking is considerably reduced or
nearly eliminated.
Degradation of Surface Finish

Sputter cleaning and ion nitriding generally enhance the surface condi-
tion of the workpiece. Ionic bombardment of the steel surface reduces
turning lines and high points. Of course, care must be exercised when
using denser gases such as argon for ionic bombardment. Argon will tend
to etch the surface to the point where the surface will become pitted.
When aggressive sputter cleaning is necessary, it is strongly recom-
mended that the argon be blended with hydrogen to a ratio of 95% H2 to
5% Ar up to a maximum of 90% H2 to 10% Ar.
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 mixed
phase compound zone of both ε and γ′ (usually around 50% each). High
internal stresses result from the different phase volumes, which means
that the crystal interfaces are inherently weak. The thicker the compound
zone, the weaker it becomes, causing weak crystal boundaries within the
zone to fail even under small loads. This is particularly evident on parts
such as H13 hot-work tool steel for aluminum extrusion dies (Fig. 8a). If
the compound zone is allowed to develop into a thick layer, the die will
bend when loaded by the aluminum billet (Fig. 8b). As the die bends, the
loaded face becomes concave and the back face becomes convex. The
compound zone on the loaded face will undergo compression and crack.
This also will occur if the effective case depth of the nitrided layer
exceeds 0.25 mm (0.010 in.).
A similar effect will occur on an H13 hot-work press forging die if the
case is too deep and the compound layer too thick. As the press loads the
die face, the surface of the die becomes compressed and the formed case
deforms at the surface, leading to surface cracking and exfoliation. This
is not serious if the exfoliation occurs on areas of the die that are not in
use, but when exfoliation occurs on a working face it becomes a major
problem.
Cracks can initiate quite easily due to impact loading or incorrect pre-
heat treatment. If the core hardness mechanical properties are lower than
the load conditions of the treated steel, then the core is likely to fail to sup-
port the nitrided case on the surface, no matter how well the nitriding has
Fig. 8 Cross section through an extrusion die. (a) Bearing face and relief
clearance area. (b) Aluminum billet for extrusion against the die face,
which is soft in the core. Side A will compress and crack; B will stretch and tear
on the case.
been done. Therefore, the thinner the compound zone (to the point of
elimination), the more ductile the steel and the better its fatigue properties.
Ion nitriding allows control of the thickness of the compound zone.
Process Gases
Metallurgical-grade nitrogen and high-purity hydrogen are the gases
primarily used for ion nitriding. Argon can be used, but only to assist in
component cleaning (known as sputter cleaning) before the nitriding
sequence. Methane can also be used to deliver controlled amounts of 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 actively
promote a dominant ε compound phase.
Accurate control of the gas delivery into the process retort ensures
accurate control of the nitriding metallurgy. Precise metering of each gas
is accomplished by mass flow controllers.
Process Parameters
In conventional gas nitriding, time, temperature, and dissociation (or
salt analysis for salt bath nitriding) are the key process parameters that
must be controlled. In ion nitriding, the control parameters include:
• Time
• Temperature
• Pressure
• Current density
• Amperage
• Voltage
• Gas flow
• Gas ratio
Depending on the type of plasma technology employed, other parameters

can be controlled to ensure repeatable process conditions.
Temperature uniformity during the nitriding process is critical. With-
out it, the formed case will be nonuniform. Some latitude in temperature
tolerance can be allowed, but usually no more than ±5.5 °C (10 °F) of the
process set point. Remember, the higher the process temperature, the
greater the risk of nitride networks due to the potential for an excess of
nitrogen in iron. This also means that nonuniform case depths will occur
across the temperature range within the process retort. For example, a tem-
perature difference of 35 °C (60 °F) over a 24 h cycle period can result in a
case depth variation of up to 0.038 mm (0.0015 in.). Obviously, the longer
the process cycle time, the greater the case depth variation. This principle
applies to gas, salt bath, as well as plasma processing techniques.
Plasma Generation Philosophies

There are two methods of plasma generation: continuous dc and pulsed dc.
Continuous dc means that plasma is generated with a particular dc
current based on a given work surface area. The voltage requirement will
vary with the pressure (Fig. 9). With this type of system, the process heat
generation normally is derived from the kinetic energy created on ionic
bombardment. This means that the furnace does not have heating ele-
ments as are often seen in conventional furnaces.
Pulsed dc can be likened to entering a darkened room and switching on
the light. On leaving the room, the light is switched off. The peak voltage
from the power source is constant, but the duration of the power to the
light is variable. Pulsed dc power supplies have the ability to switch the
power on and off as required by component geometry. The pulsed dc peak
voltage can be varied according to the part and chamber configuration
(Fig. 10). The equipment will be discussed in greater detail in Chapter 9.
Fig. 9 Continuous dc power plasma nitriding
Fig. 10 Pulsed dc plasma showing Pmax, Pmin, and Ptemp (where P = power) in
relation to power input versus time. Note that the voltage can be
adjusted, as can the duration of the pulse. Courtesy of Plateg USA
Advantages
Here are a few of the advantages of the plasma generation technique for
nitriding:
• Based on known technology

• Environmentally friendly gases
• No fire risk
• Shorter cycle times
• Better furnace utilization during continuous production
• Very low operating costs
• Minimal operator involvement
• Low process gas consumption
• Repeatable process parameters
• Repeatable metallurgy
• Ability to treat almost any steel
• Ease of selective nitriding
• Low-maintenance equipment
Environmental Impact. Because ion nitriding gases are used in such

small volumes and are nontoxic and environmentally friendly, the process
will cause no harm to the air or operational personnel. The process can be
successfully used in cell manufacturing techniques. The volumes of gases
used in the process are considerably less than those used in gas nitriding.
There is no burnoff or gas scrubbing to consider when using the plasma
nitriding process.
Oxynitriding
The oxynitriding process has grown in popularity over the past two
decades, especially in pulsed plasma nitride processing. The pulsed
plasma nitride unit offers the capability of a controlled backfill of mois-
ture in the form of either a vapor or oxygen-bearing gas. The purpose of
the oxynitriding process is to form a controlled oxide layer on the surface
of the treated steel. Once the oxide barrier has been formed, there is a
resistance to corrosion. The degree of corrosion resistance will be deter-
mined by the thickness of the oxide layer, which in turn is determined by
both time and temperature. The oxynitriding process can be performed in
gas, salt, or plasma.
The procedure is done on completion of the nitride cycle, when the 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 effect
is that the nitrided surface is deliberately oxidized to provide a corrosion-
resistant, oxygen-rich surface layer. Upon completion of the cooldown, the
furnace bell is opened and the oxynitrided work is removed.
Figure 12 shows a plain carbon-manganese steel piston rod that has
been nitrided, followed by the controlled oxynitride procedure. The center
rod shows the rod before the start of treatment. The two rods at the right
have been oxynitrided, and the two rods at the left are untreated; all four
have been subjected to salt spray testing.
The process gases used for the controlled oxynitriding procedure are
environmentally friendly and pose no threat to the ecology or the immedi-
ate environment. The surface finish of the steel after the procedure is dark
blue, almost matte black. The oxidation layer is usually around 1 µm
thick, but can be varied according to the cooldown time and the time held
at an elevated temperature (around 900 °F, or 480 °C). Generally the pro-
cedure is to go into cooldown immediately after nitriding and commence
the controlled oxidation treatment. It is a simple, effective procedure that
adds no significant time to the overall process.
Fig. 11 Schematic illustration of the oxynitriding process
Fig. 12 Examples of oxynitrided piston rods. Center rod: before treatment.

Two rods at left: untreated and subjected to salt spray testing. Two
rods at right: treated, then subjected to salt spray testing. Material is similar to UNS
G41400 and H41400 chromium-molybdenum steels. Courtesy of Plateg USA
Oxynitriding produces a pleasing cosmetic appearance and ensures a high

degree of surface protection against corrosion. The procedure can be applied
to items such as boring bars, cutting tool holders, broaches, drill bits, and
cutting tools. In fact, oxynitriding can be applied to any workpiece subject to
corrosive conditions and is not restricted to cutting tool applications.
REFERENCES
1. F. Hombeck, Forward View of Ion Nitriding Applications, Ion Nitrid-
ing, T. Spalvins, Ed., ASM International, 1987, p 169–178
2. C.K. Jones, D.J. Sturges, and S.W. Martin, Glow Discharge Nitriding
in Production, Met. Prog., Dec 1973, reprinted in Source Book on
Nitriding, P.M. Unterweiser and A.G. Gray, Ed., American Society for
Metals, 1977, p 186–187
3. Heat Treating, Vol 4, ASM Handbook, ASM International, 1991

CHAPTER 9
Ion Nitriding Equipment
ION NITRIDING EQUIPMENT falls under two categories: cold-wall

continuous 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 of
the 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
Fig. 1 Schematic of a typical cold-wall continuous dc plasma nitriding sys-

tem. Source: Ref 1
conventional vacuum furnace construction. As the furnace is essentially a

vacuum furnace, it will require:
• Vacuum pumping system

• Power source
• Gas supply
• Process controller
Process Parameters
Plasma nitriding involves more process control parameters than gas
nitriding does. Parameters for plasma nitriding include:
• Input voltage
• Amperage
• Chamber pressure
• Workpiece temperature
• Current density
• Nitrogen gas flow
• Hydrogen gas flow
• Methane gas flow
• Oxidizing gas for oxynitriding flow rate
• Process time
• Temperature rise rate
• Temperature cooldown rate
These parameters, along with a few derivatives of them, are controlled by

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

The plasma power generator is usually a solid-state unit designed to
produce continuous dc power from line voltage through to the variable
Chapter 9: Ion Nitriding Equipment / 91
voltage control to the furnace cathode power feedthroughs situated

within the process chamber. Because the power feedthrough allows con-
tinuous power from the plasma generator, its insulating characteristics
must be of a very high quality to insulate the cathode from the anode.
Thus an electrical bias is set up between the process chamber and the
cathodic workpiece.
Some older power generator designs used a straight dc power system.
Glow stability was sometimes a problem when plasma nitriding complex
part geometries and holes, particularly blind holes. This led to the use of
high-process voltages in order to address the nitriding of complex 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 this
chapter). Should the arc be initiated, there is a serious risk of both metallur-
gical and mechanical damage to the part, caused by overheating, possible
burning, and possible stock removal at sharp corners on the workpiece.
Therefore, some form of arc detection and suppression electronic-control
system is mandatory.
The power supply is used to set up a bias between the workpiece and
the vacuum chamber wall. The disadvantage of these units is that when
operating in the lower regions of the glow, it is difficult for the glow to
penetrate along the form of the part being treated, particularly in parts
with blind holes or complex geometries. In such cases, both pressure and
voltage must be varied; however, if high voltage and pressure levels are
selected, there is a serious risk that the arc discharge region could be
reached and arcing could occur.
Heating Elements
A cold-wall furnace normally has no heating elements. Heat into the
part is generated by the kinetic energy developed by the ionic 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 would
most likely be found within the furnace process chamber and usually are
electrically isolated to prevent them from being nitrided.
Furnace Thermocouples
Perhaps even more important than control of process temperature is
control of part temperature. Unlike more conventional heat treatment
methods, temperature generally is measured at the part rather than the
chamber. In conventional heat treating, the temperature generally is meas-
ured by a thermocouple located within the process chamber. It is often
incorrectly 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 at
any other point within the process chamber.
Many factors within the furnace chamber influence temperature unifor-

mity. Therefore, part temperature—not chamber temperature—must be
measured. The part and process temperature are measured by considering
the thinnest part and the thickest part within the process chamber. The part
temperatures are usually held to within a tolerance band of ±5 °C (10 °F).
If the thermocouple cannot be attached to the workpiece, then it should
be attached to dummy test coupons that are representative of the 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 from
the thermocouple is generated in millivolts and transmitted back to either
a process controller or a data logger to record the process control parame-
ters. The signal transmission method can be either conventional hard wire
or fiber optics.
Temperature uniformity throughout the process chamber is mandatory
to ensure uniform surface metallurgy, surface hardness, core hardness,
and case thickness. Without temperature uniformity, serious metallurgical
conditions can result.
Gas Flow
Good process gas-flow input control is critical. Some of the earlier sys-
tems used flowmeters. Although this method worked to some extent, it
was not accurate enough. Later methods used micrometer needle valves,
but these too could not offer a high degree of accuracy and repeatability.
A more accurate method of controlling gas delivery is the mass-flow
controller, an electronic device that allows precise flow control. Use of the
mass-flow controller is not limited to plasma nitriding; it can also be used
in gas nitriding, particularly when using the dilution method.
Gas flow can affect nitriding quality. The required gas flow remains
constant only if the work surface area also remains constant. Most heat-
treat shops cannot guarantee constant same-surface area loads; therefore,
the gas flow requirements will vary from load to load, and will vary in
relation to the type of surface metallurgy required.
Depending on the process retort size, the usual gas flow consumption
rate can be up to approximately 100 L/h. A 600 × 900 mm (24 × 36 in.)
furnace would use up to 30 L/h maximum. The reason for low gas 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 Pump
Chamber pressure is controlled by a simple mechanical vacuum pump, a
rotary-vane vacuum pump (Fig. 2, 3), or a combination system of a mechan-
ical pump and a roots blower. With an operating pressure of approximately
10 to 500 Pa (1.4 × 10–3 to 7.3 × 10–2 psi or 0.075 to 3.75 torr), the roots
blower improves the system pumping speed and compression. Lower-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 are
necessary.
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 in
oversizing the vacuum pump to pump down in less than 10 min.
The vacuum pump-out port on the furnace chamber is usually located in
the hearth. The furnace hearth and the vacuum pump are connected by a
flexible stainless steel connector.
Maintenance of the vacuum pump is critical but simple. The vacuum
pump oil must be checked weekly, and changed every 3 to 6 months. The
correct gas ballast setting is mandatory. Remember, one of the process
gases is hydrogen, which is a very soluble gas as well as highly flammable
and explosive if mixed in the correct combination with oxygen. If the gas
ballast setting is done incorrectly, an explosion or fire could result. Note:
Refer to the manufacturer’s operating and maintenance manual for the
correct method of setting the vacuum pump gas ballast. If the vacuum
pump should fail, the unit will not plasma nitride and production will stop.
Cathode and Anode

The vacuum vessel acts as the anode potential, and the furnace hearth is
attached to a specially designed and insulated power feedthrough, which
in turn is connected to the dc power source. The principal concern with the
Fig. 3 Four stages in the cycle of a rotary-vane pump: induction, isolation, compression,
and exhaust. Source: Ref 3
power feedthrough is that it be insulated with a very dense ceramic mate-

rial with high insulation characteristics. The insulator is then covered with
a steel cover and each cover is electrically separated.
Hot-Wall Pulsed dc Plasma Nitriding

The hot-wall pulsed dc plasma nitride furnace is similar in some
respects to the cold-wall continuous dc system. The major physical differ-
ence between the two is that the hot-wall system is fitted with an insulated
heating bell furnace around the process chamber. This means that work-
piece heating can now be separated from plasma generation, because
plasma 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 a
typical schematic layout of a hot-wall plasma furnace. Table 1 compares
the cold-wall and hot-wall plasma nitriding systems.
Fig. 4 Schematic of a hot-wall pulsed dc plasma nitriding furnace and associ-

ated equipment. Courtesy of Plateg GmbH
Table 1 Comparison between hot-wall and cold-wall plasma ion nitriding systems
Question Cold wall Hot wall
At what temperature is plasma started? Room temperature At a suitable elevated temperature, usually
around 200 °C (400 °F)
Why is plasma started at those temperatures? The cold-wall furnace uses a constant dc sys- The hot-wall furnace utilizes a partial pressure
tem, which requires plasma voltages around condition using hydrogen or nitrogen as a ther-
600 to 800 V. Mechanical and metallurgical mal conductance gas. The vacuum retort is
damage to the workpiece surface may occur by heated only by external heaters and not by
processing so close to the arc discharge region. plasma voltage. This means that the input volt-
age is not as high (400 to 500 V), and is away
from the arc discharge region.
Why is there a difference in plasma generation To heat and maintain the workpiece tempera- Because the part is preheated the power
voltages? ture, the required power (kW) corresponds to a required to maintain the workpiece tempera-
current density on the workpiece of approxi- ture at this partial pressure corresponds to 1–2
mately 10 A/m2 (1 A/ft2) at this partial pressure A/m2 (0.09–0.2 A/ft2). A lower voltage is
and voltage. enough to produce these currents.
How does the heatup rate compare? The cold wall usually requires more time for The hot-wall heatup of the port is usually about
heatup. 15 times faster.
Why pulse the power input? With a constant voltage input, there is a con- With a pulsed voltage, high voltage can be
stant heat output. Reducing voltages to reduce used without risk of overheating the part, or
temperature changes the current density and taking the part to the point of arc discharge.
the cathode fall voltage distance (glow seam); This means that the other parameters need not
therefore, other parameters also must be be changed.
changed.
What happens to the heat? With a cold-wall system the released electron The hot-wall furnace combined with the pulse
is hotter than the ion. This means that the elec- technology uses external blowers to prevent
tron goes back to the furnace wall (anode) and excessive wall heating. The wall temperature
creates heat. Heatup of the wall will continue, can safely rise to around 650 °C (1200 °F)
necessitating water cooling of the wall to dissi- without concern over heat buildup.
pate heat.
What happens if the glow seam must be pushed Increasing the operating pressure causes a cor- Using the hot-wall pulsed power system with
into a deep blind hole? responding increase in current density, followed the same pressure/temperature combination
by an increase in part surface temperature. and the same voltage/current relationship as
the cold wall, the plasma energy can be main-
tained by varying the duty cycle (pulse varia-
tion), even with a changing voltage and current
density.
Source: Ref 4
Another substantial difference is that the plasma power generation

pack is now a pulsed dc generator, rather than a continuous dc unit. At
first, this might not seem to be a striking development. However, the
Paschen curve (Fig. 5) shows that high voltages are needed to generate a
plasma glow. The use of high process voltages at ambient temperatures is
quite dangerous to the workpiece because of the potential for arc dis-
charge. Pulse technology nullifies that risk simply by pulsing the power at
such 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 are
located in the external heating bell furnace. The elements can be mounted
either in the insulation wall of the external bell (as with traditional meth-
ods) or directly onto the external wall of the process vessel. This is a more
efficient method of preheating the process chamber and allows for better
heat transfer into the process vessel (Fig. 6).
Operation of the hot-wall furnace involves these steps:
• First, the process area is loaded onto the furnace hearth.

• The process chamber and bell are closed, thus sealing the process
chamber from the atmosphere.
• The chamber is then evacuated by the mechanical vacuum pumping
system down to the appropriate vacuum level.
Fig. 5 Voltage versus current characteristics (a Paschen curve) for different

types of discharge in argon
Fig. 6 Arrangement of heating elements for hot-wall furnaces with correspon-

ding temperature profiles. Length in mm. Courtesy of Plateg GmbH
• The process chamber is backfilled with hydrogen gas, and the external
heating elements are switched on to heat the gas; the workload is
heated by convection.
• Once the process temperature has reached approximately 230 °C
(450 °F), the sputter cleaning procedure begins. Hydrogen ions atom-
ically blast the workpiece surfaces, a more thorough and effective
method 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. If
the workpiece surfaces are badly contaminated, a gaseous mixture of
up to 10% argon and 90% hydrogen should be used. To avoid surface
etching, do not use more than 10% argon in the mixture. Sputter
cleaning is further discussed later in this chapter.
• Once the sputter cleaning operation is complete, the process chamber
temperature is raised to the appropriate nitriding temperature for the
required holding time and with the appropriate gas flow to achieve
the required surface metallurgy.
The primary argument against use of the hot-wall system is that it

requires energy to heat the process retort, and that this expenditure is
unnecessary. However, no matter which surface hardening process is
chosen, be it nitriding or carburizing, either a process retort or a process
chamber must be heated. In plasma nitriding, it is necessary to heat both
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 energy
required provides external heating to the process retort and conductive
heat to the workload. This means that only the minimal energy neces-
sary to generate the plasma glow seam is required. The continuous dc
system, however, requires energy from the plasma and supplemental
internal elements.
Pulsed Power Supply

The power source used in the pulsed plasma nitriding system is the
heart of the system. It must provide the operator accurate control of puls-
ing time and required process voltage in order to create the plasma neces-
sary for surface preparation and nitriding completion with regard to both
the steel and the part geometry. The plasma generator supply must:
• Create the physical conditions for the abnormal glow discharge (see
the Paschen curve in Fig. 5)
• Provide good temperature uniformity within the workload area
• Heat the workload
• Prevent arc discharge conditions
With a conventional dc power supply, fulfillment of the above condi-

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

to use a pulsed dc power supply with the following special conditions:
• The power should not be of a sine-wave type. That means the power
will only be reduced and not completely shut off or isolated. It is thus
necessary to have a defined power-on, power-off system—that is, a
defined square form so that the power jumps from zero voltage into
the allowed gap of the abnormal glow discharge region.
• The length of the pulse should be shorter than the development time of
the arc. This means less than 100 µs so that the arc is suppressed.
Interruption of the arc will be possible during each pulse.
• The pause that follows each pulse should be short enough to allow an
easy ignition for the next pulse—for example, less than 1 µs.
• The ratio of pulse to pause should be variable over a wide range to
control the power input by the plasma in the workload so that it will be
possible to use an auxiliary heating system (such as external heating)
for better temperature within the process chamber.
With the pulsed dc system, the point at which Pplas occurs is only during
the pulse within the gap. The sum of the areas under each pulse is equal to
area under the line Ptemp. In the figure the pulse time (the width of the
square wave) is constant. The pause time can vary to balance the energy
losses Ptemp. By this method, the temperature adjustment is separated from
Fig. 8 Power characteristics (a) Continuous dc. (b) Pulsed dc. Abnormal glow
discharge (see Fig. 5) occurs between Pmin and Pmax. The power
required for plasma nitriding, Pplas, is in this region. Ptemp is the time average of
power required to maintain the workload at the desired temperature. In (b) the
pulse widths are regulated so the area under the pulses equals the area under
Ptemp. Source: Ref 5
the other process parameter. Typical pulse time values are between 5 and
100 µs, while pause time can vary between 5 to 200 µs (Ref 5).
Pulsed dc power gives the furnace user three additional process control
variables:
• Process voltage
• Time of power on
• Time of power off
The sum of the on time and off time of one pulse makes up one cycle, so
the control parameters could be described as voltage, frequency, and ratio
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 chamber
internal pressure. These additional controllable variables allow the opera-
tor to better control the process and to address such part geometries as
blind holes, cavities, and complex shapes. Occurrences such as arc dis-
charge and hollow cathode potential (discussed later in this chapter) are
still a concern, but not as great as with the continuous dc system (Ref 5).
The process controller will now control:
• Furnace temperature
• Part temperature
• Part temperature ramp rate to process temperature
• Plasma process voltage
• Partial pressure control
• Current density
• Nitrogen gas flow
• Hydrogen gas flow
• Methane gas flow (if used)
• Argon gas (if used)
• Oxygen gas (if used)
• Process cycle time
• Pulse power time on
• Pulse power time off
• Cooldown rate
Because of the numerous process variables available, greater use can be

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

There are five methods of work cooling from the process temperature to
an acceptable exposure temperature after plasma nitriding. Selection
depends on the primary design of the furnace system. The cooling methods
described in this section apply to both cold-wall and hot-wall furnaces.
Free-Cool Method. Free cooling is achieved by simply turning off the
plasma power and allowing the work to cool from the process temperature
down to below 150 °C (300 °F) under partial pressure conditions. The
time required depends on workpiece mass and surface area. The disadvan-
tage of this method is that it is very inefficient. There is insufficient
process gas within the furnace chamber under vacuum conditions to allow
effective cooling with convection methods of heat transfer.
Cooling under Partial Pressure and Convective Gas Conditions.
This method involves shutting off the plasma power and backfilling the
chamber to a partial pressure with a nonreactive gas—usually clean, dry
nitrogen (not metallurgical grade). The work will cool down somewhat
faster than with the vacuum free-cooling method because there is a con-
vection gas within the process work chamber. However, there is no means
of 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 on
sufficient nitrogen being introduced into the work chamber so that the
chamber pressure exceeds atmospheric pressure. Gas agitation is now by
an internal recirculation fan. This method is also more efficient than the
free-cooling method.
Cooling Using a Combination of Nitrogen Backfilled Gas in Con-
junction with a Water-Cooled Heat Exchanger. This method of cool-
ing is the most efficient method available to the plasma nitride furnace
user. The furnace is fitted with a finned-tube copper heat exchanger with
ambient-temperature water passing through the heat exchange coils as the
backfilled gas passes over the heat exchanger. Movement of the cooling
gas is made possible by the internal recirculation fan.
Postoxidation Treatment. The fifth option is to cool down to a specific
temperature and conduct a postoxidation treatment (oxynitriding) to
enhance the surface corrosion properties of the workpiece.
Postcooling. Once cooling has taken place, the process chamber can then
be opened safely without risk of oxidation or discoloration of the work 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 be
blue to almost black, depending on the postoxidation treatment temperature.
Other Considerations for Ion

Nitriding Equipment and Processing
Other important considerations associated with ion nitriding include the
hollow cathode effect, sputter cleaning, furnace loading, pressure/voltage
relationships, workpiece masking, and furnace configuration options.
Hollow Cathode
Hollow cathode is an area of low vacuum pressure where the plasma
glow seam does not follow the precise contour of the part being treated.
For example, if the plasma glow seam dips slightly into a blind hole, free
electrons (energy) are trapped in the area beneath the glow seam. The glow
seam holds the free electrons within the hole. The electrons as energy then
begin to migrate through the wall of the hole. This raises the temperature of
the steel, which can lead to overheating and sometimes burning (Fig. 9). If
the parts are placed too close together, a similar effect can take place and
cause localized overheating or sometimes burning (Fig. 10).
Supplementary Internal Heating Elements

Supplementary internal heating elements are sometimes used in the
cold-wall system. However, it must be remembered that everything within
the vacuum chamber will be nitrided, including the heating elements.
The hot-wall furnace operates in a completely different manner ini-
tially. The furnace is evacuated by the vacuum pump to the appropriate
vacuum level and backfilled with hydrogen to a partial pressure. The
external hot-wall heaters are started and the part is heated by convection
gases and not by plasma energy. At ambient temperatures and high plasma
Fig. 9 Illustration of trapped free electrons in a blind hole having the potential
to overheat the corners
Fig. 10 Parts too close together cause the “hollow cathode” effect, leading to
possible overheating and burning.
generation voltages, plasma energy is dangerous, as is the case with the

cold-wall system.
With the hot-wall system, when the temperature reaches approximately
230 to 260 °C (450 to 500 °F), the plasma generation voltage is started
and the part is then sputter cleaned under hydrogen gas up to the 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 work
surfaces prior to nitriding. The sputter gas typically is hydrogen, the lightest
gas, which cleans as well as acts as a reducing gas. Any surface oxide will
be reduced by the hydrogen to the base metal. If the workpiece surface is
seriously contaminated, the hydrogen can be mixed with argon to increase
the gas density and its cleaning ability. Again, apply caution when using
argon to ensure that the mixture is not so severe as to cause surface etching
of the steel component (maximum ratio 90% hydrogen, 10% argon).
The sputtering time depends on the prior surface condition of the steel
being treated. Generally, one would introduce several temperature steps
above the initial convective heating at 230, 370, and 450 °C (450, 700,
and 850 °F), then up to the final selected nitriding temperature. The hold-
ing time at the selected sputter cleaning stage would be approximately
10 min (depending, of course, on how much cleaning is necessary). Note
that as the temperature increases, the sputter cleaning continues (Fig. 11).
Fig. 11 Time versus temperature curve illustrating the sputter cleaning

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

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

It is most important to have good control over both the vacuum level
and the voltage as these two process parameters will strongly influence the
plasma glow-seam coverage of the part (or, more accurately, they will
determine the point of plasma ignition on the part). If the pressure is too
high with a normal process voltage for the part geometry, the part will
begin to lose the plasma glow-seam coverage. If this occurs, the area with-
out plasma ignition will not nitride (Fig. 12a and b).
Masking
A part can be masked quite simply by remembering the following:
What plasma can see, it will nitride. Mechanical masking involves wiring
steel shim stock material to the part. The thickness of the shim stock is
usually about 0.05 to 0.1 mm (0.002 to 0.004 in.).
If the part is placed on the furnace hearth, the side of the part that is in
contact with the hearth will not nitride. If it is necessary to nitride its
underside, then the part can be mounted on steel points (Fig. 13).
If two parts are placed directly on top of each other, the two contacting
faces will not nitride and will remain soft. Holes can be masked by insert-
ing a simple steel plug. For a threaded hole, the first two or three turns of a
stud can be screwed into it. There is no need to coat the stud threads.
Proprietary “paints” are available for masking that will resist the effects
of sputtering (unlike early paints that sputtered off the surface in particles
that made their way to the inner vessel wall, changing the electrical char-
acteristics of the anode over time). Nitriding stop-off paints will resist the
effects 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
Fig. 12 Glow-seam coverage dependence on voltage. (a) Workpiece during

plasma nitriding with continuous dc glow discharge in the normal
region (see Fig. 5). The total dc power input is not high enough to cover the 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 during
nitriding with pulsed dc glow discharge. By using the pulsed dc with a repetition
frequency of about 10 kHz, the complete surface is covered and uniform nitrid-
ing results. The average power input is the same as in (a). The peak power of each
pulse is higher so that the region of an abnormal glow (Fig. 5) is applied during
this pulse. Pulsed technology allows complete coverage of the surface with high
peak powers, but low average power input, so that workpiece overheating can be
avoided. Courtesy of Plateg GmbH.
be sputtered off during the cleaning process and transported to the anode
vessel, where they will coat the inner surface and affect its electrical char-
acteristics. Thus, copper plating is not recommended for plasma nitriding.
Additional information on masking prior to plasma nitriding operations
can be found in Chapter 15, “Stop-Off Procedures for Selective Nitriding.”
Configurations of Plasma Nitriding Units

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

• Available roof height
• Production requirements
• Part geometry
• Available furnace design
Fig. 13 Workpiece in furnace. (a) Component mounted on support points.

(b) Detail showing the effect of support on local case formation
Fig. 14 Vertically configured plasma system. This system has two chambers
or bells, so one can operate while the other is being loaded or
unloaded. Courtesy of Plateg GmbH
Summary: Advantages of Plasma Nitriding

Plasma nitriding, particularly pulsed plasma nitriding, has become a
mature technology capable of processing workloads efficiently and 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.
• No fire risk: Hydrogen, used for sputter cleaning and as a dilution gas,
presents no fire risk whatsoever because of the minuscule amounts of
oxygen remaining within the process chamber under vacuum conditions.
• No obnoxious smells: Neither nitrogen nor hydrogen will produce any
offensive smells, nor any adverse skin reactions.
• Minimal distortion: Because the process can control the type of surface
metallurgy created on the workpiece, compound zone thickness can be
better controlled, leading to lower overall size growth. In addition, better
advantage can be taken of the lower process temperatures. The process
temperature will act as a stabilizer, or an additional tempering proce-
dure. Thus any retained austenite that might be present as a result of the
prehardening and tempering procedure will be decomposed, leading to
better dimensional stability.
• Clean work: The work surface is usually very clean. However, oxygen
from the air can discolor the work surface if the vessel is opened at too
high a temperature. The work surface can also be discolored by an
overly intensive sputter-cleaning program.
• Repeatable results: Consistent and repeatable results can be achieved
that suit the application.
• Elimination of nitride networks: Problems associated with nitride net-
works can be overcome by manipulating the nitriding potential. This
is achieved simply by reducing the amount of required nitrogen for the
process.
• Process management: The system is almost self-managing, taking full
advantage of PC/PLC combinations for better control. Preventive main-
tenance can be better planned, and remote troubleshooting and process
control 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 operator
supervision (load/unload/program/initiate), provides good utiliza-
tion of floor space, and results in reduced energy costs because of
shorter cycle times compared to traditional nitriding. The capital
cost of the equipment is higher than for gas nitriding but is offset by
better plant utilization due to faster process cycles and more repeat-
able metallurgy.
• Integration into cell manufacture: The pulsed plasma nitride furnace
can be integrated into the manufacturing line. Figure 15 shows a pre-
cision gear manufacturing facility that has successfully integrated
heat-treatment production requirements into the gear cutting line. A
variety of small helical and spiral gear pinions and planetary gears are
nitrided. This system can also be built into a robotic handling system,
completely automating the process line (though only with a single
product line, generally automotive components). Such automated sys-
tems can considerably reduce operating costs.
Fig. 15 Single bell unit suitable for integration into a gear manufacturing pro-
duction line. Courtesy of Plateg GmbH
REFERENCES
1. D. Pye, Nitriding Techniques and Methods, Steel Heat Treatment Hand-
book, G.E. Totten and M.A.W. Howes, Ed., Marcel Dekker, 1997, p 744
2. N. Harris, Oil Sealed Mechanical Pumps, Modern Vacuum Practice,
McGraw Hill, 1989, p 71
3. N. Harris, Oil Sealed Mechanical Pumps, Modern Vacuum Practice,
McGraw Hill, 1989, p 72
4. D. Pye, “Practical Nitriding” course notes, 1986
5. R. Gruen, Pulse Plasma Treatment: The Innovation for Ion Nitriding,
Ion Nitriding Proceedings, ASM International, 1987, p 143–147

CHAPTER 10
Nitriding in
Fluidized Beds
A FLUIDIZED-BED FURNACE SYSTEM can be used for the gas
nitriding process. The fluidized-bed furnace is a unique metallurgical pro-
cessing tool that enables the user to complete most heat treatment
processes, including surface treatments. The discussion here focuses on
fluidized-bed nitriding.
Previous chapters have discussed gas nitriding, salt bath nitriding, and
plasma ion nitriding. Each of these procedures is conducted at a process
temperature of approximately 500 °C (930 °F). The fluidized-bed furnace
uses ammonia gas for its nitrogen source, whereas the salt bath uses
cyanide. Fluidized-bed nitriding is similar in process technique to gas
nitriding and similar in the method of heat transfer to salt bath nitriding.
The fluidized bed exhibits the same characteristics as a liquid, with one
exception: A fluidized bed is not wet. The technique of fluidization involves
the disturbance of a bed of finely divided particles, which behave as a liquid
would behave. This is accomplished by passing a gas at sufficient volume
and pressure so as to separate microscopically the fine particles. If the 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 are
critical. The bed does not require large volumes of gas, only that necessary
to separate the particles (Ref 1) (Fig. 1).
The particles in this instance are aluminum oxide. Good heat transfer
takes place from the heating medium to the aluminum oxide particles,
which in turn transfer the heat to the workpiece. When the bed is in opera-
tion, its surface of finely divided aluminum oxide particles bubbles just
like water bubbles when air passes through it (Fig. 2).
Heating Method
The methods of heating a fluidized bed are very similar to those used to
heat an atmosphere-type furnace or a salt bath. The heating system can be
electrical or gas, and internal or external.
Fig. 1 Various types of contact in fluidized beds. Source: Ref 1
Fig. 2 Liquid-like behavior of gas fluidized beds. Source: Ref 1
External Resistance Heating. The aluminum oxide is usually contained

in the heat-resisting material located inside the furnace casing, with the
heating elements mounted onto the furnace installation material (Fig. 3).
Several gases can be used to fluidize the aluminum oxide particles:
• Nitrogen • 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 very

similar to that of the immersed electrodes in a salt bath. The difference in
this instance is that the internal heating elements are usually sheathed. It is
a very simple method of heating the fluidized bed and provides good heat
transfer between the heating elements and the aluminum oxide.
Gas-Heated Fluidized Beds. Gas heating is usually less expensive
than electrical heating (depending, of course, on locality and geography).
Chapter 10: Nitriding in Fluidized Beds / 113
Fig. 3 External resistance heating. (a) Fluidized-bed furnace with external

heating by electrical resistance elements: (1) pivoting cover in two parts;
(2) insulation; (3) refractory material; (4) fluidized bed; (5) resistance elements; (6)
intake for fluidized gas (air or nitrogen); (7) parts to be treated. (b) Recirculation of
fluidizing gas in a fluidized-bed externally heated by electrical resistance. Source:
Ref 1
Once again, heating of the fluidized-bed process chamber can be either

external or internal.
External Combustion Heating. When the fluidized-bed container is
heated by an external gas combustion system, the container is manufac-
tured from the same heat-resisting material as it would be if manufactured
Fig. 4 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 to
be treated. Source: Ref 1
for an electrical heating system (Fig. 5). The fluidized-bed container is

contained within the furnace combustion chamber, which in turn is 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 temperature
after the workload has been removed from the furnace and a new load
introduced makes it an attractive choice for nitriding. The bed does not
require conditioning or purging when changing from one atmosphere sys-
tem to another, making it productive and economical in terms of operating
costs and throughput. The fluidized bed also permits easy change from
one process system to another (e.g., from nitriding to ferritic 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 complete
the 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
Fig. 5 Immersed-element ceramic retort fluidized-bed. Source: Ref 1
Fig. 6 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
on volume of ammonia gas in relation to workpiece surface area. Because

the 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 grades
of steel can be processed effectively using the fluidized-bed technique.
The types of steels that can be treated, including all the stainless steel
grades, will be described in Chapter 12. Stainless steels particularly
require good gas flow control and the appropriate gas dissociation.
The fluidized process cycle times, as in the gas nitriding process, are
governed by the laws of diffusion. In other words, the diffusion rate of nas-
cent nitrogen into the steel surface is the same for fluidized-bed nitriding as
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 8
shows typical processing times for nitride case depths.
Oxynitriding
Oxynitriding can be accomplished in the fluidized-bed furnace much
like it is accomplished in gas nitriding. This means that on completion of
the nitriding cycle, controlled amounts of moisture are added to the process
chamber.
The oxynitriding process forms a very thin surface oxide layer on the
immediate surface of the workpiece. This deliberately oxidized surface
layer is resistant to some aspects of corrosion (though not all). In general,
the process is used for applications where an expensive material such as
stainless steel is being replaced by low-carbon steel with an enhanced sur-
face condition. The oxynitriding process is gaining in popularity, particu-
larly in the automotive industry, in both Europe and the United States.
Fig. 7 Heating rates to 1000 °C (1830 °F) for cylinders of varying diameter in
different types of heat treatment furnaces. Source: Ref 1
Fig. 8 Total nitride case depth versus time in a fluidized bed at 525 °C (975 °F).
Source: Ref 1
Operating the Fluid Bed for Nitriding

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

Measurement of the gas process dissociation is not the same as for gas
nitriding. The gas dissociation measurement of ammonia in the 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 an
effective method of measuring the gaseous activity within the nitriding
process chamber.
With a fluidized bed, the method is effective only if the exhausted ammo-
nia can be captured and the dissociation measured. Such measurement is
Table 1 Recommended fluidized-bed nitriding procedures

Recommended time Temperature
Material min max °C °F
Carbon and low-alloy steels 1h 2h 580±5 1075±10

Tool and die steels (structural) 30 min 3h 540–580 1000–1075
Tool steels (cutting) 5 min 1h 540–580 1000–1075
Corrosion- and heat-resistant steels 1h 2h 580±5 1075±10
Ductile, malleable, and gray cast iron 1h 4h 580±5 1075±10
Powder metal products (ferrous) 30 min 2h 580±5 1075±10
Source: Ref 1
Table 2 Depth of compound layer after fluidized-bed nitriding

Case depth
mm in.
Material min max min max
Carbon and low-alloy steels 0.0038 0.03 0.00015 0.001

Tool and die steels (structural) 0.003 0.013 0.0001 0.0005
Tool and die steels (cutting) 0.003 0.0001 ...
Corrosion- and heat-resistant steels 0.0038 0.03 0.00015 0.001
Ductile, malleable, and gray cast iron 0.0038 0.03 0.00015 0.001
Powder metal products (ferrous) 0.0038 0.0 0.00015 0.001
Source: Ref 1
based on process gas flow in relation to work surface area being treated.
This means that if the work surface area is not constant (as in a typical com-
mercial heat treatment shop), and if the gas flow rate remains constant for
each load of work processed, then the gas dissociation will be different for
each load. Studies on various materials and various load surface areas
must be made to ensure a reasonable chance of consistent and repeatable
nitriding.
ACKNOWLEDGMENT
Grateful acknowledgment is given to Ray Reynoldson of Quality Heat
Treatment Pty Ltd., Turbo Drive, North Bayswater 3153, Australia, for his
valued assistance with this chapter.
REFERENCES
1. R.W. Reynoldson, Theory and Practice, Heat Treatment in Fluidized
Bed Furnaces, ASM International, 1993, p 3–9
2. C. Dawes and D.F. Tranter, Nitrotec Surface Treatment Technology,
Heat Treat. Met., Vol 12 (No. 3), 1985, p 70–76

CHAPTER 12
Steels For Nitriding
MANY STEELS are commercially nitrided:
• Aluminum-containing low-alloy steels, including the Nitralloy group

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

One of the most difficult tasks in nitriding is to select a steel in relation
to the operating environment of the part that will not only ensure good
nitriding results but will also be cost effective, easy to machine or fabri-
cate, and functional. Several questions must be addressed during the
selection process:
• What is the product to be manufactured and how complex is the part

geometry?
• Under what type of operating conditions will the workpiece operate?
Will there be compressive loads, cyclic loads, impact loads, or tensile
loads?
• Will the workpiece operate under abrasive conditions?
• Will the workpiece operate under corrosive conditions?
• What will be the operating temperature of the part? Will it be in a hot
or cold environment?
Table 1 Compositions of selected nitridable steels

Composition, %
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 0.1
32 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 ... 0.90–1.30
Nitralloy 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 3.0

L6 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 0.7
D2 1.65 ... ... 11.5 0.6 ... 0.1 0.5
S1 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 4.3
H10 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 9.5

T4 0.8 4 0.7 1.6 18 5
T1 0.75 4 ... 1 18 ...
T15 1.5 5 ... 5 12.5 5
M42 1.08 4 9.5 1.2 1.5 8
M41 0.92 4 5 1.8 6.5 5
M3 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 preharden
and temper operation.
• Will there be adequate part lubrication?

• Is further machining after nitriding a consideration? For example, will
the part undergo grinding, lapping, or polishing?
After the engineer has gathered the necessary information, the search
for the appropriate steel can begin. If several steels are suitable for a par-
ticular application, price and availability become additional considera-
tions (Ref 1).
Chapter 12: Steels for Nitriding / 127
Requirements for a Nitriding Steel

In the early years of nitriding, Adolph Fry at Krupp Steel (Ref 2) recog-
nized that certain steels responded better from a metallurgical standpoint
in terms of surface hardness, core hardness, distortion, cycle time at tem-
perature, and the formation of stable nitrides. Fry discovered that certain
elements will respond more readily than others to form stable nitrides dur-
ing the nitriding process, and this led to the development of the Nitralloy
group of steels (see Table 1 for compositions). Of the alloying elements
commonly used in commercial steels, aluminum, chromium, vanadium,
tungsten, and molybdenum are beneficial in nitriding because they form
nitrides that are stable at nitriding temperatures. The effects of specific
alloying elements are discussed later in this chapter. Figures 1 and 2 show
the influence of alloying elements on hardness after nitriding and depth of
nitriding.
Aluminum will form very hard nitrides in the nitrided steel surface.
Generally the maximum amount of aluminum permitted in the steel is in
the region of 1.5%. Above 1% Al will lead to surface cracking under
extreme surface load conditions. This is because the core hardness of the
material is usually very ductile. If a highly ductile workpiece undergoes
severe loading, then there is a strong possibility that the surface of the case
will lead to crack propagation.
Fig. 1 Effect of alloying elements on hardness after nitriding. In steels 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. 2 Effect of alloying elements on depth of nitriding measured at 400 HV.

Nitriding was carried out at 520 °C (970 °F) for 8 h. Source: Ref 3
Molybdenum will form stable nitrides at the nitriding temperature and

will reduce the risk of surface embrittlement at the nitriding temperature.
Chromium will also form stable nitrides at the nitriding temperature;
however, the high chromium content found in some stainless steels makes
them more difficult to nitride. Chromium reacts with oxygen to form a
chrome oxide barrier on the surface, which must be broken down by depas-
sivation in order for nitriding to be effective. The higher the percentage of
available chromium at the steel surface, the more difficult the steel will be to
nitride. The positive side of this is usually high surface hardness values.
Vanadium in a nitriding steel also is conducive to the formation of sta-
ble nitrides. In addition, fine grain toughness will be exhibited within the
formed case.
Tungsten enables the steel to retain its hardness at high operating tem-
peratures with no loss of surface hardness. Depending on the tungsten
content and the general composition, the nitrided steel is able to operate at
temperatures up to 590 °C (1100 °F) with enhanced wear characteristics
and no appreciable loss of surface hardness.
Silicon is considered to be a good nitride former. Though it is usually
present as either an oxidizer or a stabilizer, silicon generally is not of 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. Steels
that do not contain those elements, such as the mild steels and low-carbon
steels, will also nitride but will have lower surface hardness because the
case formation is limited to pure iron nitride. However, the corrosion
resistance of low-carbon steels will be greatly enhanced, successfully
withstanding a minimum 100 h salt spray test. Studies have shown that
with a deep case (90 h cycle on gas nitriding), low-carbon steels exhibited
resistance to 20% salt spray solution up to 150 h.
Can Stainless Steels Be Nitrided?

In general, most stainless steels can be nitrided but with some adverse
effects on corrosion resistance. Hardness values are generally in the range
of 1000 HV or more, depending on the nitriding method. With ion nitrid-
ing and control of the nitriding potential, some hardness values greater
than 1400 HV have been achieved.
Austenitic stainless steels are perhaps the most difficult to nitride. The
following types have been successfully nitrided using gas, plasma, and
salts: 301, 302, 303, 304, 305, 309, 310, 316, 321, and 347. However,
when nitriding these steels at conventional nitriding temperatures, corro-
sion resistance is seriously impaired, in some cases up to 1000%.
The material should be in the annealed condition, which will reduce the
risk of blistering or flaking. The oxide film must be removed prior to
nitriding, or it will form a barrier that will be difficult to decompose.
Removal can be accomplished by wet or vapor blast pickling solution,
molten salts, or sputter cleaning. One technique is to use a chloride-based
solution as a cleaning agent. Once depassivated, the surface should not be
touched by hand; deposition of body oil from fingerprints will inhibit the
nitriding effect at the point of contact.
Remember, if the core is annealed, the surface can accommodate only
abrasion resistance. The core will be too soft to support any kind of con-
stant or cyclical load on the surface. If a load is applied, the case likely
will collapse, crack, and flake.
Martensitic stainless steels will nitride without exception. This group
can be preheat treated to give a supportive core for abrasive and impact
applications, as well as torque loads.
Once again, remember that the surface must be depassivated before
nitriding. Corrosion resistance will be adversely affected but may be 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 considerations
apply 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), with
time 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 for
alloy steels or tool steels.
Example 1: Plasma Nitriding of AISI Type 422 Stainless Steel

Surface modification of type 422 stainless steel was accomplished
using the pulsed plasma nitriding process. The workpiece design called
for an operational service temperature up to 645 °C (1200 °F) and the 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 0.25
Manganese 1.00
Phosphorus and sulfur 0.025 max
Silicon 0.7
Chromium 12
Nickel 0.8
Molybdenum 1.10
Vanadium 0.25
Tungsten 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 martensite
that had been hardened and tempered to 340 to 360 HV (~35 to 37 HRC).
This was accomplished via the following heat treatment procedure:
Process 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 temperature
Quench Air cooled
Stabilization 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 actual

results 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 h
Sputter clean voltage 650 V
Sputter retort pressure 100 Pa (0.015 psi)
Process temperature 525 °C (975 °F)
Cycle time 16 h at process temperature
Operating pressure 300 Pa (0.05 psi)
Operating voltage 500 V
Cooling Partial-pressure N2
Summary Results. The microhardness surveys of four cycles of pulsed

plasma nitriding each exhibited similar results with negligible deviations.
Previous gas nitriding results had exhibited a surface hardness of 852 HV,
which just made the minimum specification requirement. In addition, the 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 gas
nitriding process, due to a slight roughening of the surface. This necessi-
tated two further grinding operations. The deterioration in surface finish
after pulsed plasma ion nitriding was 13.3%. This was within the accepted
surface finish requirement, requiring no further machining. Thus, in some
instances it is not necessary to grind after plasma nitriding, and the work-
piece can be used directly.
The part was thermally sensitive to distortion. Precise measurements
were taken prior to pulsed plasma nitriding and the growth/distortion was
determined to be within the dimensional tolerance levels. The computer-
controlled system regulated not only temperature levels, but also current
density, process pressure, gas flows, gas ratios, and power levels.
Example 2: Nitriding of AISI Type 440A Stainless Steel

A martensitic type 440A stainless steel specimen was nitrided using
the pulsed plasma ion process in order to establish the minimum nitriding
temperature, bearing in mind that high as-quenched hardness values for
Fig. 3 Comparative hardness of plasma nitrided versus gas nitrided type 422
stainless steel. Courtesy of Seco/Warwick Corporation
martensitic stainless steels begin to decrease at temperatures as low as

150 °C (300 °F).
Typical composition of type 440A is:
Element Wt%
Carbon 0.7
Manganese 0.85
Phosphorus 0.040 max
Sulfur 0.030 max
Silicon 0.9
Molybdenum 0.75
Chromium 17
The preheat treatment of the steel to achieve the core properties and
hardness 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 Oil
Stabilize Cryogenic treatment using liquid N2 for 20 min
Temper 190 °C (375 °F)
The nitriding specification was:
Characteristic Required value

Case depth 0.0375 mm (0.0015 in.) max
Case hardness 800 HV (~64 HRC)
Compound zone 0.0025 mm (0.0001 in.) max γ′
The process parameters were:
Process Parameter
Sputter time 1h
Sputter voltage 650 V
Cycle time 16 h
Cooling Partial pressure with clean, dry N2
Summary Results. The process temperature of 190 °C (375 °F) was too
low for effective nitrogen diffusion. Diffusion did occur, but not suffi-
ciently to form a commercially usable case. The tempering curve was fur-
ther examined for a comparable martensitic grade containing a slightly
higher carbon content (AISI type 440C with 1.10% C) (Fig. 4). Based on
Fig. 4 Tempering curve for type 440C stainless steel. Composition: 1.02 C,
0.48 Mn, 0.017 P, 0.011 S, 0.18 Si, 0.54 Ni, 16.90 Cr, 0.64 Mo. Heat
treated at 1040 °C (1905 °F), 2 h. Oil quenched from 66 to 94 °C (150 to 200 °F).
Double stress relieved at 175 °C (345 °F), 15 min. Water quenched. Tempered 2 h.
Heat treated, 9.78 mm (0.385 in.) round. Tested, 9.53 mm (0.375 in.) round. At
260 to 540 °C (500 to 1000 °F). Also, heat treated, 14 mm (0.550 in.) round.
Tested, 12.8 mm (0.505 in.) round. At 295 to 760 °C (1100 to 1400 °F). Source:
Republic Steel
the tempering curve, it was concluded that a process temperature of 370 °C

(700 °F) could be used. Temperatures above 370 °C (700 °F) cause rapid
deterioration of the corrosion resistance of the 440A stainless steel.
On that basis, a further cycle was made at a process temperature of 370 °C
(700 °F) with the following process parameters:
Process Parameter
Sputter time 1h
Cycle time 16 h
Cooling 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 at

temperature, nitrogen diffusion will take place and form stable nitrides,
but with a very shallow case depth. A higher process temperature causes
serious deterioration of the surface corrosion characteristics. The corro-
sion resistance of another martensitic stainless steel is seen in Fig. 5, with
and without nitriding, and compared to a high strength 4140 steel.
Fig. 5 Reduction in corrosion resistance after nitriding of type 422 stainless

steel
Example 3: Nitriding of AISI Type 630 (17-4 PH) Stainless Steel

Type 630 stainless steel is widely used in the aircraft and aerospace indus-
try for gear manufacture and other critical performance items. It is known as
a precipitation hardening steel, providing good tensile strength and impact
values when heat treated by solutionizing and precipitation treatments.
Pulsed plasma ion nitriding technology was considered for this steel
because of its ability to provide controlled, repeatable metallurgy without
affecting the core. Ion nitrided type 630 also maintains good corrosion
resistance.
Typical composition of type 630 is:
Element Wt%
Carbon 0.07 max
Manganese 0.89
Silicon 1.00
Nickel 3.8
Copper 3.9
Niobium 0.3
Tantalum 0.4
Chromium 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 Oil
Precipitation 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.) max

Core hardness required 830 HV (65 HRC) min
Compound zone required 0.005 mm (0.0002 in.) max
The plasma nitride process parameters were:
Process Parameter
Sputter time 1h
(continued)
Process Parameter
Cycle time 4 h (at temperature)
Cooling Partial pressure using
clean dry nitrogen gas
Summary Results. The microhardness survey (Fig. 6) showed a surface

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

The following temperature factor values are based on the Harris for-
mula (Ref 5):
Case depth = Square root of time × Temperature factor
Fig. 6 Microhardness of AISI 630 (17-4 PH) stainless steel after pulsed plasma
ion nitriding. Source: Ref 4
Suggested temperature factors for the process cycle time in relation to the
required case depth are:
Temperature
°C °F Temperature Factor
460 865 0.00221
470 875 0.00233
475 885 0.00259
480 900 0.00289
500 930 0.0030
510 950 0.0033
515 960 0.0035
525 975 0.0037
540 1000 0.0038
The factors are based on the steel at the selected process time. They do
not pertain to nitriding of the higher alloyed steels such as stainless
steels. The diffusion rate of nitrogen into the steel surface will reduce
dramatically as the alloy content increases. The factors also do not con-
sider the furnace loading and load density, and positioning in relation to
potential 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 the
cycle time for a particular load. The process technician should keep a
record of:
• Load surface area

• Load mass
• Selected process temperature
• Plasma power conditions
• Process pressure
• Process gas flows
• Grade or grades of steel being processed
• Case depth achieved
REFERENCES
1. D. Pye, Nitriding Techniques and Methods, Steel Heat Treatment
Handbook, G.E. Totten and M.A.H. Howes, Ed., Marcel Dekker, Inc.,
1997, p 721–764
2. A. Fry, The Nitriding Process, ASST Nitriding Symposium, 1929,
reprinted in Source Book on Nitriding, American Society for Metals,
1977, p 99–106
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 Surface
Modification of Stainless Steels AISI 422, 440A and 630, Surface
Modification Technologies VI, T.S. Sudarshan and J.F. Braza, Ed.,
Minerals, Metals & Materials Society, 1993, p 195–216
5. F.E. Harris, Case Depth, Metals Progress, Vol 44, Aug 1944, p 265

CHAPTER 13
Control of the
Process Gas in
Plasma Conditions
PROCESS GAS CONTROL for plasma (ion) nitriding is a matter of

estimating the flows necessary to accomplish the required surface metal-
lurgy. Conventional gas nitriding systems measure the dissociation of the
ammonia process gas. This is not quite so simple with plasma systems
because the process is operating under partial pressure conditions, making
it difficult to introduce an effective sampling system into the unit. 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 understanding
process gas control in plasma nitriding and its influence on compound
zone formation. Emphasis is placed on the effect of sputtering on the
kinetics of compound zone formation. Additional information on gas
ratios and gas flow can be found in Chapters 8 and 9.
Analysis by Photo Spectrometry

A significant development in process gas control analysis can be cred-
ited to N. Ryzhov of Moscow State University, who developed what
appeared to be a workable control system of the gas species activity of the
plasma glow seam (Ref 1). The work, which has not yet been commercial-
ized, is based on the solubility limit of nitrogen in iron and the amount of
atomic nitrogen available for diffusion. The gas is evaluated by line-of-
sight through an observation port in the furnace process chamber.
The system uses photo spectrometry as the principle of operation. The
observation unit, sighted onto the plasma glow seam, develops an electrical
signal that is then transmitted to a personal computer/programmable logic
control (PC/PLC) combination (Fig. 1). This controls the gas-delivery system
Fig. 1 Ion nitriding furnace incorporating process gas control analysis system.
(a) Furnace layout. (b) Schematic of equipment layout. Source: Ref 1
to the process chamber to manipulate the process gases around the required
gas ratios and required surface metallurgy.
Analysis by Mass Spectrometry

Szabo and Wihelmi discussed mass spectrometric diagnosis of the sur-
face nitriding mechanism in a direct-current (dc) glow discharge (Ref 2).
Their hypothesis was that it makes no difference whether the nitrogen
source is ammonia, or nitrogen and hydrogen. Use of hydrogen, they
stated, is an important function in the nitriding process. The hydrogen acts
Chapter 13: Control of the Process Gas in Plasma Conditions / 141
as a reducing gas to reduce (with heat) surface oxides on the steel and,
perhaps more importantly, to influence and regulate the composition of
the compound zone (white layer) as shown in Fig. 2.
They also investigated the ability of alloying elements to form nitrides
in the steel surface, concluding that the following elements will readily
form nitrides (listed in increasing order of ease of nitride formation):
• Iron
• Manganese
• Silicon
• Tungsten
• Molybdenum
• Chromium
• Vanadium
• Titanium
• Aluminum
To observe the reactions of the process gases in the nitriding chamber,

they connected a mass spectrometer. They determined that they could 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 not
only observed, but also controlled.
Difficulties Associated with Gas Analysis

Because the plasma nitriding process is under vacuum conditions, it is
difficult to evaluate accurately the gaseous species activities within the ion-
ized gas glow seam and, more importantly, the amount of nitrogen diffusion
into the steel surface in relation to the solubility limits of nitrogen in iron.
The Russian method of glow seam observation and analysis of the gaseous
activities seemed to hold the most promise as a method of accurately con-
trolling not only the solubility limit of nitrogen in iron, but also the solubility
limit of carbon in austenite.
Control of the gas activities has eluded scientists and metallurgists in
the field of pulsed plasma ion nitriding. Using sensors to measure gas
Fig. 2 Commencement of nitride formation on a steel surface. Note: The

hydrogen now acts as a reducing agent. Source: Ref 3
dissociation during gas nitriding or derivatives of that technique remains

the 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 of
chromium-molybdenum-vanadium steels were carried out by Roliński and
Sharp (Ref 4). Their work was performed at 540 °C (1000 °F) in a mixture
of 30% nitrogen and 70% hydrogen. They found that the process could be
described by a half-order polynomial equation and used TableCurve 2D
software (SPSS Science, Inc.) to determine the effect of sputtering rate on
compound layer growth and composition. A complete account of the work
of Roliński 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. Roliński and G. Sharp
The role of sputtering in plasma nitriding has been a subject of many

studies in the last 35 years (Ref 5–19). A qualitative approach to analysis of
the sputtering rate (SR) in plasma nitriding has been proposed by Keller
(Ref 6), who concluded that the parabolic growth of the compound zone
(zone containing a mixture of Fe4N and Fe2–3N nitrides) is affected by a lin-
ear removal rate of the surface atoms due to sputtering. The first numerical
analyses of these processes were done by Marciniak (Ref 13, 14) and Sun
and Bell (Ref 19). The calculated and experimentally verified SRs for 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 of
the compound layer growth for a specific steel, the specific nitriding condi-
tions, and the assumed SR. The assumed values of SR were between 0.1
and 0.5 µm/h (Ref 19).
If the assumptions of Keller (Ref 6) and Marciniak (Ref 14) are correct,
then the kinetic of the compound zone growth y in plasma nitriding can be
described by the half-order polynomial equation:
y = a + bx + c√x (Eq 1)
where a is compound zone thickness formed during the ramp-up time, b is

SR, c is a coefficient of compound zone growth due to diffusion, and x is
nitriding time. The experimental results of Edenhofer (Ref 8) for the
kinetics of the compound zone formation on En 9 (AISI 4142) steel
plasma nitrided at 450, 530, and 570 °C (840, 985, and 1060 °F) were fit-
ted to Eq 1 using the TableCurve 2D software (Ref 20) and are presented
in Fig. 3. The fit of the data for the available 45 h range is very good, since
the coefficient of determination, r 2, is high: 0.993, 0.995, and 0.962,
respectively. The calculated SR is 0.057 µm/h for 450 °C (840 °F), 0.124
µm/h for 530 °C (985 °F), and 0.427 µm/h for 570 °C (1060 °F). The
graph extrapolated to 100 h of nitriding time shows the tendency of the
curves: the maximum, respectively, at approximately 65, 50, and 35 h of
nitriding and the diminishing values thereafter. At the same time, the
growth due to diffusion (the coefficient c) is larger for higher temperatures
and smaller for lower nitriding temperatures.
Similar graphs based on the experimental data of Marciniak (Ref 14) for
36 H3M (0.36C-3Cr-0.6Mo-0.6Mn-0.27Si) steel nitrided at 530 °C (985 °F)
Fig. 3 Compound zone thickness versus nitriding time for 42Cr Mo4 (AISI
4142 steel) plasma nitrided 3.3 mbar in the atmosphere of 25% nitro-
gen + 75% hydrogen at 570 °C (1060 °F) (upper curve), 530 °C (985 °F) (middle
curve), and 450 °C (840 °F) (bottom curve) based on the experimental data of
Edenhofer (Ref 8, 24). The graph is extrapolated over the original limit of 45 h. The
fit equations are y = 3,8722 – 0.236x + 2.361√x, y = –0.789 – 0.124x + 1.958 √x,
and y = 0.758 – 0.057x + 0.918√x and r 2 = 0.993, 0.995, and 0.962, respectively.
are presented in Fig. 4. The curve with a well-distinguished maximum at

about 16 h drawn for nitriding with 50% nitrogen and 50% hydrogen 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 the
kinetic equation, which represents a nitriding potential, is higher for the
samples nitrided with 50% nitrogen and 50% hydrogen than for samples
nitrided with 15% nitrogen and 85% hydrogen. However, all of the above
predictions for the range of time exceeding the experimental data range may
not be accurate since extreme caution is advised in relying on polynomials
for extrapolations and foreasts beyond the range of the dataset (Ref 20).
The diminishing value of the compound zone thickness during a long
nitriding time may have important practical meaning; the compound zone
could eventually disappear completely and therefore the nitriding rate
would drop down and/or a denitriding of the steel could take place (Ref 9).
Wells and Strydom suggested that the compound zone growth might also
be affected by redeposition of the sputtered material (Ref 16). This phe-
nomenon could be enhanced by oxygen, which may always be present in a
small quantity in industrial systems, and, therefore, the compact portion of
the compound zone can be significantly reduced by formation of the
oxynitride (Ref 16).
Consequently, in our studies, we researched the actual kinetics of com-
pound zone formation in long nitriding with an aim to establish the SR in
such a cycle. The 3% Cr-Mo-V steel we investigated is used in the gear
industry to achieve an exceptionally deep case. It was then very important
Fig. 4 Compound zone thickness versus nitriding time for 36H3M 3% Cr-Mo
steel plasma nitrided at 530 °C (985 °F) in the atmosphere of 50%
nitrogen + 50% hydrogen (upper curve) and 15% nitrogen + 85% hydrogen (bot-
tom curve) based on the experimental data of Marciniak (Ref 14). The graph is
extrapolated over original limits of 16 and 36 h, respectively. The fit equations are
y = 1.275 – 0.425x + 3.295√x, y = 0.5 – 0.036x + 0.479√x, and r 2 = 0.994 and
0.986, respectively.
to verify if the thickness of the compound zone would become thinner to

the point of disappearing completely with prolonged nitriding.
Experimental Parameters
Material and Processing. A quenched and tempered 3% Cr-Mo-V
(DIN 39CrMo V13.9) nitriding steel was used for these studies. The
Brinell hardness was 321 to 363. The test samples were 25 × 25 × 150 mm
(1 × 1 × 6 in.) bars with a ground surface finish of Ra 1.6 µm or better. All
samples were blasted with 180 grit aluminum oxide before nitriding. The
equipment and processing details are described elsewhere (Ref 21). The
nitriding was carried out at 540 °C (1000 °F) nominal temperature, and
the nitriding times were from 4 to 400 h. The samples were treated in a
direct current (dc) plasma in the atmosphere of 30% nitrogen and 70%
hydrogen and a pressure of 3.2 mbar. The ramp-up time was 4 h.
Testing Procedure. The samples were cut in half and prepared for 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 a
single measurement was ±0.1 µm. There was a minimum of five samples
used in each run. The x-ray diffraction phase analysis was performed
using Cr Kα radiation.
Results and Discussion

The compound zone thickness changes with the nitriding time, as
shown in Fig. 5. A possible maximum value of approximately 13.5 µm is
Fig. 5 Compound zone thickness versus nitriding time for 3% Cr-Mo-V steel
plasma nitrided at 540 °C (1000 °F). The fit equation is y = 6.158 –
0.0294x + 0.933√x and r2 = 0.952. Confidence and prediction intervals represent
normal distribution and standard error (small interval) at 95%.
achieved after about 250 h of nitriding, and the final value after 400 h is
about 13 µm. Micrographs of the compound zone are presented in Fig. 6.
They do not reveal any significant presence of the vapor-deposited, possi-
ble oxygen-contaminated layer, as was suggested by Wells and Strydom
(Ref 16). The top portion of the compound zone contains some porosity
and probably a layer of the nitrides deposited from plasma; however, its
main fraction stays very compact.
Optical microscopy of the microsections also revealed a presence of fre-
quent conical structures at the surface. Similar features were also observed
by others (Ref 22). The intensity of these peaks was higher on samples
nitrided longer. The x-ray diffraction showed that initially after 4 h of
Fig. 6 Optical micrographs of surface layers produced on 3% Cr-Mo-V steel

by plasma nitriding at 540 °C (1000 °F) for (a) 4 h, (b) 25 h, (c) 144 h,
(d) 289 h, and (e) 400 h. Bright field, etched with 2% nital
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 25
h of nitriding, only two weak diffraction peaks from the (100) and (101)
planes could be detected. The 400 h nitriding produced a single-phase γ′
compound zone on the steel surface. It is very likely that the ε-nitride 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 the
steel during nitriding (Ref 9, 17, 23, 24). In the short nitriding processes,
carbon atoms were diffused toward the surface and, by reacting with iron
and nitrogen, helped in the formation of the ε-phase. During the long expo-
sure to the plasma, the surface was decarburized as carbon was sputtered
away and replaced by nitrogen. With a lack of carbon atoms, the nitriding
potential of the plasma was shifted toward γ′; the lower nitrogen phase and
the ε-phase disappeared completely.
The results of the compound zone thickness versus nitriding time stud-
ies clearly demonstrate the effect of sputtering: a possible maximum value
at 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 of
nitrogen and sputtering. Since sputtering was taken into account only
from a time when the final temperature was reached, the a value was
affected by an error of not counting the material removed. In fact, this
value is only about 0.06 µm if we assume that SR was the same during
ramping and the final soaking.
Fig. 7 X-ray diffraction patterns of surface layers produced on 3% Cr-Mo-V

steel plasma nitrided at 540 °C (1000 °F) for (a) 4 h, (b) 25 h, (c) 144 h,
(d) 289 h, and (e) 400 h. Cr Kα radiation
The SR of 0.0295 µm/h from this experiment agrees very well with the
value of 0.036 µm/h calculated by us for the literature data of nitriding
performed in the atmosphere of 15% nitrogen and 85% hydrogen on simi-
lar steel (Ref 14). However, our kinetic studies as well as the studies based
on the results of others (Ref 8, 9, 14, 15) could be affected by not taking
into account the fact that a phase composition of the steel surface may
change during a long nitriding process. In the plasma nitriding of the 3%
Cr-Mo-V steel, the equilibrium at the surface was not achieved very 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 on
pressure, temperature, gas composition, and the plasma power density, but
mainly on the phase composition of the compound zone that formed on
the surface. Some of these parameters were probably different in our
experiment than in the experiments carried out by the other researchers
(Ref 8, 14, 19, 24, 25); however, as long as we consider γ′ compound zone
formation, its SR stayed low.
In comparatively short nitriding cycles with a sufficiently high nitro-
gen content carried out by others (Ref 8, 14), a mixture of the γ ′ and ε
was produced and, consequently, the SR was much higher than when the
“pure” γ ′-phase was produced. This can be seen in Fig. 8, which repre-
sents the nitriding kinetics for shorter (up to 25 h) cycles. The graph is
not biased by the phase composition change toward a pure γ ′; the SR is
about 0.049 µm/h and the coefficient c about 1.221. If “sputtering-free”
plasma nitriding is hypothetically assumed, Eq 1 can be used with the
coefficient b equal to zero to see the effect of the remaining (diffusion)
fraction of the equation on the kinetic.
Fig. 8 Compound zone thickness vs. nitriding time for 3% Cr-Mo-V steel
plasma nitrided at 540 °C (1000 °F). This is a modified form of Fig. 5
from which the data for a “pure” γ′ compound zone were removed (144, 289,
and 400 h). The fit equation is y = 5.438 – 0.049x + 1.221 √x and r 2 = 0.958.
Confidence and prediction intervals represent normal distribution and standard
error (small interval) at 95%.
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 nitriding
of 3% Cr-Mo-V steel and the gas nitriding of Armco iron are very similar;
they achieve a final value of 25 to 30 µm after 400 h of nitriding. The final
value for the plasma nitriding curve of 36H3M steel is about 10 µm.
Regardless of the sputtering effect, the plasma nitriding process in the
atmosphere of 30% nitrogen and 70% hydrogen can be considered a low
nitriding potential process.
In a plasma process, activation of the cathode due to ion bombardment
from the atmosphere containing sufficiently high nitrogen is very effec-
tive, and, therefore, a full coverage of the surface with the compound
zone after only a few minutes of nitriding is achieved (Ref 19). This was
also evidenced in our experiments by the fact that the curve did not start
at the beginning of the coordinate, but earlier (Fig. 5). In a low-potential
gas nitriding process, the γ ′-phase nucleates on ferrite extremely slowly
and only after substantial time becomes a continuous, compact layer
(Ref 26).
The experiments showed also that the equilibrium between plasma and
the steel surface was not achieved quickly when the atmosphere of 30%
nitrogen and 70% hydrogen was used. Instead, an initially formed mix-
ture of γ ′- and ε-phases was slowly converted into a single γ ′-phase
structure. The phase composition changes resulted in a reduction of the
SR, as well as a reduction in the diffusional growth of the compound
zone. Sputtering of the surface in plasma nitriding has then an additional
Fig. 9 Comparison of a hypothetical, “sputtering-free” kinetic of compound

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

tial and enhances the ability to reduce the thickness of the compound
zone.
Conclusions
The experimental data of the compound zone formation taken from the
literature were analyzed and presented in graphic forms. It was found by
using the TableCurve 2D software that the SR for the γ′ (0.036 µm/h) was
smaller than for the ε-type compound zone (up to 0.425 µm/h). At the
same time, it was confirmed that the growth of the compound zone due to
the diffusion of nitrogen was also slower for the γ′-type, which could be
expected (Ref 26).
The graphs extrapolated over the experimental data range showed that
the kinetic curves may achieve a maximum value and that the sputtering
may cause a complete disappearance of the compound zone after long
nitriding. These types of kinetic characteristics were more likely for the
ε-type compound zone than for γ′.
The experiments carried out on 3% Cr-Mo-V steel did not show any
disappearance of the compound zone, and its final value after 400 h of
nitriding was still about 13 µm. The SR calculated from our experiment
was about 0.03 µm/h. This agrees well with the value calculated for
kinetic data taken from the literature for a similar steel for presumably the
γ′-type nitride (Ref 14). The analysis limited to cycles not exceeding 25 h
resulted in a higher rate of sputtering as well as a faster diffusional
growth. This fact can be attributed to the presence of the ε-phase in the
compound zone.
Based on our research, it seems to be likely that the γ′ compound zone
can disappear completely because of sputtering after an extremely long
nitriding time. Instead, it is more likely that it will become more porous
and the specific surface area will be greater. However, this will still need
to be proved by additional experimental work.
ACKNOWLEDGMENT
The Appendix is reprinted (with minor changes) from E. Roliński 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 to
448 (reproduced by permission of ASM International).
REFERENCES
1. D. Pye, A Review of the Gas Species Activity and Control of Pulsed
Plasma Technology during the Nitriding, Carburizing, and 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 Compound
Layer), Härt. Tech. Mitt., Vol 47 (No. 3), May-June 1992, p 169–174
4. E. Roliński and G. Sharp, Effect of Sputtering on Kinetics of Com-
pound Zone Formation in the Plasma Nitriding of 3% Cr-Mo-V
Steel, J. Mater. Eng. Perform., Vol 10 (No. 4), Aug 2001, p 444–448
5. H. Knüppel, K. Brotzman, and F. Elserhard, Stahl Eisen, 1958, Vol 75
(No. 26), p 1871
6. K. Keller, Schichtaufbau Glimmnitrierter Eisenwerkstoffe, Härt.
Tech. Mitt., Vol 26, 1971, p 120 (in German)
7. M. Hudis, J. Appl. Phys., Vol 44, 1973, p 1489
8. B. Edenhofer, Heat Treatment ’76, Proc. 16th International Heat
Treatment Conf., Stratford-upon-Avon, 6–7 May 1976, The Metals
Society, 1976, p 7
9. B. Edenhofer, Proc. Heat Treatment ’79, Birmingham, 22–24 May
1979, TMS/ASM, 1979, p 52
10. G.G. Tibbets, J. Appl. Phys., Vol 44, 1974, p 5072
11. E. Roliński, Ph.D. dissertation, Warsaw University of Technology,
1978 (in Polish)
12. T. Karpinski and E. Roliński, Proc. 8th National Conf. Heat Treat-
ment, Bratislava, Slovakia, 1978, Dom Techniky ČSVTS, p 27 (in
German)
13. A. Marciniak, “Processes of the Cathode Heating and Nitriding
under a Glow Discharge Condition,” Ph.D. dissertation, Warsaw
University of Technology, 1983 (in Polish)
14. A. Marciniak, Surf. Eng., Vol 1, 1985, p 283
15. A. Wells and I. Le R. Strydom, Surf. Eng., Vol 2, 1986, p 283
16. A. Wells and I. Le R. Strydom, Surf. Eng., Vol 4, 1988, p 55
17. T. Lampe, S. Eisenberg, and G. Laudien, Surf. Eng., Vol 9, 1993, p 69
18. H. Michel, T. Czerwiec, M. Gantois, D. Ablitzer, and A. Ricard,
Surf. Coating Technol., Vol 72, 1995, p 103
19. Y. Sun and T. Bell, Mater. Sci. Eng., Vol A224, 1997, p 33
20. TableCurve®2D, Version 4, SPSS Inc., 1998
21. E. Roliński, F. LeClaire, D. Clubine, G. Sharp, D. Boyer, and R. Not-
man, J. Mater. Eng. Performance, Vol 9, 2000, p 457
22. Y. Sun, N. Lou, and T. Bell, Surf. Eng., Vol 10, 1994, p 279
23. J. Zysk, Metaloznastwo I Obrobka Cieplna, No. 6, 1973 (in Polish)
24. B. Edenhofer, Ibsen Industries International GmbH, private commu-
nication, Aug 2000
25. J.G. Conybear and B. Edenhofer, Proc. 6th Int. Conf. Heat Treat-
ment of Materials, Chicago, IL, 28–30 Sept 1988
26. L. Maldzinski, W. Liliental, G. Tymowski, and J. Tacikowski, “New
Possibilities of Controlling the Gas Nitriding Process by Utilizing
Simulation of Growth Kinetics of Nitride Layers,” presented at the
18th ASM Conf., Rosemont, IL, Oct 1998

CHAPTER 14
Processing with Nitriding
THE NITRIDING PROCESS can be applied to various materials and

part geometries. This chapter focuses on tool steels, pure irons, low-alloy
steels, and maraging steels.
Hot-Work Tool Steels

The hot-work group of tool steels is usually considered to be the AISI
H-series and includes chromium-base, tungsten-base, and 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, or
ion nitriding.
Forging Dies
Selection of a hot-work steel grade depends on the forge die applica-
tion. For many forging steel applications, the steel of choice is H13, which
is classified as a deep-hardening chromium hot-work steel containing 5%
Cr and 0.40% C. This steel can be readily water cooled while in service
and has a good toughness factor after nitriding—provided that the preheat
treatment has been done correctly in terms of core hardness for case 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 is
placed on the steel component (Fig. 1).
Core hardness generally is determined by the hardness at which the
steel can be cut, rather than the best hardness for supporting the case. The
appropriate hardness for case support is around 44 to 47 HRC, which pro-
duces a tough, springlike condition. This core hardness will allow some
flexibility in the die without taking a permanent set for deformation.
Another advantage of the nitrided case is that it will withstand high-
temperature operating conditions with no significant loss of surface hard-
ness. Nitrided hot-work tool steels are unlike carburized steels that rely on
the diffusion of carbon and then a phase transformation to martensite,
Fig. 1 Core support of the nitrided case on a forging die. Source: Ref 1
which necessitates a quench (and results in distortion and the possibility

of mixed phase conditions).
If a mixed phase is present, the nitriding process will decompose it. It
must be expected that with the decomposition of the mixed phase, some
size change will occur due to transformation of the retained austenite to
martensite.
Other considerations are the surface metallurgy requirements of the die,
including case depth, compound layer formation, and temperature. Each
factor is discussed later in this chapter.
Case Depth. A deep case is unnecessary. With a deep case formation,
the surface will begin to lose its flexibility, no matter how well the preheat
treatment has been conducted. Inflexibility leads to surface cracking, and
thus to press downtime for die repair.
Compound Layer Formation. The thickness and phase construction
of the compound layer significantly influence die performance. If a thick
case is produced, there is a strong likelihood that a thick, inflexible com-
pound layer will be produced on the die surface. Therefore, the cycle must
be run with a thin diffused nitrided case with a thin compound surface
layer. Recommended case depth is a maximum of approximately 0.25 mm
(0.010 in.). The thinner the diffused case, the thinner the compound layer.
With gas and salt bath nitriding methods, the thickness of the compound
layer can be expected to be roughly 10% of the total measured case depth.
However, these values will change with both controlled nitriding (dilu-
tion) and particularly with ion nitriding. The latter two methods offer a
greater degree of surface metallurgy control.
Temperature. Process temperature selection plays a significant role in
the thickness of the diffused case and formed compound layer. If higher
process temperatures are selected for gas or salt bath nitriding, there is an
inherent danger of nitride networking on corners. This is a very brittle
phase condition, and care should be taken to minimize its potential.
Process temperature traditionally has been selected based on the temper-
Chapter 14: Processing with Nitriding / 155
ing temperature of the steel, without considering the potential for nitride
networking at higher temperatures. If the surface metallurgy and the
nitride potential can be controlled, the possibility of nitride networking
can be greatly reduced.
Aluminum Extrusion Dies

Aluminum extrusion dies typically are manufactured using the hot-
work steel H13 described earlier. The main area of concern with extrusion
dies is the bearing face area (Fig. 2).
The mechanics of operation are based on the press load factor on the
aluminum billet to be extruded. The aluminum billet is usually preheated
to around 425 °C (800 °F). At room temperatures, the surface of the alu-
minum billet will oxidize to form aluminum oxide. Aluminum oxide is
extremely abrasive and will begin to abrade and wear the bearing surface
of the extrusion die. The die aperture will become larger and out of toler-
ance, resulting in a costly shutdown of the press. Typical extrusion press
configurations are shown in Fig. 3.
Nitriding—gas, salt bath, or fluidized bed—enhances the hardness of the
die bearing surface and reduces wear. Again, a deep case is not required; a
shallow case will suffice up to a maximum of 0.25 mm (0.010 in.), with
formation of a compound layer in the region of 10% of the total case depth.
The compound layer will wear off as the extruded aluminum is pushed
over the bearing surface. Aluminum oxide formation becomes much more
aggressive due to the frictional forces now being developed as the hot
Fig. 2 Schematic cross section of an aluminum extrusion die made from H13
steel showing the bearing (wear) surface and a core with hardness of
38 to 44 HRC. Source: Ref 1
Fig. 3 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, hydraulic
power unit; 2, tie rods; 3, butt shear; 4, extrusion platen; 5, container shifting
cylinders; 6, swiveling operator’s console; 7, die slide; 8, container; 9, container
housing; 10, billet loader; 11, press base; 12, billet loader cylinders; 13, pressing
stem; 14, crosshead; 15, side cylinders; 16, cylinder platen; 17, main cylinder
aluminum is pushed through the die (Fig. 4). Thus, the wear factor becomes
even greater.
Ideal die surface metallurgy will reduce formation of the compound
layer on the surface. This can be accomplished by controlling the process
gas using the controlled nitride method (dilution) or ion (plasma) nitrid-
ing. The two-stage process, which uses a higher process temperature of
565 °C (1050 °F), is not recommended because of the possible risk of
nitride networking.
High-Speed Steel Cutters

Enhancing the surface hardness of high-speed steel cutters via the
nitriding process offers many advantages. However, once again, a deep
case is not required. A short cycle is necessary only to diffuse a case of no
greater than 0.038 mm (0.0015 in.). The formed case will be in the region
Fig. 4 Bearing surface of an aluminum extrusion die, demonstrating the wear

process due to hot aluminum extrusion. (a) Untreated die. (b) Die with
nitrided surface
of 1100 HV or harder and will be supported by a substrate material hard-

ness in the region of 850 HV. Remember, the nitride processing will give
the high-speed steel a further temper, producing a dimensional stabilizing
effect. There will be a very slight, but insignificant, increase in size due to
the diffusion.
Another procedure is to follow the nitriding with a thin-film deposition
using the plasma-assisted deposition technique. The cutter is nitrided, and
then the nitrided surface is immediately coated with titanium nitride. Two
process techniques are accomplished during the same furnace operation.
Gears
Nitriding is finding greater acceptance as a surface hardening method
for precision manufactured gears, although it requires careful steel selec-
tion. The steel selected must not be a high-carbon grade; otherwise, there
is a risk of forming ε as the dominant phase in a dual-phase compound
zone. This could lead to brittle fracture of the compound layer, depositing
fine pieces of fractured steel from the gear surface that will usually lodge
between the meshing teeth, thus causing further deterioration of the gear
pressure face.
Furthermore, the steel must not have a high aluminum content. 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, pressure
must be controlled to ensure that the plasma glow seam is uniformly posi-
tioned over the entire gear tooth surface. This is necessary for uniform
case depth on the pressure face and tooth root (Fig. 5).
As described below, surface metallurgy—in terms of growth, compound
layer formation, and case uniformity—is critical to gear performance.
Growth. By selecting a process temperature around 485 to 500 °C (900
to 925 °F) and then controlling it, the compound zone will not form as thickly
Fig. 5 Illustration of ion nitriding pressure that is too low (toward high
vacuum), resulting in no nitriding at the tooth root. Source: Ref 1
as at higher temperatures and the risk of nitride networking will be reduced,

along with the risk of edge or corner chipping. Remember, the solubility
limit of nitrogen in iron increases as temperature increases. The solubility
limit using a process temperature of approximately 485 °C (900 °F) is
approximately 6 to 7%. If compound layer thickness is reduced, growth
also is reduced.
If surface growth is reduced, less stock will have to be removed by lap-
ping or grinding operations. In fact, the gear pressure face could be
machined slightly undersize and “grown” into size, thus reducing expen-
sive machining time. Selection of a lower process temperature thus offers
important advantages to the gear manufacturer.
Gear teeth must be fully deburred before ion nitriding. Otherwise,
localized hot spots will likely occur at burrs. When this happens, the gear
tooth will appear dark around the area of the burr.
Compound Layer Formation. Process temperature affects formation of
the γ′- and ε-phases within the compound zone. A lower process tempera-
ture tends to reduce formation of the ε-phase, whereas higher process tem-
peratures encourage its formation—particularly if the steel has a higher
carbon level. A greater presence of ε-phase means that the immediate gear
surface will have superb wear resistance properties but no impact value.
The pressure face will begin to crack and chip with high impact loading.
If the gear has been previously tempered at 510 to 540 °C (950 to
1000 °F) and the core hardness is temperature sensitive, the core hard-
ness may be reduced by up to 2 to 4 Rockwell points due to the process
time at the nitriding temperature. This could take the gear core hardness
out of specification. Pretreatment of the gear to establish the case sup-
port core hardness is extremely important. Emphasis should be placed
on producing a core hardness that will ensure a good case support and
improve fatigue bending performance of the gear tooth rather than a
core hardness that will ease machining.
Case Uniformity. If a uniform case depth is not maintained, then the
gear tooth could fail under service loads. Nonuniform case formation 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 all
gear surfaces. With gas nitriding, it can be a serious problem due to gas
stagnation in the root of the tooth. This can be caused by inadequate gas
circulation. With ion nitriding, using too low a process pressure or failing
to adequately control the process pressure can cause a nonuniform glow
seam that does not reach the root of the gear tooth (Fig. 6).
Pure Irons
Pure iron can be successfully nitrided, even with gas and salt bath
methods. (Please note that this discussion is restricted to nitriding and not
ferritic nitrocarburizing, which will be discussed in Chapters 18 to 23.)
Fig. 6 Effect of process pressure in ion nitriding of gear teeth. Courtesy of

Plateg GmbH
The case that forms is strictly of iron nitride. However, the resulting 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 region
of 60 HRC (700 HV). This is accomplished by manipulating the process
gas ratios so that the ratio of nitrogen to hydrogen is approximately 5:1,
higher than would be expected from the decomposition of ammonia gas,
which encourages iron nitride formation at the surface. The hardness
value will be no greater than 700 HV, but corrosion resistance will signifi-
cantly improve. This technique often is used when corrosion resistance is
more important than wear resistance.
Low-Alloy Steels
Low-alloy steels that are nitrided using the gas or salt techniques have
low surface hardness values but improved torque and corrosion resistance.
The same applies to the ion nitriding technique. Increasing the nitrogen-
to-hydrogen ratio significantly raises surface hardness values (though they
will not be high) and further improves corrosion and torque resistance.
Remember, low-alloy steels (and even the cast irons) contain no alloying
elements to form significant stable nitrides in the steel surface. Therefore,
the nitriding potential of the process gas must be raised (via increased
nitrogen content) to encourage the formation of iron nitrides. This means
high nitrogen flows in relation to hydrogen during ion nitriding. Once
again, it is important to consider the solubility of nitrogen in the Fe-N
phase diagram (Ref 2) (see Chapters 1 and 3 for examples of the Fe-N
binary phase diagram).
Maraging Steels
Maraging steels can be nitrided, but the surface hardening mechanism
is very different from that of conventional nitriding. As the process tem-
perature passes through 450 °C (850 °F) on heat up, a phase change from
ferrite to austenite occurs—along with a subsequent size change. The
nitrogen diffuses into the maraging steel at that point, and the solubility
limit of nitrogen in iron is exceeded. After the cycle time is completed, the
maraging steel must be cooled quickly in the process chamber under nitro-
gen. This will produce a shallow layer on the steel surface due to the rapid
cooling of fine nitrogen-austenite products. This product is detrimental in
terms of surface mechanical properties.
Therefore, the nitriding process temperature for maraging steels should
be lower than for conventional steels. The process temperature must be
lower than the ferrite-to-austenite phase transformation temperature of
approximately 450 °C (850 °F). A process temperature of between 425
and 450 °C (800 and 850 °F) will produce a surface hardness of approxi-
mately 67 HRC (900 HV). This lower-than-normal process temperature
results in an extended case formation time (Ref 3).
The maraging steels can be successfully ion nitrided at lower tempera-
tures to avoid the phase change and subsequent growth, with better con-
trol of the hydrogen-to-nitrogen process gas ratios. The ratio generally
would be approximately 5 parts hydrogen to 1 part nitrogen or even 6
parts hydrogen to 1 part nitrogen. This means that the solubility limit of
nitrogen in iron is not reached, and thus there can be no risk of possible
corner networking.
Higher Alloyed Steels

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

CHAPTER 15
Stop-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 density
and thickness, its adhesion to steel, and part surface finish.
Methods for Selective Gas Nitriding

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

50 µm (0.002 in.) for rough finishes
• Nickel: More effective, allowing thinner coatings, up to 25 µm
(0.001 in.); also exhibits excellent resistance to nitrogen penetration
• Silver: Fairly thick deposits, up to 100 µm (0.004 in.)
Copper and silver are relatively easy to strip after completion of nitriding.
However, copper stripping can be expensive due to effluent problems caused
by the cyanide-base plating solution, which must be neutralized. Nickel is
difficult to strip, and the stripping process could affect part surface quality.
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 and
zinc. 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 that
will not leave bristles on the painted steel surface. When the steel is heated to
the process temperature, loose bristles on the surface will burn off and
expose the steel immediately beneath. Nitriding will take place in the area of
the burnt-out paint bristles, which can wreak havoc with cutting tools.
Some stop-off paints are water-based and must be correctly cured. Any
water remaining in the stop-off paint will manifest itself during nitriding. The
water will form a steam pocket under the paint, producing a bubble. When
the bubble bursts, a localized hard spot will form on the exposed steel.
Stop-off paint residues can be reduced by brushing or washing, or
removed by light blasting with fine abrasives.
Methods for Selective Salt Bath Nitriding

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

Masking an area for plasma ion nitriding follows this simple rule:
“What plasma can see, it will nitride; what plasma cannot see, it cannot
nitride.” The area to be masked should be covered with a piece of shim
stock steel attached to the component. The shim stock serves as an 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 surface
will begin to transfer the paint onto the furnace wall or the anode. The
Chapter 15: Stop-Off Procedures for Selective Nitriding / 165
sputtered paint will migrate and adhere to the process wall, leading to a
change in the electrical response characteristics of the wall. Once a cycle
has been completed, the inner wall should be cleaned of deposits.
Another stop-off consideration is that when a component is in contact
with the cathodic hearth, it will not nitride. Therefore, a support system is
needed. As shown in Fig. 1, the part can be mounted onto steel points. The
points will maintain the cathodic contact with the furnace hearth and
nitriding can take place. The case depth will be slightly more shallow at
the point of contact but not to a significant degree.
Threaded holes must be protected from nitriding. This is accomplished
simply by partially inserting a threaded stud into the hole. Screwing three
or four threads deep is sufficient (Fig. 2).
Fig. 1 Extrusion dies lifted from the furnace hearth to allow the plasma glow
seam to cover the die completely. Note the die supports (steel tubes)
used to maintain cathode potentials.
Fig. 2 Masking of blind tapped holes on a component for plasma nitriding.


CHAPTER 16
Examination of the
Nitrided Case
EXAMINING AND EVALUATING the nitrided case is generally

accomplished by hardness testing and microscopic examination. This
chapter discusses both characterization methods, as well as sample
preparation.
Hardness Testing
Hardness testing is perhaps the most widely used method for evaluating
a nitrided case. Hardness of a steel is used as a cross-reference to many
other properties, such as tensile strength and impact strength. Generally,
hardness testing of a nitrided case is accomplished with a low load or
microhardness load below 1 kg. Do not use the Rockwell machine with a
150 kg load, which will punch through the nitrided case and provide an
incorrect reading.
Hardness testing can be categorized as either macrohardness or 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 load

applications below 1 kg. It can be carried out by:
• Rockwell superficial testing

• Vickers hardness testing
• Knoop hardness testing
Testing Hardness Profiles through the Case

The 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 microhardness
test method (Fig. 2, 3). This method of hardness testing requires either
(a) a test coupon of the same material that is in the same metallurgical
condition or (b) a sacrificial component.
Sample Preparation
The methods discussed in this section apply to preparation of samples
for both microhardness testing and microscopic examination.
Cutting, or sectioning, of a sample must be carried out under the coolest
conditions possible. This is accomplished by “flood cutting,” where the
Fig. 1 Step-ground specimen for hardness traverse method of measuring

depth of medium and heavy cases. Arrows show locations of hardness-
indenter impressions.
Fig. 2 Cross section of a round test coupon cut through to expose the nitrided
case in preparation for a microhardness traverse test. The exposed sur-
face must be polished before testing.
Chapter 16: Examination of the Nitrided Case / 169
Fig. 3 Results of a microhardness traverse test on a nitrided test coupon
sample is sectioned using an abrasive cutoff wheel while submerged in the

cutting fluid. Do not use excessive pressure or localized overheating likely
will occur, adversely affecting the cut surface metallurgical condition and
subsequent test results. Mixing a rust inhibitor into the coolant will reduce
the risk of surface oxide formation on both the sectioned sample and the
remaining sample portion. Diamond sawing is another method of cold 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 finish
considerably reduces the pregrind time (Fig. 4, 5).
When using an abrasive wheel cutoff unit, the wheel residuals and the
metallic fines that result from sectioning must be removed from the recir-
culating cooling system. The fines will build up quickly in the drain and
delivery hoses, significantly reducing the pipe diameter and thus restricting
Fig. 4 Laboratory abrasive cutoff device
Fig. 5 Diamond saw used for sectioning of very hard specimens. Very thin
cuts can be made with this tool.
the coolant drainage and delivery system. This applies to the machine
sump and the recirculating reservoir, both of which must be cleaned fre-
quently. In addition, the bearings on which the drive motor is connected to
the wheel load handle must remain well lubricated. If the bearings become
tight, then the force required to depress the abrasive wheel could lead to an
excessive load being placed on the sample being sectioned, creating the
potential for localized overheating of the sample surface.
Vapor degreasing is often used to clean the specimen after it has been
machined prior to mounting. Hot vapors of a chlorinated or fluorinated
solvent remove oils, greases, and waxes that may be on the specimen.
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 both
hardness traverse testing and microexamination after etching. Tables 1 and
2 list the properties of commonly used thermosetting and thermoplastic
mounting materials. Tables 3 and 4 show typical problems with compres-
sion mountings and castable materials.
There are essentially two types of mounting systems: the cold-mount
system and the hot-mount system. Each has its own advantages and
disadvantages.
The two-component cold-mount system comprises a polymer and a 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 be
placed with the sample to identify the surface of the nitrided sample to be
examined (Fig. 6).
Table 1 Typical properties of thermosetting molding resins

Molding conditions Heat Coefficient
distortion of thermal Abrasion Polishing
Temperature Pressure temperature(a) expansion
Time, 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 and
alkalies
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 and
alkalies
(a) Determined by method ASTM D 648. (b) Specimen 100 mm2 (0.15 in.2) in area abraded on slightly worn 600-grit silicon carbide under load of 100 g at rubbing speed of
105 mm/min (4 × 103 in./min). (c) 25 mm (1 in.) diam mount on a wheel rotating at 250 rpm covered with synthetic suede cloth and charged with 4 to 8 µm diamond paste
Table 2 Typical properties of thermoplastic molding resins

Molding conditions
Heating Cooling Coefficient
Heat distortion of thermal Abrasion Polishing
Temperature Pressure Temperature Pressure temperature(a) expansion,
Time Time 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 materials

Problem Cause Solution
Thermosetting resins
Too large a section in the given mold Increase mold size; reduce specimen
area; sharp cornered specimens size.
Radial split
Excessive shrinkage of plastic away Decrease molding temperature; cool

from sample mold slightly prior to ejection.
Edge shrinkage
Absorbed moisture; entrapped gases Preheat powder or premold; momen-

during molding tarily release pressure during fluid
state.
Circumferential splits
Too short a cure period; insufficient Lengthen cure period; apply sufficient
pressure pressure during transition from fluid
state to solid state.
Burst
Insufficient molding pressure; Use proper molding pressure; increase

insufficient time at cure temperature; cure time. With powders, quickly seal
increased surface area of powdered mold closure and apply pressure to
materials eliminate localized curing.
Unfused
Thermoplastic resins
Powdered media did not reach Increase holding time at maximum

maximum temperature; insufficient temperature.
time at maximum temperature
Cottonball
Inherent stresses relieved upon or after Allow cooling to a lower temperature

ejection prior to ejection; temper mounts in
boiling water.
Crazing
Fig. 6 Epoxy mounts with binder clips to hold the specimen perpendicular to
the polished surface
Table 4 Typical problems of castable mounting materials

Problem Cause Solution
Acrylics
Too violent agitation while blending Blend mixture gently to avoid air
resin and hardener entrapment.
Bubbles
Polyesters
Insufficient air cure prior to oven cure; Increase air cure time; decrease
oven cure temperature too high; oven cure temperature; correct
resin-to-hardener ratio incorrect resin-to-hardener ratio.
Cracking
Resin-to-hardener ratio incorrect; Correct resin-to-hardener ratio; keep

resin has oxidized containers tightly sealed.
Discoloration
Resin-to-hardener ratio incorrect; Correct resin-to-hardener ratio; blend

incomplete blending of resin-hardener mixture completely.
mixture
Soft mounts

incomplete blending of resin-hardener mixture completely.
mixture
Tacky tops
Epoxies
Insufficient air cure prior to oven cure; Increase air cure time; decrease
oven cure temperature too high; oven cure temperature; correct
resin-to-hardener ratio incorrect resin-to-hardener ratio.
Cracking
Too violent agitation while blending Blend mixture gently to avoid air
resin and hardener mixture entrapment.
Bubbles
Resin-to-hardener ratio incorrect; Correct resin-to-hardener ratio; keep

oxidized hardener containers tightly sealed.
Discoloration

incorrect blending of resin-hardener mixture completely.
mixture
Soft mounts
The primary disadvantage of the cold-mount system is the difficulty of

obtaining two parallel faces after curing and retrieval from the mold. Its
primary advantage is its low cost.
The second method is the hot-mount system, also known as compres-
sion molding. The mold is usually made of a thermosetting material:
• Bakelite (phenolics)
• Epoxy
• Diallyl phthalate
This system requires a pressure mold equipped with a heater system to

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

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

length, laid out on a smooth, flat surface. Once again, sufficient water flow
is important to ensure that the silicon carbide strip is well flushed. The
secret to reducing the polishing time is careful pregrinding.
Polishing follows the pregrinding operation. Do not let the sample sit
after the pregrind for any length of time. If the exposed sample surface is
wet, it will begin to oxidize, which could cause very slight pitting. The
polishing operation should start immediately after pregrinding to reduce
the risk of oxide formation. One of two polishing mediums can be used:
• Diamond paste. While diamond paste (Fig. 10) is perhaps the hardest
material and will produce very highly polished surfaces, additional
lubrication is necessary to ensure good distribution of the paste over
the polishing cloth to improve polishing efficiency. Recommended
particle sizes for the final polish are 1 µm or 0.3 µm. Diamond pastes
are very expensive.
• Aluminum oxide slurry in suspension. Aluminum oxide slurries for
final polishing can have particle sizes of 1 µm down to 0.5 µm. The
particles are suspended in water or in paste or powder form. Selection
depends on personnel and availability. Such slurries are self-lubricating
and economical.
The flat wheel of the polishing machine should be tightly fitted with a
well-lubricated soft-nap polishing cloth. The rotational speed of the
wheel should be approximately 400 to 450 rpm. Once the polishing is
complete, rinse the polished sample under lukewarm running water, then
Fig. 10 Application of diamond paste to a nylon cloth for preliminary polishing

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

The sample should be etched only after microhardness tesing has been
completed. If etched beforehand, the hardness indentations may not be
clear enough to read accurately.
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 mL
HNO3 (nitric acid) and 94 to 96 mL ethyl alcohol (or denatured alcohol).
The common error when using nital involves etching time. The sample
is usually oversoaked in the etchant, resulting in a surface that is too dark
for accurate examination. The specimen should be etched for approxi-
mately 3 to 8 s, taking care not to overetch. The exact etching time, of
course, depends on the type of steel being examined.
Nital will reveal the compound zone on the immediate surface and sta-
ble nitrides in the case directly below. It will not, however, show the two
phases of the compound zone. The compound zone will show as the sur-
face white layer. In the case of a low-alloy steel or a plain carbon steel, the
iron nitride precipitates will be seen as well-defined straight lines, below
which the steel core will exhibit iron nitride precipitates.
Picral is another common etching medium used to examine a nitrided
surface. It is not complex, consisting of 4 g picric acid and 100 mL
ethanol. The addition of 0.5 to 1% zephiran chloride improves etch rate
and uniformity. Specimens should be immersed for 3 to 15 s, taking care
not to overetch.
Care should be exercised when using picric acid (refer to the Material
Safety Data Sheet, MSDS). Proper storage of picric acid powder is critical
to operator and technician safety, and the powder should be kept very
moist or even under water.
Tips for Using Etchants. Here are a few helpful guidelines:
• When mixing etchants, always add reagents to the solvent unless spe-
cific instructions indicate otherwise.
• Where water is given as the solvent, distilled water is preferred
because the purity of tap water varies.
• Methanol is usually available only as absolute methanol. When using
this alcohol, it is imperative that approximately 5 vol% of water be
added whenever an etchant composition calls for 95% methanol. Gen-
erally the alcohol used is denatured alcohol.
• For conversion of small liquid measurements, there are approximately
20 drops per 1 mL.
• Etching should be carried out on a freshly polished specimen.
• Gentle agitation of the specimen or solution during etching will result
in a more uniform etch.
• The etching times given are only suggested starting ranges and not
absolute limits.
• In electrolytic etching, a direct current (dc) is implied unless other-
wise indicated. An economical source of dc current for small-scale
electrolytic etching is a standard 6 V lantern battery.
• Microscope objectives can be ruined by exposure to hydrofluoric acid
fumes from etchant residue inadvertently left on the specimen. This
problem commonly occurs when the specimen or mounting media con-

tain porosity and when the mounting material (such as Bakelite) does
not bond tightly to the specimen, resulting in seepage along the edges.
• In all cases, extreme care should be taken to remove all traces of the
etchant by thorough washing and complete drying before placing the
specimen on a microscope stage.
Etchant Removal. Once etching has been completed, quickly wash the
sample under clean, cold running water. Otherwise, the etchant will con-
tinue to etch the specimen surface. After thorough water rinsing, rinse the
surface with clean denatured alcohol. Then take a household hair dryer set
on very low heat and blow-dry the alcohol across the face of the specimen
in one direction. Ensure that the sample is dry on all sides of the mount
and then examine. If surface stains have occurred, lightly repolish the
specimen on the rotating polishing cloth wheel and re-etch, completing
the procedure once again.
Safety Precautions
Here are some practices that will contribute to the safe preparation of
specimens. The list is not exhaustive. Obviously, care must be taken when
machining samples.
Preparing Etchant. When preparing an etchant, the operator or 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, rubber
gloves, rubber apron, long sleeves, and shoe protection.
• Mix acid to water, not water to acid (Fig. 11).
• Use proper solution mixing, storage, and handling equipment and
facilities (e.g., acid-resistant glass).
• Wipe up all spills and leaks, no matter how small, using the “spill kit”
and “spill treated wipers.”
• Dispose of all solution not correctly identified by composition and
concentration. When in doubt, throw it out—but do so in a 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 chemical
manufacturer.
Vapor degreasing is a cleaning process that uses the hot vapors of a

chlorinated or fluorinated solvent to remove soils, particularly oil, greases,
and waxes. Extreme care should be taken in handling the vapor. Follow
these precautions (Ref 2):
• Provide adequate ventilation.

• Do not breathe the vapor fumes.
Fig. 11 Add acid to water, not vice versa. Wear a full apron, a full face mask,
rubber gloves, long sleeves, and shoe protection.
• Wear protective clothing, including gloves, face mask, and apron.

• Do not allow any of the liquid to come in contact with the skin.
• Ensure that the vapor does not overheat or come in contact with 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 be
taken in its handling. Use cotton gloves or rags when moving the compo-
nent into the nitriding retort. Otherwise, contamination from fingerprints
will cause resistance to nitrogen penetration, leading to soft spots.
Optical Light Microscopy

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

• Is it necessary to observe only the nitrided case?
• Is a permanent record (i.e., a micrograph) needed?
• Is visual imaging analysis required?
Microscopes range widely in cost and performance. The availability of

spare units and service from the supplier is another consideration. For qual-
ity control and troubleshooting, light microscopy up to 1000 to 1200× will
cover most metallurgical requirements. When research of the development
of the case and surface metallurgy is involved, the user can usually obtain
scanning electron microscopy services through a local metallurgical or
materials science school. Examples of case microstructures are shown in
Fig. 12 to 18.
Fig. 12 AMS 6470 steel with 0.15 to 0.35% Pb added, oil quenched from
900 °C (1650 °F), tempered 2 h at 605 °C (1125 °F), surface activated
in manganese phosphate, and gas nitrided 30 h at 525 °C (975 °F). Structure is a
white layer of Fe2N and a matrix of tempered martensite. 2% nital. 400×
Fig. 13 Same material and heat treating conditions as described in Fig. 12,
except nitrided 36 h. The depth of the nitride layer has increased, and
platelets of iron nitride can be seen in the case. 2% nital. 400×
Fig. 14 Same steel and processing as Fig. 12, except slack quenched and
ground heavily before nitriding. Because the surface was not chemi-
cally activated before nitriding, nitrogen diffusion was retarded. 2% nital. 400×
Fig. 15 4140 steel, oil quenched from 845 °C (1550 °F), tempered 2 h at 620
°C (1150 °F), surface activated in manganese phosphate, and gas
nitrided 24 h at 525 °C (975 °F). Structure is white layer of Fe2N, Fe3N, and
Fe4N, and tempered martensite. 2% nital. 400×
Fig. 16 H13 steel, heated to 1030 °C (1890 °F) in a vacuum, quenched in

nitrogen gas, triple tempered at 510 °C (950 °F), surface activated in
manganese phosphate, and gas nitrided 24 h at 525 °C (975 °F). White surface
layer is iron nitride. Grain-boundary networks of nitride are present throughout
the martensitic case. 2% nital. 300×
Fig. 17 18% Ni maraging steel (300 CVM), solution treated 1 h at 815 °C

(1500 °F), surface activated, and gas nitrided 24 h at 440 °C (825 °F).
Etching has made the nitride surface layer and grain-boundary nitrides appear
black. Modified Fry’s reagent. 1000×
Fig. 18 4140 steel, quenched and tempered to 30 HRC, then ion nitrided
24 h at 510 °C (950 °F). Monophase surface layer of Fe4N, plus a dif-
fusion zone of nitride containing tempered martensite. Nital. 750×
REFERENCES
1. N.I. Sax, Handbook of Dangerous Materials, Reinhold Publishing, 1951
2. H.B. Elkins, Chemistry of Industrial Toxicology, John Wiley & Sons,
1959
SELECTED REFERENCES
• B.L. Bramfitt and A.O. Benscoter, Metallographer’s Guide: Practices
and Procedures for Irons and Steels, ASM International, 2002
• C. Johnson, “Metallography, Principles & Procedures,” Leco Corpo-
ration, St. Joseph, MI

CHAPTER 17
Troubleshooting
PROBLEMS often occur during nitriding, just as with any other heat
treatment process. They can take the form of:
• Process problems
• Steel problems
• Machining problems
Troubleshooting is a process of elimination and plain old detective

work. One must be both observant and systematic during the trouble-
shooting procedure.
Gas Nitriding
Surface Cleanliness. A typical problem with gas nitriding originates
from part surface cleanliness. If the work surface is contaminated with a
hydrocarbon-based substance such as cutting oils, machining oils, lapping
compounds, or fingerprints, the affected area will not successfully nitride,
and the area below it will be soft. The remedy is to ensure absolute surface
cleanliness prior to gas nitriding. This can be accomplished by degreasing
using either chemical or ultrasonic methods.
If the part surface is contaminated with a cutting fluid that is chloride-
or sulfide-based, then serious surface pitting and loss of hardness may
result. It can be safely said that any grease or acidic compound on the sur-
face of a steel is very likely to cause nitriding problems.
Loss of Gas Dissociation. When gas dissociation cannot be achieved
during the nitriding process, something within the process hardware is
taking the gas flow away from the components being nitrided. Check that
there is no restriction in the feedline from the ammonia gas source to the
process chamber. If the line is internally oxidized, it will restrict the
ammonia flow. Internal oxidation of the process gas-delivery system
results from the incorrect pipe material selection. Check also that the inlet
pipe to the process chamber is clean and clear.
If the support fixture for the work is made from a low-alloy material
such as mild steel, then it will pull the dissociated gas away from the
workpiece. This also applies to the process chamber fabrication material.
Even if the process chamber is made from the correct material, it can
begin to pull the ammonia necessary for the nitriding process away from
the workpiece if it is reaching the end of its useful life, or if it has become
saturated with nitrogen. This can be remedied by shot blasting the inside
surface of the process chamber or by regenerating the chamber by heating
up to a temperature of 785 °C (1450 °F), followed by light shot glass bead
blasting or sandblasting.
Surface Discoloration. When the process chamber is opened after
nitriding, the parts occasionally may be discolored with an almost rain-
bow effect. The discoloration is an oxide formation on the steel surface
and is in no way detrimental to the part. In fact, it will enhance the surface
corrosion resistance. This is the principle behind the oxynitride 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 nitriding
process.
Case Exfoliation. When the case exfoliates, or peels off, this is usually
indicative of surface decarburization present on the immediate surface of
the steel. Surface decarburization can occur as a result of:
• Insufficient stock removal from the steel surface, leaving a decar-

burized layer that will result in defective surface metallurgy. The
remedy is to ensure removal of more than 10% of the surface stock
thickness.
• A decarburizing condition during the preheat treatment operation that
has left the surface seriously decarburized
Components that exhibit these surface conditions cannot be salvaged and

are scrap.
Should the surface exhibit a failure of the case due to chipping (e.g., on
the pressure face of a gear), it is usually indicative of an aluminum-bearing
(Nitralloy) steel. The aluminum-bearing steels have very high surface
hardness values and cannot be subjected to an impact load condition. If a
gear tooth tip begins to chip off, it generally indicates a supersaturated case
formation of nitrogen, where the solubility limit of nitrogen in iron has
been exceeded, resulting in nitride networking and a brittle case. This usu-
ally takes place on sharp corners and edges. The only course of action is to
investigate the gas dissociation control.
If the surface of the case exhibits light, flaky peeling, some type of sur-
face contamination may be present. Investigate the cleaning treatments
Chapter 17: Troubleshooting / 187
prior to nitriding. Surface contamination is the most probable cause of any

flaking or exfoliation of the nitrided surface. Contamination may be due to:
• Grease
• Oil
• Residual cutting fluids
• Residual heat treatment salts (see discussion below on salt bath nitriding)
• Insufficient precleaning
If the flaking is persistently seen on most loads processed through the fur-
nace, the source of contamination may likely be within the furnace process
retort. Examine the internal faces of the retort for signs of contamination.
The interior walls may simply require a cleaning operation or wipe down. If
necessary, the wipe down can be accomplished by using an alcohol-based
solution and hand wipes. (Appropriate ventilation should be used to ensure
that the alcohol fumes do not overcome the operator.) It may be advisable to
consider either a burnout of the process retort or complete replacement.
Orange Peel Effect. If the workpiece surface shows what can only be
described as an orange peel effect, with the appearance of dimples, then
once again it can be traced back to surface decarburization. The remedy is
to investigate the surface machining and the preheat treatment conditions.
The part is not salvageable unless there is a deep case and the surface 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 can
be seen on the surface of nitrided gears. It can be avoided by:
• Increasing the case depth

• Improving the core hardness by reducing the previous tempering 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 and
ensure reasonably repeatable nitriding conditions: process salt chemistry
and bath cleanliness.
Process Chemistry. The subject of discussion here is salt bath nitriding
and not ferritic nitrocarburizing. There is a great distinction between the
two processes. It is important to understand the chemical makeup of the
nitriding salt. Once the salt chemistry is understood, along with the basics
of salt titration for measuring nitriding activity, the bath should be titrated
at least once per day in order to maintain a consistent bath activation. The
usual practice is to titrate for cyanide only, not carbonate.
Bath Cleanliness. Because there are many sources of bath contamina-

tion, the bath must be regularly desludged. Desludging should take place
at least once per day, or even once per shift if the bath is operating on a
shift basis. This is extremely important. The main contaminant in the bath
comes from iron oxide and the precipitation of products of the carbonate
content. The sources of iron oxide are fixtures, soft iron support wire for
suspending parts into the bath, and, if the bath is made from mild steel,
from the inside of the bath itself. Nitriding baths usually are fabricated
from mild steel or a low-carbon or low-alloy steel but in some instances
can be made from a heat-resisting stainless steel.
Postwashing Treatments. If the workpieces have been previously heat
treated through salt baths such as neutral salts followed by a quench into a
marquenching salt, thorough postwashing is necessary to prevent exfolia-
tion or flaking of the case.
Salt Bath Nitriding of Alloy and Tool Steels. When nitriding the
higher alloyed steels and tool steels, the nitrogen diffusion rate will not be
as great as for the lower alloy steels. Therefore, do not confuse a shallow
case as being caused by the cyanide level of the bath; it is simply a matter
of a slower rate of diffusion. Cycle times are usually short, in the region of
5 min up to 180 min, without quench.
Ion Nitriding
With the ion nitriding process, process problems as well as metallurgical
problems can occur. Process problems generally will affect the process
metallurgy. Care of the furnace is mandatory, even in terms of initial instal-
lation of the equipment. Considering the cost of water, the water-cooling
system usually is a closed-loop system that goes through a water chiller or
cooler. The system must be piped from the inlet of the cooler/chiller to the
return of the chiller/cooler. If the pipework is made from galvanized steel
tube, then very aggressive corrosion will begin to occur that over time will
restrict water flow. Use either copper or stainless steel tubing.
O-Ring Seal. Another common problem involves the overuse of vac-
uum grease on the O-ring seals before commencing the process cycle. The
grease can be your friend, but it can also be your enemy in terms of vac-
uum accomplishment. Like sticky flypaper, too much grease on the seal
will attract dust. The amount of vacuum grease required is as much as it
takes to make the tip of your finger look wet. It is not necessary to squeeze
the grease from the tube into the palm of your hand. Wipe the seal clean
before closing the process chamber and then apply the vacuum grease
sparingly. Do not use any other type of grease.
Process Vessel Cleanliness. Bell-type plasma nitriding furnaces
require regular cleaning. Each week, wipe the inside of the process cham-
ber walls to remove any carbon particles and sputtered particles, taking
care 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 of

level and condition. Ensure that the vacuum pipe connecting the vessel to
the pump is fitted with a valve or trap to prevent the possibility of back-
streaming of vacuum pump oil when a pressure differential occurs
between the process chamber and the vacuum pump.
Overheating at a workpiece corner or edge can be visually seen through
the nitride furnace sight port. This likely is due to a problem with high volt-
age, high pressure, or a combination of both. Check the vacuum pressure
level and decrease the pressure. Alternatively, reduce the process voltage.
These adjustments, separately or together, will reduce the overheating. The
same applies when a hollow cathode is observed. If the localized overheat-
ing is within 55 °C (100 °F), then nonuniform case metallurgy will occur.
Temperature uniformity within the process retort, be it gas or plasma
nitriding, is mandatory.
Loss of Nitriding. If the plasma glow seam provides only partial cover-
age, the portion of the workpiece that is not covered by the glow will not
be nitrided. Partial coverage is caused by a process chamber pressure that
is too high toward atmospheric pressure. The only way to correct this is to
reduce the pressure until the glow seam uniformly covers all the surfaces
to be nitrided (Fig. 1).
Fig. 1 Complete glow seam coverage, ensuring complete nitriding of all sur-
faces. Courtesy of Plateg GmbH
Another condition that can cause the loss of glow seam coverage is
when the surface area available for nitriding exceeds the original design
characteristics of the furnace. This is particularly true for continuous
direct current (dc) systems. The design basis of the continuous dc unit is
the power ratio of current density in relation to surface area to be treated.
Therefore, if the surface area exceeds the original design capabilities of
the system, then there will be a loss of plasma coverage and subsequent
partial nitriding of the component.
Arc discharge is likely to occur if the plasma pulse duration is too
long. The solution is to shorten the pulse duration time to the point where
no discharge occurs. Too high a pulse voltage can cause the discharge,
especially if the pulse duration time is long. All of this is based on the
assumption that the operator can see the arc discharge through the furnace
viewing port. If the operator is not in attendance, the arc discharge will not
be seen. It will, however, be evidenced by the appearance of the work-
piece after the cycle is complete. The furnace technician must understand
the possible causes of the arc discharge phenomenon and be able to pro-
gram the unit so that arc discharging will not occur. For continuous dc
plasma furnaces, the technician can only adjust the process voltage and/or
the process pressure (Fig. 2, 3).
Chipping on corners and edges is usually caused by the oversaturation
of nitrogen in iron. Remember, the solubility of nitrogen in iron at tradi-
tional nitriding temperatures is between 5 and 7 wt% (changing, of course,
with changes in alloying elements in the steel). Oversaturation means that
excess nitrogen is not in solution, but has migrated to grain boundaries at
corners and edges, forming nitrides and thus weakening the area.
It is only possible to observe this phenomenon by light microscopy. The
excessive nitrides at the grain-boundary areas will be seen as excessive
Fig. 2 Workpiece during plasma nitriding with continuous dc glow discharge.

Numerous micro-arcs are visible on the workpiece surface and may
produce microscopic damage. A large concentration of micro-arcs can result in
an avalanche-like increase in power. A big arc will form, destroying the surface
by melting due to overheating. Courtesy of Plateg GmbH
Fig. 3 The same workpiece as in Fig. 2, but during plasma nitriding with
pulsed dc glow discharge. Conditions such as vacuum pressure, gas
mix, and power input remain the same. By using pulsed dc with a repetition fre-
quency of about 10 kHz, the formation of micro-arcs is suppressed. Courtesy of
Plateg GmbH
white lines in the corner or edge area. The corrective action is to reduce
the nitrogen-to-hydrogen ratio or reduce the process temperature. How-
ever, once the nitride networking has occurred, it cannot be removed, but
only adjusted for on the next process cycle. It is thus important to place a
test coupon in the load (preferably of the same material and condition as
the parts being processed) for metallurgical examination after processing.
Low surface hardness after ion nitriding can be attributable to a 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 can
be 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 a
time period and then off too long means a loss of nitriding effect on both
the surface and the case depth.
Surface contamination is another possible cause of low surface hard-
ness. Some cutting fluids are silicone based; any residual deposits will act
as a barrier to nitrogen diffusion.
Surface flaking is usually caused by some form of surface contamina-
tion. The contamination can be due to decarburization from previous heat
treatment operations, machining operations using coolants that contain
silicones, or inadequate cleaning after salt bath treatment. Another 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 of
what system is used.

CHAPTER 18
What Is Meant by
Ferritic Nitrocarburizing?
FERRITIC NITROCARBURIZING accomplishes surface treatment of

a part in the ferrite region of the iron-carbon equilibrium diagram (Fig. 1).
As the process takes place in the ferrite region, both nitrogen and carbon
diffuse into the steel surface. The process is categorized as a 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 carbon
atoms into a solid solution of iron, thus entrapping the diffused atoms in the
interstitial lattice spaces in the steel structure (Ref 1).
As with the nitriding procedure, there are many methods and deriva-
tives of ferritic nitrocarburizing. These are discussed in the chapters that
follow.
Process Benefits
Ferritic nitrocarburizing improves the surface characteristics of plain
carbon steels, low-alloy steels, cast irons, and sintered ferrous alloys. As
described in later sections of this chapter, resistance to wear, fatigue, and
corrosion are improved with the introduction of nitrogen and carbon.
Scuffing resistance means the resistance to wear on the metal sur-
face. This is accomplished by changing the nature of the surface com-
pound layer, which is also known as the white layer. The completed
compound layer will form with both epsilon (ε) and gamma prime (γ ′)
phases. The dominant ε-phase resists abrasive wear.
Fatigue properties of steel are greatly improved by altering the com-
position of the compound layer. This means that treated steel has greater
resistance to fatigue failure than an untreated steel (Ref 1).
Corrosion Resistance. After ferritic nitrocarburizing, steel parts can
withstand many hours in a salt spray environment, whereas an untreated
plain carbon steel will fail the corrosion test very rapidly.
Fig. 1 The iron-carbon equilibrium diagram. The nitrocarburizing process is carried out in the ferrite region
(alpha iron) of the diagram.
Low Distortion. Another major advantage of the ferritic nitrocarburiz-

ing process is that the procedure is carried out at a low temperature that
prevents phase changes in the steel (from ferrite to austenite), thus reduc-
ing the risk of distortion. Distortion is the result of the release of induced
stresses, the thermal shock of quenching, and the risk of incomplete trans-
formation to martensite. No phase change occurs during the ferritic nitro-
carburizing treatment.
Chapter 18: What Is Meant by Ferritic Nitrocarburizing? / 195
Early History of Ferritic Nitrocarburizing

Ferritic nitrocarburizing has been a proven process for many years and
is now gaining much acceptance by engineers. This increased interest in
the process, the author believes, is due to engineers gaining a better under-
standing of materials selection and metallurgists gaining a greater under-
standing of process capabilities and restrictions. In addition, many furnace
manufacturers want to serve their clients by developing new and more
efficient process methods and equipment.
The early methods of ferritic nitrocarburizing were accomplished in
low-temperature (550 °C, or 1020 °F) salt baths working on the principle
of the decomposition of cyanide to cyanate (in the ferrite region). 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 the
formation of:
• Nitrides: The nitrides were formed as a result of the nitrogen compo-

nent contained in the cyanide salt. The nitrogen diffused into the steel
to form iron nitrides in low-alloy steels and stable nitrides in higher
alloyed steels.
• Carbon: The carbon was supplied from the salt in limited quantities
and formed carbides, interspersed with the formed nitrides.
• Sulfides: The sulfur addition to the salt formed sulfides in the case,
providing a self-lubricating property.
The action of the molten salt at the process temperature also caused
slight surface porosity on the treated steel. This allowed the surface pores
to become minute reservoirs, retaining lubricant on the immediate surface.
The net result was that the treated component resisted scuffing and
exhibited excellent resistance to frictional wear problems. The process
was a great success with high-speed spindles and high-speed cutting tools.
It did, however, require careful salt bath analysis on a daily basis (Ref 1).
Another challenge of the process was that the salt was not very water solu-
ble. The treated component required extensive hot water cleaning after
treatment. Cleaning became a major issue.
Problems associated with salt bath processing led to experimentation
with gaseous methods of ferritic nitrocarburizing. Experiments were con-
ducted in the late 1950s with gaseous methods by Cyril Dawes of Joseph
Lucas Ltd. in England. The company successfully applied for a patent on
the process in 1961 (Ref 3).
The gaseous procedure produced a porous layer very similar to the layer
produced with the Sulfinuz process (with the exception of forming surface
sulfides), which claimed to provide good antifrictional properties. The
process patent stated that the gaseous atmosphere consisted of ammonia,
with a hydrocarbon gas and other small amounts of carbon-containing

gases (Ref 3).
An important study that contributed greatly to the scientific understand-
ing of gas nitrocarburizing treatments and compound layer structure was
published 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 process
technique of using salt baths for the ferritic nitrocarburizing process. If
anything, it spurred on the salt manufacturers to develop more environ-
mentally friendly salts and cleaner procedures. Degussa of Germany
developed the salt bath process of “Tufftride,” a two-component process
that formed both nitrides and carbides in the immediate surface of the
steel (Ref 1). The process will produce only very shallow case depths,
approximately 0.05 mm (0.002 in.) deep, but with high surface hardness
values, good fatigue properties, and excellent corrosion resistance. The
process cycle times are relatively short (in the region of 1.5 h), followed
by a quench (Fig. 2).
Once again, the process relies on the decomposition of cyanide to
cyanate, which is accelerated by the introduction of a titanium aeration
tube. The aeration tube passes air through the molten salt from the bottom
of the salt pot. The system requires good operational maintenance in terms
of regular bath desludging, salt analysis, and periodic regeneration. This
requires raising the bath temperature to 575 °C (1070 °F) and holding for
approximately 2 h, followed by another desludging operation. The pur-
pose is to precipitate out of the molten salt any free iron originating from
work support baskets and fastening wire used to wire the components in
place in the work basket (Fig. 3).
Fig. 2 Typical time-temperature process cycle for a ferritic nitrocarburizing

procedure using salt baths
Fig. 3 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 made
of sheet metal, also insures that parts are coated with salt before nesting together.
Basket may be made of carbon or alloy steel rod and steel wire mesh. Work must
be free from oil, or the parts will stick together. Parts must be dry. (b) Inconel 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 with
trays, intended for small parts. Trays provide a maximum of loading space with-
out adversely affecting circulation. Entire fixture is made of Inconel. (d) Netted
fixture, of Inconel, for holding small parts with a head or shoulder. (e) Methods of
wiring small parts. Black annealed steel wire is used for parts weighing less than
10 lb; annealed stainless wire is used for heavier parts. (f) Hooks, made of nickel
alloy rod, for holding circular parts. (g) Method for holding large parts in which
tapped handling holes are available or can be provided. Nickel alloys are used
for such fixtures because of the need for high-temperature strength. Resistance to
oxidation is not a factor, as liquid carburizing salts are reducing. (h) Rack for
holding six small crankshafts; exploded view shows a crankshaft in position.
(i) Special rack for carburizing the outside diameters of bearing races. Holding
plates are made of mild steel; rods, of Inconel.
With the advent of pulsed plasma technology in the early 1980s for ion
nitriding, it did not take long to realize that another method of ferritic nitro-
carburizing had been discovered. This procedure was soon commercial-
ized. Advantages include faster process cycle times, less surface cleaning
and preparation, deeper case formation, and better control of surface metal-
lurgy formation. Equipment is now being built that is capable of performing
both pulsed plasma ion nitriding and ferritic nitrocarburizing in the same
process chamber and with the same pulsed power pack. The procedure
offers a more controllable, repeatable surface metallurgy.
Why Ferritic Nitrocarburize?

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

Cost of Insurance and Freight. What type of crating or container will

be used? After loading is complete, a visual inspection should be carried
out; if possible, digital photographs should be taken of the load in its posi-
tion in the event of possible insurance claims when the contents are later
uncrated. Other related costs and concerns include:
• Loading and delivery from the manufacture site to the point of departure
• Type of shipping line (conference or nonconference)
• Paperwork delays. If the equipment is shipped internationally, incom-
plete or improper paperwork can cause serious delays at either the port
of departure or the port of arrival. Such delays can be very costly.
Check on daily demurrage rates and duties payable.
• Insurance. The equipment must leave the manufacturing facility fully
insured. Before installation, check the suitability of the intended site
with the insurance carrier. Does the room or building have the neces-
sary fire protection? Would the existing fire protection system damage
the new equipment?
• Road transport from the port of arrival. Road transportation permits
may be necessary if the vehicle load is considered a wide load.
• Access. Be sure that before the equipment arrives, doors and wall
apertures are large enough to allow easy access of the equipment into
the facility. It can be embarrassing if the furnace will not fit through
the door. Preplanning models can sometimes be used to navigate large
equipment through plants.
Operating costs include materials, energy, disposal of spent chemicals,

labor (including training), rejected materials, and time. All these costs must
be evaluated on a per item or other basis before making a final decision.
Training
To ensure that the furnace goes together the first time (and hopefully
starts the first time), at least two primary discipline people—the operating
person and the maintenance person—should visit the manufacturing site
when the furnace is being assembled. They also should be present after the
hot trials to see how the furnace is dismantled. Photographically 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, and
the 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, as
well as reactions of the process in relation to part geometry.
While many more considerations may arise when preparing to ship a

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

CHAPTER 19
Salt Bath Ferritic
Nitrocarburizing
FERRITIC NITROCARBURIZING is a diffusion process that is a

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

• Time • Sealed and controlled environment
• Suitable nitrogen source
Fig. 1 Various trade names for gaseous, salt bath, and ion (plasma) ferritic
nitrocarburizing processes. Fluidized-bed procedures also are available.
It is now more than 40 years since the development of the low-temperature

cyanide-based salt bath nitriding process. Salt bath nitriding will form not
only the compound layer at the steel surface, but will allow nitrogen as
well as carbon to diffuse into the surface to improve fatigue strength, tor-
sional strength, and tensile strength. In addition, the process will further
improve wear resistance, galling resistance, and corrosion resistance.
Another advantage of both ferritic nitrocarburizing and nitriding is that
they are low-temperature processes. This means that distortion will be
kept to an absolute minimum. However, to say these processes are distor-
tion free is misleading.
Low-Cyanide Salt Bath Ferritic Nitrocarburizing

Salt bath nitrocarburizing was first established in the late 1940s when
high-cyanide nitrocarburizing salt baths were introduced. Environmental
considerations and the increased cost of detoxification of cyanide-containing
effluents have led to the development of low-cyanide nontoxic salt bath
nitrocarburizing treatments. This section summarizes the development of
one commercially successful low-cyanide treatment known as the Melonite
process developed by Houghton Durferrit (Ref 4).
Chapter 19: Salt Bath Ferritic Nitrocarburizing / 203
Melonite Process
The Melonite process, also known as Meli 1, uses a molten salt bath of a
special composition. It provides a wear- and scuff-resistant surface on
steels, sintered irons, cast irons, and similar materials. Treated parts exhibit
excellent wear and corrosion resistance and good sliding characteristics.
The process salt bath is mainly an alkali cyanate plus an alkali 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 the
component surface to form alkali carbonate. Either continuously or inter-
mittently, the bath is treated to regenerate the carbonate product back to a
cyanate.
The process forms a complete multilayer surface case that comprises a
compound layer and a diffusion layer. The surface compound layer, consist-
ing of different compounds of iron, nitrogen, and oxygen, resists abrasion
corrosion and scuffing and is fairly stable at elevated operating tempera-
tures. Surface hardness depends on the steel that is being treated, and
researchers at Durferrit claim surface hardness values ranging from 800 to
1500 HV. The diffusion layer consists of nitrides of the appropriate alloying
elements and formed carbides.
Like the surface hardness, the case depth will vary according to the type
of steel being treated. Simply stated: The lower the alloy content of the
steel, the lower is the resulting hardness value. The case depth, however,
is deeper. The higher the alloy content of the steel, the higher is the result-
ing hardness value. The case depth, however, is shallower (Fig. 2).
Compound Layer. During salt bath nitrocarburizing by the Melonite
process, a nitrocarburized layer is formed consisting of the outer com-
pound layer (ε-iron nitride) and the underlying diffusion layer. The base
material determines the formation, microstructure, and properties of the
compound layer.
The compound layer consists of compounds of iron, nitrogen, carbon,
and oxygen. Due to its microstructure, the compound layer does not possess
Fig. 2 Influence of chromium on diffusion layer hardness and total case depth
in various 0.40 to 0.45% C steels. Source: Ref 5
metallic properties. Particularly resistant to wear, seizure, and corrosion, it

also is stable almost to the temperature at which it was formed. Compared
with plasma or gas nitrocarburizing, compound layers with the highest
nitrogen content can be obtained by the Melonite process. Layers with high
nitrogen content give better protection against wear, particularly corrosion,
than those with a low content.
Depending on the material used, the compound layer will have a hard-
ness of about 800 to 1500 HV. Figure 3 compares surface layer hardnesses
produced by various processes.
In the metallographic analysis of salt bath nitrocarburized components,
the compound layer is defined clearly from the underlying diffusion layer
as a slightly etched zone. During the diffusion of atomic nitrogen, the
compound layer forms. The growing level of nitrogen results in the solu-
bility limit in the surface zone being exceeded, which causes the nitrides
to precipitate and form a closed compound layer.
In addition to the treatment parameters (temperature, duration, and bath
composition), the levels of carbon and alloying elements in the materials
to be treated influence the obtainable layer thickness. As stated previously,
the higher the alloy content of the steel, the higher the resulting surface
hardness value. The case depth, however, will be shallower.
The profile shown in Fig. 4 is for Meli 1 bath at 580 °C (1075 °F) with
the usual treating duration of 60 to 120 min. The compound layer obtained
was 0.1 to 0.2 mm (0.0004 to 0.0008 in.) thick on most materials. 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. Here
we 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, the
lower the nitrogen penetration depth at equal treating duration. On the
other hand, hardness increases with higher alloying content.
1400
480°C(895°F) 1.5 h
1200 480°C(895°F) 3.0 h
Vickers hardess HV 0.1
480°C(895°F) 6.0 h
1000
580°C(1075°F) 6.0 h
800
600
400
200
4 8 12 16 20 24 28 32 36 40
Distance from surface, in. × 10–4
Fig. 3 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
In the case of unalloyed steels, the crystalline structure of the diffusion

layer is influenced by the cooling rate after nitrocarburizing. After rapid
cooling in water, the diffused nitrogen remains in solution. If cooling is
done slowly, or if a subsequent tempering is carried out, some of the nitro-
gen could precipitate into iron nitride needles in the outer region of the
diffusion layer. This precipitation improves the ductility of nitrocarbur-
ized components. Unlike unalloyed steels, part of the diffusion layer of
highly alloyed materials can be better identified metallographically from
the core structure due to improved etchability. However, the actual nitro-
gen penetration is also considerably deeper than the darker etched areas
visible metallographically.
Cooling does not influence diffusion layer formation to any noteworthy
extent. Figure 5 shows the compound layer thickness for various materials
in relation to treatment time.
Q P Q
580 °C (1075 °F)
Air
350–400 °C 350–400 °C 350–400 °C
Temperature
(660–750 °F) (660–750 °F) (660–750 °F)

Oxidizing + cooling
Nitrocarburizing
Postoxidizing
Preheating
Polishing
Time
Fig. 4 Time versus temperature profile for the QPQ nitrocarburizing treat-
ment cycle. Source: Ref. 6
10 Carbon steels
Compound layer thickness
Melonite MELI 1 at e.g., 1015, 1045
580°C (1075°F) Low-alloy steels
18 e.g., 4140
Compound layer in ten thousands
High-alloy steels
6 e.g., D3, H11, 304
Cast iron
0
0.5 1 2 3 4
Treating time in the MELI 1 bath, h
Fig. 5 Compound layer thickness in relation to treating time for various mate-
rials. Note that the time scale is logrithmic. Source: Ref 4
Surface Hardness and Tensile Strength. The surface hardness obtain-

able by the Melonite treatment is influenced primarily by the composition
of 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 bath
nitrocarburized steels.
Wear Resistance and Running-In Properties. Due to the intermetallic
composition of the compound layer, friction and the tendency to weld with a
metallic counterpart are reduced. Melonite treated components exhibit excel-
lent sliding and running-in properties, as well as greater wear resistance.
Wear tests and practical applications repeatedly confirm the superior
wear resistance of salt bath nitrocarburized parts over traditional case-
hardened, induction-hardened, or hard-chrome-plated surfaces. In many
cases the wear resistance of the compound layer is further improved by an
oxidative post-treatment. For example, components such as transmission
shafts, plug gages, and hydraulic aggregates have a longer service life
after Melonite treatment than after hard chrome plating.
What about the wear resistance of the diffusion layer? Figure 6 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 that
ran against a salt bath nitrocarburized camshaft made from chilled cast
iron. Although the surface hardness of the case-hardened rocker arm was
slightly reduced by nitrocarburizing, the much improved wear resistance
due to the presence of the compound layer to approximately 80 h running
time is clearly visible. After 70 to 80 h, the wear profile runs parallel to that
of the case-hardened rocker arm, which is attributable to the protection
Table 1 Tensile strength and surface hardness for various nitrocarburized steels
Tensile strength after Approximate surface hardness after
hardening and tempering at 90 min at 580 °C (1075 °F)
Steel designation 600 °C (1110 °F), MPa using the Melonite process
Germany U.S. 2h 6h HV 1 HV 10 HV 30
Ck15 1015 600 550 350 300 200
C45W3 1045 750–850 700–800 450 350 250
Ck60 1060 750–900 700–800 450 350 250
20MnCr5 5120 800–950 800–900 600 450 400
53MnSi4 ... 850–950 800–900 450 400 350
90MnV8 O2 1000–1200 900–1100 550 450 400
42CrMo4 4140 900–1200 900–1100 650 500 450
X19NiCrMo4 ... 900–1100 900–1000 600 500 450
55NiCrMoV6 L6 1200–1400 1150–1300 650 550 500
50NiCr13 ... 1200–1350 1100–1200 600 500 450
X20Cr13 420 1000–1200 1000–1200 >900 600 450
X35CrMo17 ... 1000–1200 1000–1200 >900 700 550
X210Cr12 D3 1500–1700 1400–1600 >800 600 450
X210CrW12 ... 1500–1800 1400–1650 >800 600 500
X165CrMoV12 ... 1400–1900 1400–1700 >800 650 500
45CrMoW58 ... 1500–1800 1400–1700 800 700 600
X32CrMoV33 H10 1700–1800 1600–1750 >900 850 700
X38CrMoV51 H11 1700–1900 1500–1700 >900 850 700
X37CrMoW51 H12 1700–1900 1600–1800 >900 800 700
X30WCrV93 H21 1500–1800 1500–1700 >900 850 800
Source: Ref 7
afforded by the diffusion layer. A spontaneous increase in wear after the

loss of the compound layer was not observed (Ref 7).
This test again showed that a high surface hardness does not automati-
cally mean that the protection against wear is also very high. Assessment of
a material or mating materials depends on the wear mechanism involved.
Nitrocarburized running partners have proved themselves to be very good,
particularly under adhesive wear conditions. Their tendency to seize is
much lower than that of other surface layers. Figure 7 shows the results.
Salt Bath Nitrocarburizing plus Post Treatment (Ref 6)

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

Loss of weight of the running surface, g
Case hardened and

0.20 Melonite treated
Rocker arm: CrMo alloyed

0.15 case hardening steel
Camshaft: chilled cast iron,
Melonite treated
0.10
Test conditions: 1000 rpm
Load: 745-845 MPa
0.05 (110-120 ksi)
Oil: SAE 10W30
(80°C, or 175°F)
0 20 40 60 80 100
Running time, h
Fig. 6 Influence of surface treatments on rocker arm wear. Curves on left are
for case-hardened components. Curves at right are for case-hardened
and Melonite-treated components. Source: Ref 4
Fig. 7 Wear behavior of drawing dies after different surface treatments. SNC,
salt bath nitrocarburizing. Source: Ref 5
10 0.016
0.0112 0.011
(7.2) (7.1)
Weight loss, g /mm2

Weight loss, g /in.2
5 0.008
0.004
(2.3)
0.0005
(0.3)
0 0
Chromium Nickel SBN QPQ
20 µm 20 µm
Fig. 8 Corrosion resistance of various surface treatments on steel based on

field immersion tests. Test conditions: Full immersion for 24 h in 3%
sodium chloride plus 3 g/L hydrogen peroxide. Salt bath nitrocarburized with no
post-treatment. Source: Ref 6
treatment are carried out on the nitrocarburized surface. The quench-polish-

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

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

100
88 h
75
Surface area corroded, %

88 h
72 h
72 h
50
25
336 h
88 h
0
Hard Electroless QPQ
chrome nickel
Fig. 9 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
As the ferrous component is immersed into the molten salt, a catalytic sur-
face reaction takes place, causing the cyanate to decompose and release
the nitrogen and carbon for the process.
The decomposition of the salt to supply the nitrogen component of the
surface metallurgy satisfies the solubility limit of nitrogen in iron of 6 to
9%. The surface carbon content will be in the region of 1% in a solid solu-
tion in the ferrous component surface. In addition (as seen in the traditional
nitriding process), a surface compound zone begins to nucleate in which
the ε-nitride is the dominant phase in partnership with the γ′-phase.
Dimensional Stability
The SBN process is conducted at subcritical temperatures of 580 °C
(1075 °F); therefore, thermally induced microstructural phase changes
and accompanying volume changes do not occur. This permits the treated
component to be cooled from the SBN bath without drastic quenching,
thus minimizing dimensional changes and distortion. Accordingly, 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 of
595 °C (1100 °F). Stampings and other parts with thin cross sections are
well suited for SBN because of the capability for dimensional stability and
no distortion.
The SBN Processing Sequence

Referring to Fig. 10, the process begins with a prewash and preheat
cycle at 400 °C (750 °F) to ensure that the parts are clean and dry. A load
Fig. 10 Kolene QPQ process cycle. Source: Ref 9
of components that has been uniformly preheated will reduce thermal

shock and permit more efficient recovery of the SBN bath temperature.
The load is then transferred to the salt bath and held at 580 °C (1075 °F)
for a predetermined dwell time, depending on the required depth of com-
pound layer. From the salt bath, the parts are quenched into an oxidizing
salt bath (KQ-500) at a lower temperature, typically 400 °C (750 °F), and
held from 5 to 20 min. This intermediate quench cycle is a key ecological
feature of the SBN process since during this exposure to the oxidants,
cyanates and any cyanides generated within the nitriding bath and con-
tained as part of the dragged-out salt are effectively destroyed. Intermedi-
ate quenching also retards the cooling rate, thus minimizing thermally
induced distortion.
After the oxidizing quench, parts are cooled to room temperature,
rinsed, and, if required, subjected to post treatment. This may include
mechanical polishing, if surface finish is of concern, or the Kolene QPQ
treatment to develop maximum corrosion protection and/or enhance the
cosmetic appearance. The QPQ treatment provides a lustrous dark finish.
Process Control
An important factor in producing the desired monophase ε iron nitride
when nitrocarburizing is control of the nitrogen and carbon activities of
the processing environment. For SBN, this is accomplished by monitoring
and regulating the cyanate (CNO–) concentration within the operating salt
bath to 34 to 38%. As more parts, and thus more iron, are put through the
bath, cyanate content decreases until regeneration becomes necessary.
Based on results of standard titration analyses, an appropriate amount of
an organic polymer regenerant is added. This reacts with the carbonate in
the bath and converts it to cyanate.
Metallurgical Results
The 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-forming
alloying elements is not required to produce a compound zone. Formation of
nitride in low-carbon steel is seen in Fig. 11. Cast gray iron and ductile
(nodular) iron are also treatable by this diffusion process. The response to
SBN as measured by depths of compound zones and diffusion zones formed
and by the associated hardnesses of both zones is directly related to the mate-
rial composition. In general, with an increase in carbon content and/or in the
amount of alloying elements, the rate of diffusion decreases, thus resulting in
shallower depths of nitrogen penetration (Fig. 12a and b). Diffusion rates are
particularly sensitive to materials containing nitride-forming elements such
as chromium and molybdenum.
Carbon and Low-Alloy Steels. The SBN process time cycles are most
often limited to 2 h since diffusion rates decrease with time and long cycles
become increasingly nonproductive. Typically, a compound zone depth of
7.5 to 20 µm (0.0003 in. to 0.0008 in.) would be produced in 90 min for
carbon or low-alloy steels with diffusion zone depths ranging from 0.4 to
0.75 mm (0.015 to 0.030 in.), depending on the material (Fig. 13a and b).
Tool Steels. Compound zone depth requirements and associated time
cycles for tool steels depend on the application. High-speed cutting tools
require a very thin compound layer and thus are treated for only 10 to 20 min
at a reduced temperature (540 °C, or 1000 °F) to maintain base-material
hardness. This provides resistance to chip welding at the cutting edge.
The SBN treatment also benefits die casting and forging dies by reducing
heat checking, soldering, diffusion, and wear from molten metals (Fig. 14).
Cycle times for hot-work die steels may require 2 h to achieve the desired
properties. The performance of stamping dies is also increased by reducing
galling and seizing (Fig. 15).
Fig. 11 Microstructure of low-carbon steel treated by SBN plus Kolene

KQ-500 process. 500×. Source: Ref 9
Fig. 12 Thickness as a function of nitrocarburizing time for various alloys.

(a) Rate of diffusion decreases with increasing carbon and alloying
content, resulting in shallower penetration. (b) Total nitriding depth of specific
alloys. Source: Ref 5
Fig. 13 Low-carbon steel SBN treated and then aged to precipitate nitride
needles. Source: Ref 10
Fig. 14 Gross heat checking in a low-alloy tool steel forging die due to
excessive temperature. Heat checking occurred after an undeter-
mined number of 225 kg (500 lb) nickel-base alloy preforms had been forged
from an average temperature of 1095 °C (2000 °F). Source: Ref 11
8
Taber abraser test Untreated
7 AISI 1117 Steel
6
Wear, in. × 10–4
5
Case hardened
4 to HRC 54
2 SBN
1
0
0 10 20 30 40 50 60 70
Cycles × 103
8
Taber abraser test
7 AISI 1045 Steel
Hardened and
6 tempered to
Wear, in. × 10–4
HRC 28
5
SBN
4
2
1
0
0 10 20 30 40 50 60 70
Cycles × 103
Fig. 15 Wear resistance comparison between SBN treated specimens and

alternate treatments for two steels. Source: Ref 10
Stainless steels are SBN treated primarily to introduce wear resistance

and antigalling properties. While developing these properties, there could
be significant degradation of the stainless (corrosion resistance) qualities of
the material, which should be considered when specifying any 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. Cycle
times of 90 to 120 min are normally specified to generate compound zone
depths of 75 to 200 µm (0.003 to 0.008 in.) and diffusion zones to 0.4 mm
(0.015 in.). Both gray iron and nodular iron (Fig. 18a and b) are generally
treated to improve wear resistance and increase fatigue strength.
Fig. 16 Compound layer 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
Diffusion Curves
304 & 316 SS, Annealed
35 SBN in Nu-Tride
Compound layer depth, in. × 10–4
*Modified Nu-Tride
30 630°C
(1165°F) 580°C (1075°F)
25
+
20
510°C (950°F)*
15
10
5 455°C (850°F)*
400°C (750°F)*
0
0 2 4 6 8 10
SBN time, h
Fig. 17 Diffusion response of annealed types 304 and 316 austenitic stain-
less steel to SBN in Nu-Tride at 400 to 625 °C (750 to 1160 °F).
Source: Ref 13
Fig. 18 Cast irons treated in SBN. (a) Gray iron. SBN, 580 °C (1075 °F) for 2 h, followed by salt bath
quenching. Nital. Original magnification 500×. (b) Ferritic nodular iron. SBN, 580 °C (1075 °F) for
90 min, followed by salt bath quenching. Nital. Original magnification 500×. Source: Ref 10
A comparison of the microstructures of various ferritic metals after

SBN is given in Fig. 19.
Engineering Properties
The end result of SBN is the presence of two distinct layers of a nitrogen-
bearing microstructure, the outermost identified as the compound zone
and the subjacent layer called the diffusion zone. Each of the zones con-
tributes to improving performance by enhancing specific engineering
properties such as wear resistance, lubricity, corrosion resistance, and
fatigue strength. From these, other benefits in performance are realized,
including excellent running-in properties, antigalling, and antiseizing
characteristics, and reduced tendency for fretting corrosion.
Wear resistance is perhaps the most significant property resulting
from SBN. The ability of the compound zone to resist wear depends on
whether the wear is adhesive or abrasive. Adhesive wear occurs when two
components are in relative motion in an essentially abrasive-free environ-
ment. Under these conditions, the intrinsic physical characteristics of the
compound zone (i.e., hardness and lubricity) notably improve the sliding
and running-in behavior, and consequently increase the resistance to adhe-
sive wear (Fig. 20). The phase composition of the compound zone that
demonstrates the best wear resistance consists predominantly of ε-nitride
phase (monophase preferred) with a very small amount of γ′-phase.
Resistance to abrasive wear depends on the relative hardnesses of the
abrading substance and of the compound zone. For unalloyed carbon steels,
the compound zone hardness is equivalent to approximately HRC 55,
thus providing only short-time resistance to abradants of higher hardness.
One method of increasing hardness, and thus abrasive wear resistance, is
to increase the content of nitride-forming elements of the base material.
Fig. 19 Microstructures of various ferritic materials that have undergone salt

bath nitrocarburizing. All etched in 3% nital. All 500×. Courtesy of
Kolene Corp. (a) Ferritic nodular iron; 90 min at 580 °C (1075 °F), oxidizing molten
salt quench. (b) Low-carbon steel; 90 min at 580 °C (1075 °F), oxidizing molten salt
quench. (c) Type 304 stainless steel; 90 min at 580 °C (1075 °F), oxidizing molten
salt quench. (d) AISI D2 tool steel; 90 min at 580 °C (1075 °F), oxidizing molten salt
quench. (e) H13 medium-carbon hot-work tool steel; 120 min at 580 °C (1075 °F),
oxidizing molten salt quench
Progressive increases in chromium from 0% (carbon steel) to 17% (stain-

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

0.4
No lubrication
Coefficient of friction, µ
Lubricated with
0.3 SAE 30 oil
0.2
0.1
0.0
Chrome plated, Case SBN 90 min
30 µm hardened 580°C
(0.0012 in.) (1075°F)
plus SBQ
Fig. 20 Frictional properties for various surface treatments. SBQ, salt bath
quenching. Source: Ref 10
Corrosion resistance increases via SBN, most notably with the Kol-
ene QPQ process. Figures 8 and 9 show how QPQ treatments improve
corrosion properties.
Fatigue strength under bending and torsional loading develops as a
result of nitrogen present in the diffusion zone and particularly the amount
in solid solution. Improvement in fatigue strength produced through SBN
occurs in a range of materials and in varying degrees (Table 2). Low-
carbon steels with relatively low strength display the greatest increase in
fatigue strength (from 80 to 100%), whereas the higher alloy steels and
cast irons exhibit from 20 to 80% improvement in strength.
Other Methods for Salt Bath Nitrocarburizing

Sursulf, a ferritic nitrocarburizing procedure developed in France, also
is based on the principle of the decomposition of a low-percentage
cyanide to cyanate that is mixed with a low-temperature carbonate salt for
the activated carbon in the salt. The salt also has a sulfur compound in the
salt analysis to assist with the formation of surface sulfides as well as
some surface porosity. The porosity will form small reservoirs to hold sur-
face lubricants. The lubricants will also work with the sulfur to form an
almost self-lubricating surface with high wear resistance and improved
corrosion resistance.
Tenoplus is a two-stage high-temperature nitrocarburizing process. The
first stage is conducted at 625 °C (1160 °F) and held at that temperature,
followed by a cooldown to a lower temperature of 580 °C (1075 °F). Sub-
sequent cooldown takes place in an oxidizing bath to a lower temperature
of 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).
Table 2 Improvement in fatigue strength

Alloy treated Fatigue strength improvement, %
Low-carbon steels 80–100
Medium-carbon steels 60–80
Stainless steels 25–35
Chrome/manganese (low-carbon) steels 25–35
Chrome alloy (medium-carbon) steels 20–30
Cast irons 20–80
Source: Ref 10
REFERENCES
2. Cassel Manual of Heat Treatment and Case Hardening, 7th ed.,
Imperial Industries Ltd., United Kingdom, 1964
3. B. Prenosil, Hutn. Listy, 1962, Vol 17 (No. 6), p 414–424 (available
as translation BLL RTS 9520)
4. R. Willing-Lepenies and C. Faulkner, “New Developments in Salt
Bath Nitrocarburizing,” product literature, Houghton Durferrit, Val-
ley Forge, PA
5. G. Wahl and S. Alwart, Improvement of Tribological Properties
through Nitrocarburizing
6. Nitrocarburizing, Surface Hardening of Steels: Understanding the
Basics, J.R. Davis, Ed., ASM International, 2002, p 195–212
7. R. Willing-Lepenies and C. Faulkner, “Melonite-QPQ-Process,”
product literature, Houghton-Durferrit, Valley Forge, PA, p 9
8. “The Theory and Practice of Molten Salt Bath Nitriding,” product
literature, Kolene Corporation, Detroit
9. J.R. Easterday, “The Kolene QPQ Process,” product literature,
Kolene Corporation, Detroit
10. J.R. Easterday, “Salt Bath Ferritic Nitrocarburizing,” product 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 Bath
Nitrocarburizing, 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

CHAPTER 20
Gaseous Ferritic
Nitrocarburizing
GASEOUS FERRITIC NITROCARBURIZING, like salt bath nitro-

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

The gaseous nitrocarburizing process was first developed in 1961 at
Joseph Lucas (Industries) Ltd. in England. This treatment produced on
mild steel a porous layer that was claimed to have good antifrictional prop-
erties. The published patent revealed that the gaseous atmosphere consisted
of ammonia and hydrocarbon or other carbon-containing gases of unspeci-
fied proportions and that the treatment was undertaken in the temperature
range of 450 to 590 °C (840 to 1095 °F). During the 1960s, further research
led to the development of a large range of gas nitrocarburizing processes
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 gaseous
source, and oxygen. Ammonia is considered to be the most readily and
actively available source of nitrogen for the process and is blended with the
other process gas supplies of carbon (from a hydrocarbon gas) and oxygen
from other sources. As in gas nitriding, the cracked nascent ammonia gas
will dissociate at the steel surface and react with the hydrocarbon gas to
form both nascent nitrogen and free carbon. The gases will allow carbon
dioxide to be generated in relation to the classical water gas reaction:
CO2 + H2 ↔ H2O + CO
If we assume that ammonia is supplied at a constant pressure to the

process, a drop in the partial pressure of hydrogen will occur. This in turn
will increase the nitriding potential of the process (Ref 3) and lead to:
NH3 + CO ↔ HCN + H2O
This means that some hydrogen cyanide (HCN) gas will form as a by-
product of the process. The hydrogen cyanide will contribute both nitro-
gen and carbon to the process reaction, thus improving the nitriding
potential of the total process gas.
The metallurgical results of the process are very similar to the classical
nitriding process, with the exception that now there is carbon in the layer.
The thickness of the compound layer is determined by:
• Time
• Temperature
• Process gases
• Pretreatment of the steel
Chapter 20: Gaseous Ferritic Nitrocarburizing / 221
The rate of growth of the compound layer is approximately the square root
of the process time at the chosen process temperature.
The primary objective of the ferritic nitrocarburizing process is to form
nitrides and, most importantly, surface carbon, which will encourage the
ε-nitride phase in the surface of the steel. The surface carbon comes from
three sources:
• Decomposition of the process gas

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

Nitrocarburized Components
The properties obtained as a result of the gaseous ferritic nitrocarburiz-
ing process are as varied as those obtained with other process techniques.
Surface properties, however, can be modified simply by modifying the
process parameters.
The compound zone properties formed by the ferritic nitrocarburizing
process are generally very similar. There will be a significant difference if
the immediate surface has been oxidized after completion of the ferritic
nitrocarburizing treatment. The nonoxidized surface will show a signifi-
cant increase in:
• Wear resistance
• Torsional strength
• Corrosion resistance (except on stainless steels)
• Surface hardness
• Fatigue strength
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-alloy
steel. However, the process can also be applied to medium- and high-carbon
steels with various metallurgical results. Typical components that undergo
gaseous nitrocarburizing include:
• Machine spindles
• Cams
• Timing gears
• Powder metal components (see discussion below)
• Die casting dies (see discussion below)
• Shot sleeves
• Exhaust valves
• Cylinder liners
• Camshafts
• Crankshafts
• Steel water valves
• Ductile iron pump housings
• Automotive components, including suspension struts, stickshift levers,
and window winding mechanisms
• Gas spring pistons
• Door lock mechanisms
• Gears (see discussion below)
• Machine slides
• Cylinder liners
• Water pump components
Many other components can be successfully ferritic nitrocarburized.

Reduced material and processing costs can result in significant savings.
When an engineer is considering the use of gaseous ferritic nitrocarbur-
izing, he or she should consult with the heat treater or metallurgist to dis-
cuss process advantages and limitations as well as the metallurgical
behavior 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 that
will improve wear resistance, improve release properties, and resist the
buildup or welding of aluminum, magnesium, and zinc on mold and core
surfaces. This has led tool and die makers and mold makers to review the
properties and performance of ferritic nitrocarburized treatments in terms
of applications and performance on tool steels.
Gear Heat Treatment. Most gear manufacturing depends on control

tolerance and precision treating in terms of distortion. This means that a
process that can offer a significant reduction in distortion and an 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 a
quench 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 improves
antiscuffing properties and compression stress at the gear surface.
Sintered Steels. Extensive research work has been done on sintered
powder metallurgy (P/M) ferrous alloys, and ferritic nitrocarburizing is
now a fully commercial process for the P/M industry. The degree of suc-
cess depends on the level of compaction of the sintered part. Both gaseous
and plasma nitrocarburizing processes have found commercial use.
Safety Considerations
Because the gaseous ferritic nitrocarburizing process involves a com-
bustible atmosphere that is explosive when operated below the self-ignition
temperature, safety precautions must be rigorously enforced. It is mandatory
that all safety purge systems are in place and fully operable and that the cor-
rect purge sequencing has occurred before combustible gases are admitted to
the furnace. The process operating temperature is in the region of 570 °C
(1060 °F). The normal combustion gas ignition temperature is 750 °C (1350
°F), so any temperature below this will cause an explosion if the process gas
is not admitted according to the furnace manufacturer’s written procedures.
Appendix: Gaseous Nitrocarburizing—

A Suitable Alternative for the Heat
Treatment of Automotive Crankshafts
K. Bennett, BOC Ltd.
and
Q. Weir and J. Williamson, Leyland Vehicles Ltd.
The remarkable properties conferred on low-alloy steels by nitrocarbur-
izing have been highlighted comprehensively in the literature in the past
few years (Ref 4–6). In addition to enhanced fatigue resistance, the
increase in wear resistance, attributed to the formation of a thin compound
layer (Fig. 1) composed of the single-phase ε-nitride or carbonitride, has
been shown to be dramatic. Typically, a compound layer of some 12.5 to
20 µm has been reported as a result of a conventional 3 h nitrocarburizing
treatment at a temperature of 570 °C (1060 °F) (Ref 4).
Fig. 1 Typical compound layer on nitrocarburized mild steel
It has been suggested recently (Ref 7) that the properties obtained by

the 60/80 h gaseous nitriding of classical nitriding steels can be matched
by a 12/20 h nitrocarburizing treatment (Fig. 2). In this context, an added
advantage of nitrocarburizing would be the replacement of the brittle
duplex white layer, characteristic of gaseous nitriding, by the more ductile
single-phase ε compound layer. This would eliminate the frequent neces-
sity for postheat treatment machining.
Six-cylinder engine crankshafts, manufactured in B.S. 970: 1970
708A42 steel (En 19C), are normally nitrided for a process period of 60 h.
The service conditions in respect of the crankshaft journals are such that
while good wear resistance is required, a high indentation resistance is not
essential. A prime requirement exists for a good fatigue resistance.
Following consultation with BOC Ltd., Leyland Vehicles Ltd. decided
to initiate a program of work aimed at evaluating the properties to be
obtained by short-cycle nitrocarburizing (3 h duration in this instance) of
such automotive crankshafts. If it were possible to produce crank-shafts
with acceptable resistance to both wear and fatigue by such a treatment,
then the cost savings would be considerable relative to gaseous nitriding.
It was decided that the evaluation should be based upon surface hard-
ness tests, microhardness traverses, metallographic examination, and
fatigue tests. Such tests would be reinforced by both engine testing and
road trials on the actual components.
Process Technique
The geometry of the components is such that it was considered essential
to treat vertically in order to control distortion (Fig. 3). The only suitable
equipment available at Leyland for this purpose was a Wild Barfield pit fur-
Fig. 2 Comparative hardness profiles produced by 80 h nitriding and 16 h

nitrocarburizing
Fig. 3 Furnace load of crankshafts
nace which was normally used for gas carburizing. This is shown schemati-
cally in Fig. 4. Some reservations were expressed as the normal technique
of venting ammonia-bearing exhaust gases to atmosphere could not be
accommodated. The location of the furnace and the height of the heat treat-
ment shop presented additional problems in this respect. However, it was
Fig. 4 Schematic diagram of the Wild Barfield pit carburizing furnace used in
the nitrocarburizing trials
found in practice that the exhaust burnt quite readily during processing at
the nitrocarburizing temperature. The furnace was modified by the inclusion
of a pilot flame at the exhaust pipe. It was decided that the furnace atmos-
phere would be the standard nitrogen/ammonia/carbon dioxide nitrocarbur-
izing system (Ref 5–7). A further furnace modification was carried out to
facilitate the separate delivery of carbon dioxide to the furnace chamber. (It
is an accepted fact that, should ammonia and carbon dioxide be delivered
through the same manifold, then a reaction will occur between the two
gases. The manifold will quickly become blocked with heavy deposits of
ammonium carbamate.)
With a furnace volume of 2 m3 (70 ft3), the standard purge atmosphere
flow employed equated to six volume changes per hour. During process-
ing, the standard atmosphere flow was four volume changes per hour. The
full treatment details were as follows:
• Clean and degrease workpieces and jigs thoroughly.

• Load furnace standing at ambient temperature. Seal and purge with
nitrogen at purge flowrate for 1 h.
• Switch on heat after 1 h and reduce nitrogen flowrate to four volume
changes per hour.
• 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 process
flowrate 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 furnace
temperature.
• Reduce furnace temperature to below 400 °C (750 °F).
• Unload furnace and cool components either by oil quenching or slow
cooling.
It was found necessary in practice to reduce the furnace temperature

below 400 °C (750 °F) in order to control dimensional movement and to
avoid possible cracking in thin-wall areas. The cooling rate below 400 °C
(750 °F) would not appear to affect the metallurgical properties of the
treated components. It was evident, however, that the slower-cooled 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 h
nitrocarburizing were as discussed below.
Hardness. Surface hardness of the order of 580 to 650 HV5 were
achieved. Hardness traverses on sectioned samples indicated microhard-
nesses in the compound layer in excess of 700 HV0.01, with the highest
values being recorded close to the interface of the compound layer and the
substrate. Core hardness was 320 to 330 HV5.
Case Depth. Hardness profiles (Fig. 5) showed effective case depths
(to 500 HV) of 0.125 to 0.15 mm (0.005 to 0.006 in.).
Metallographic examination revealed a typical continuous unbroken
compound layer (Fig. 6). The depth of the compound layer was of the
order of 25 to 37.5 µm, which is considerably deeper than that expected of
a conventional nitrocarburizing treatment of the same duration on the
same material.
Fatigue Resistance. Fatigue testing of treated crankshafts was carried out
at the Motor Industry Research Association (MIRA). The results obtained
(Fig. 7) point to a fatigue resistance twice that of untreated components.
Wear Resistance. Crankshafts nitrocarburized in the 3 h treatment
were run in a six-cylinder turbocharged engine on an experimental test
bed in a 1884 h endurance trial. No wear was observed on the crank pins;
a maximum of 6 µm (0.00025 in.) wear occurred on the main journals.
Fig. 5 Typical hardness profile on a crankshaft after 3 h nitrocarburizing
Fig. 6 Compound layer on a crankshaft after 3 h nitrocarburizing
Fig. 7 Fatigue tests results

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 substantial
and comparable with that produced by conventional 60 h gas nitriding
of the same material. In view of the fact that the effective case depth
of the nitrocarburized component is approximately 50% of that of a
nitrided component, the wear performance can only be attributed to
the properties of the compound layer.
• The fatigue resistance obtained by 3 h nitrocarburizing is within speci-
fication for the automotive crankshaft.
• The adoption of a short-cycle nitrocarburizing treatment in this
instance would yield a saving in heat treatment costs of some 85%.
In addition it has been shown that the nitrocarburized surface facilitates

the use of a cheaper bearing than that required for the more friable surface
of the nitrided component.
ACKNOWLEDGMENT
This appendix was reprinted with minor changes from K. Bennett,
Q. Weir, and J. Williamson, Gaseous Nitrocarburising—A Suitable Alter-
native for the Heat Treatment of Automotive Crankshafts, Heat Treatment
of Metals, Vol 8 (No. 4), 1981, p 79–81. Reproduced by permission of
Wolfson Heat Treatment Centre.
REFERENCES
1. T. Bell, Gaseous and Plasma Nitrocarburizing, Heat Treating, Vol 4,
ASM Handbook, ASM International, 1991, p 425–436
3. J. Grosch, Heat Treatment with Gaseous Atmospheres, Steel Heat
Dekker, 1997, p 663–719
4. T. Bell, Ferritic Nitrocarburising, Heat Treatment of Metals, Vol 2,
1975, p 39–49
5. C. Dawes, D.F. Tranter, and C.G. Smith, Reappraisal of 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 of
Materials (Detroit, 6–8 May 1980), p 146–160

CHAPTER 21
Equipment for
Ferritic Nitrocarburizing
EQUIPMENT for the ferritic nitrocarburizing process is quite diverse,

covering salt bath furnaces, atmosphere furnaces, and plasma furnaces.
Each of these is examined in this chapter.
Salt Bath Furnace Equipment

Early on, ferritic nitrocarburizing was accomplished in salt baths.
Advantages of the salt bath furnace include:
• Simple design
• Fueled by either natural gas or electricity
• Easy to operate
• Excellent heat transfer and temperature uniformity
• Not capital intensive
Major disadvantages of the salt bath procedure have been cleanliness

and effluent disposal. Good housekeeping of the salt bath work area is
essential not only in terms of cleanliness but also personal safety.
Effluent disposal is a subject addressed by all of the process salt manu-
facturers, who now produce process salts with very low cyanide contents,
to less than 4 wt%. Although the salt still contains cyanide, decomposition
and subsequent neutralization of the cyanide effluent are much easier.
Salt bath equipment used for nitrocarburizing is essentially similar in
design to salt bath furnaces used for other processes. Although batch
installations are most common, semicontinuous and continuous 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 is
nonreactive with the process salt. The pot can be made entirely of titanium
or can be titanium lined to reduce the capital cost of the process bath
equipment.
Fig. 1 Internally heated salt bath furnace with immersed alloy electrodes and
ceramic tile lining
Fig. 2 Internally heated salt bath furnace with submerged graphite electrodes
and a modified brick lining for use with carburizing salts
Chapter 21: Equipment for Ferritic Nitrocarburizing / 233
Atmosphere Furnace Equipment

Atmosphere nitrocarburizing requires a furnace that is both gastight and
airtight to ensure safety of the operator and the equipment. Furnace design
is of paramount importance to prevent an ingress of unwanted oxygen.
The furnace can be either the cold-wall vacuum design for partial pres-
sure processing or the integral quench atmosphere design. It should be
noted that because ferritic nitrocarburizing takes place below the atmos-
phere design limits of an integral quench furnace, the furnace manufacturer
should be consulted. Integral quench furnace doors are typically safety
interlocked. On a normal integral quench furnace, the external door cannot
be opened at a chamber temperature below 760 °C (1400 °F). This feature
is to eliminate the risk of an explosion. Consult the furnace manufacturer
regarding low-temperature operations such as ferritic nitrocarburizing.
Plasma-Assisted Furnace Equipment

Plasma-assisted ferritic nitrocarburizing systems (Fig. 3) often differ
from plasma nitriding systems in terms of process vessel materials of con-
struction. This is because the equipment operates at process temperatures
up to 110 °C (200 °F) higher than those used in plasma-assisted nitriding.
Process temperatures can reach a maximum of 650 °C (1200 °F).
The principle of operation remains the same, however, with the process
conducted in a partial-pressure condition. A major difference is that the
plasma-assisted ferritic nitrocarburizing furnace usually will have an
Fig. 3 Typical plasma nitrocarburizing furnace and associated control system.

Courtesy of Plateg GmbH
additional source of oxygen for the subsequent oxynitride procedure,

which provides surface enhancement by deliberate surface oxidation.
Improvement in corrosion resistance is shown in Fig. 4 and 5.
Process Control of Surface Metallurgy

Because most steels can be treated by plasma-assisted ferritic nitrocar-
burizing, it is necessary to thoroughly understand the steel being
processed in relation to process capabilities. In the case of low-alloy
steels, there are insufficient alloying elements to accomplish any real 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 (other
than iron nitrides), iron nitride formation can be encouraged by supplying
Fig. 4 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. 5 Towing ball hitch treated by the oxynitride process (right) and
untreated (left). Both subjected to salt spray test. Courtesy of Plateg
GmbH
additional nitrogen. This is accomplished by increasing the nitrogen-to-

hydrogen ratio at the flowmeters (or mass flowmeters). The classical gas
nitriding ratio of nitrogen to hydrogen is:
2NH3 ↔ 2N + 3H2
or a ratio of 1 part nitrogen to 3 parts hydrogen.

A plain low-carbon steel will nitride at a conventional nitriding temper-
ature of around 500 °C (925 °F) but with a resulting hardness of only
about 35 HRC. Corrosion resistance will improve dramatically but not
surface hardness. A higher process temperature of approximately 580 °C
(1075 °F) should be considered, and temperatures can reach as high as
620 °C (1150 °F). Determination of the process temperature depends on
the steel analysis and the surface metallurgical requirements.
A plain low-carbon steel also requires a higher nitriding potential. This
is accomplished by simply increasing the process gas nitrogen-to-hydrogen
ratio by volume to a ratio of approximately 5:1 and up to 7:1. This will
result in surface hardness values of up to 700 HV (59 HRC). As previously
stated, the carbon source can come from the steel itself or by the addition of
a hydrocarbon gas to the process gas flow. In the case of a plain low-carbon
steel, there is insufficient carbon in the steel to promote formation of the
dominant ε-nitride phase; therefore, hydrocarbon gas (methane or propane)
must be added. The usual recommendation is approximately 2% by volume
of the total gas flow of nitrogen and hydrogen. This will be sufficient to 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 from
process monitoring points on the furnace. The computer will then respond
to any set point signal deviation and take whatever corrective action is
necessary (Fig. 6). To accurately control the gas flow, mass flow con-
trollers are used. A mass flow controller can control flow rates ranging
from 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 during
the process time. A general estimate is approximately 1.5 L/min/m3 of the
volume of the process vessel with a total work surface area of 22 ft2
(2 m2). Based on 1.5 L/min and given that there are approximately 4 L per
U.S. gallon:
90 L/h
1.5 × 60 min = = 22.5 gal by volume of process gas
4
This makes plasma-assisted ferritic nitrocarburizing cost competitive

when compared with the conventional gaseous procedure. The reason for
the lower process gas consumption is that gas is being consumed strictly
for plasma generation and is not being used as a “sweep gas” to prevent
gas stagnation within the process chamber. Gaseous nitriding and ferritic
nitrocarburizing processes often require an overpressure condition to
ensure gas uniformity throughout the process chamber, particularly when
considering blind holes.
In addition, shorter process cycles greatly improve equipment 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 carburizing
other than by using high temperatures of around 1040 °C (1900 °F).
The rate of solid-state diffusion of any element into the surface of any
steel is governed by the laws of the physics of diffusion. In other words,
the diffused element cannot go into the steel any faster than the laws of
physics will allow it to go. Many claims are made of very deep case
depths being accomplished with very short cycle times at process temper-
atures in the region of 580 °C (1075 °F). The problem here is that “case
depth” usually is not defined. Is the claim being made for total case depth
or effective case depth? What is meant by core hardness? Is it the core
plus 5 HRC points, or is it the actual core hardness itself? This is 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 the
metallurgist than has been previously thought. It would make great sense
to 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 time

and temperature for carburizing can serve as an approximate guide for
plasma-assisted ferritic nitrocarburizing (Ref 1). The guide is based on a
plain low-carbon steel, using the formula:
Case depth = K √t
where the case depth is in inches (for case depth in millimeters divide by
25.4), t is the time in hours, and K is the temperature factor given in Table 1.
The rate of nitrogen and carbon diffusion will begin to slow as the
alloying content of a steel is increased by the addition of chromium,
Table 1 Temperature factor for estimating case depth

Temperature
°C °F Temperature factor, K
495 925 0.00046
510 950 0.00056
525 975 0.00068
540 1000 0.00081
550 1025 0.00097
565 1050 0.00116
580 1075 0.00136
585 1085 0.00146
595 1100 0.00160
605 1125 0.00187
620 1150 0.00218
635 1175 0.00252
650 1200 0.00291
660 1225 0.00334
675 1250 0.00382
molybdenum, aluminum, tungsten, vanadium, manganese, and especially

nickel. It is difficult to say by how much the diffusion will be retarded
because of the many available alloying element variations. For a steel
containing all of these elements (not accounting for percentage 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 manner
into the process chamber.
The process gas must be carefully selected. Steam may cause problems
with electrical equipment such as power feedthroughs and control sys-
tems. Nitrous oxide or oxygen is preferred; nitrous oxide tends to be more
user friendly to valves and control systems than oxygen. The thickness of
the oxygen-bearing compound zone after treatment will be determined by
the time at temperature and the cooldown time. A typical part that has
been ferritic oxynitrocarburized is shown in Fig. 7.
The primary reason for oxygen treatment after ferritic nitrocarburizing is
to enhance surface corrosion resistance. The procedure is comparable to
Fig. 7 Milling cutter after ferritic oxynitrocarburizing. Courtesy of Plateg GmbH

Fig. 8 Shafts treated by the oxynitride process (right) and untreated (left). Both
subjected to salt spray test. Courtesy of Plateg GmbH
the black oxide type of treatment and will enhance the cosmetic surface
appearance of the steel component (Fig. 8). Just as with salt bath treat-
ments, the surface finish of the component will depend on the surface finish
of 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 salt
spray corrosion tests, exceeding 200 h by a considerable margin. Resis-
tance to the salt spray test will be determined by the resulting oxide layer
thickness.
The oxidization treatment following ferritic nitrocarburizing has almost
no cost attached to it, other than a portion of the amortization of the equip-
ment. Generally, there is little or no power consumption. The only other
associated expenses would be the cost of the oxidation process gas, fol-
lowed by furnace occupancy.
REFERENCES
1. Heat-Treating, Cleaning, and Finishing, Vol 2, Metals Handbook, 8th
ed., p 98

CHAPTER 23
Evaluating the Process
PROCESS EVALUATION begins just as for the traditional nitriding

methods. The same precautions must be taken so as to not disturb the sam-
ple face being investigated for case depth, case hardness profile, and
visual microscopic evaluation. Sample cutoff, surface pregrind, polish,
wash and rinse, and etch are conducted exactly as described in Chapter 16,
“Examination of the Nitrided Case.”
Case Depth Evaluation

Case depth can be measured by one, or all, of three methods:
• Total case depth to core hardness

• Total case depth to core hardness plus 5 HRC scale points
• Effective case depth to 513 HV hardness value
Each measurement method should be stated on the part drawing or by the

customer, and whoever processes the work must be very clear as to what
is required in terms of case depth.
The most effective and accurate method of measuring case depth is by a
microhardness traverse from the surface through the case of a sectioned,
preground, and polished sample. Optical examination via light micro-
scopy will show the case metallurgy, from which an assessment of the
case depth can be made, but this method is not completely accurate.
The load mass for the traverse will be determined by the accomplished
case depth to be measured. Shallow case depths should be measured with
a light load, up to 200 g. The load selection can go as low as 10 g; how-
ever, as the impression becomes smaller it will be become difficult to see
and measure with an older microscope eyepiece.
Case Hardness
It is strongly recommended that the case hardness not be measured
directly on the steel surface with a heavy load (for example, at 150 kg as
used with the Rockwell C scale). The indenter is likely to penetrate

through the formed case and into the core, thus giving a false hardness
reading. For a nitrided case depth from 0.5 to 1.0 mm (0.020 to 0.040 in.),
it is acceptable to use the Vickers hardness test system with a light load up
to 10 kg (Fig. 1).
The sample surface should be clean and free from decarburization and
must be presented squarely 90° to the central axis of the penetrator. Any
deviation from this will result in a false reading.
The microhardness tester is usually used for a case traverse at right
angles to the case (Fig. 2, 3). Microhardness testing requires a highly pol-
ished specimen surface in order to clearly see the diamond indenter
impression in the case (Fig. 4). The sample must be held firmly in an
appropriate holding device (Fig. 5).
What If the Formed Case Has Low Hardness Values?

There are two reasons for low case hardness; each is a result of the
process control. First, poor precleaning of the material surface can reduce
nitrogen diffusion into the steel. The steel surface must be clean and free
from contaminants such as chlorides, sulfides, silicon products, cutting
oils, and lapping compounds.
Fig. 1 Vickers hardness testing machine

Chapter 23: Evaluating the Process / 247
Fig. 2 Microhardness testing machine
Fig. 3 Microhardness testing system with features including cameras and

computer imaging. Courtesy of NewAge Industries
Fig. 4 Comparison of identations made by Knoop and Vickers indenters
Fig. 5 Typical fixtures used for holding and clamping workpieces for micro-
hardness testing
The low hardness generally indicates that the case has not been
processed to the correct depth specification. In other words, the case depth
is too shallow and the indenter punches through it. The remedy is to either
increase the case depth requirement or, if the case depth is correct, select a
lighter load for the hardness test.
The second reason is an imbalance of the furnace atmosphere. If the
gaseous method of ferritic nitrocarburizing is being used, check the gas
flow rates and relative volumes. Next, check the gas decomposition within
the furnace. This can be accomplished by a gas analyzer that measures the
carbon monoxide, carbon dioxide, and hydrogen contents and adjusts the
gas composition and flow rates accordingly.
Chapter 23: Evaluating the Process / 249
If using the salt bath method of ferritic nitrocarburizing, analyze the salt
composition for cyanide and cyanate composition. If the resulting analysis
shows an excess of carbonate or cyanate, regenerate the bath. This is 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 a
shallow case, then:
• The process cycle time is too short to accomplish the required case
depth (assuming that the gas flow rates are correct).
• The gas flow rates could be incorrect. Increase the nitrogen flow rate
to increase the nitriding potential of the process.
In the fluidized-bed method of ferritic nitrocarburizing, a shallow case

can occur due to low nitrogen volumes. Set the process gas flows 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 originating
from an external environmental source? If so, check the operating atmos-
phere or liquid for acidic content and adjust the process parameters
accordingly. The corrosion may also be generated by an uncontrolled salt
bath, gas mixture, or plasma condition. If this is the case, check that the
salt bath composition is within specified operating limits and adjust
accordingly. Check also that the gas dissociation and gas ratios are
within the required operating limits and that the process temperature is
correct. Lastly, check that the plasma conditions are correctly set in terms
of power density, voltage, amperage, pressure, gas ratios, and process
temperature.
Material selection may also play a role in initiating corrosion. Does the
selected material have the necessary corrosion resistance for the given
environment? Even with improved corrosion resistance imparted by the
nitrocarburizing treatment, a more corrosion-resistant base material may
have to be selected.
Distortion
Distortion is always a contentious subject. Despite the low nitrocarbur-
izing process temperature, there is no absolute guarantee that distortion
will not occur. Distortion can be classified as either shape distortion
(warpage) or size distortion (Ref 1).
Shape distortion originates from induced residual stresses from such
operations as forging, rolling, and machining and will result in twisting,
Melting
chemistry Annealing
variations temperature Rough machine Stress relieve
Distortion
Rolling Normalize Preheat Final machine Ferritic

or forging temperature treatment nitrocarburize
temperature
Fig. 6 Flow diagram showing factors contributing to distortion
bending, and ovality. The degree of distortion that occurs will be deter-
mined by the amount of residual stress present in the steel. The residual
stress can be in the steel as a result of mixed phases due to incomplete
transformation at the preharden and temper operation. The only way to
reduce the induced mechanical stress is to stress relieve prior to final
machining and prior to nitriding. To deal with the possible mixed phase
problem, resulting from pre-heat-treatment, it may be necessary to cryo-
genically treat the steel, followed by tempering to deal with any trans-
formed retained austenite to martensite. This will stabilize the steel prior
to the ferritic nitrocarburizing procedure if retained austenite was present.
Size distortion results from thermochemical treatments such as ferritic
nitrocarburizing, and from variations in the steel chemistry from heat to
heat. If the steel has been prehardened and tempered, the prior austenitiz-
ing temperature, as well as time at the austenitizing temperature, will
influence grain size (Ref 2).
The flow diagram in Fig. 6 presents factors that influence distortion
during final heat treatment. Remember, both ferritic nitrocarburizing and
nitriding will act as a stress-relieving process. If residual induced stress
has not been dealt with beforehand, then the procedure will act as a stress
relieve operation and distortion will occur. Steel will respond to heat from
any source.
REFERENCES
1. G.E. Totten, C.E. Bates, and N.A. Clinton, Handbook of Quenchants
and Quenching Technology, ASM International, 1993, p 443
2. G.E. Totten and M.A.H. Howes, Chapter 5, in Steel Heat Treatment
Handbook, Marcel Dekker, 1997, p 251–292
Index_Nitriding.qxd 10/1/03 3:37 PM Page 251

Index
A nitriding 35–36
for retorts 43
SAE 5115, salt bath nitrided 57(F)
Alloy steels salt bath ferritic nitrocarburizing 211, 211(F),
AISI 4140, microstructure 182(F), 183(F) 212(F), 216(F)
compositions 126(T) Case depth. See also Nitrided case; Surface
compound zone thickness 143(F), 144(F), hardness
145(F), 146(F), 148(F), 149(F), 182(F), automotive crankshafts 227
183(F) carburizing vs. nitriding 13(F), 14(F), 15–16,
hardness 35, 160, 161 15(F), 16(F)
salt bath ferritic nitrocarburizing 211, 212(F) determination of 68–69, 69(T), 136–137
salt bath nitriding 188 extrusion dies 155
Alloying elements, effect of 4–5, 7–8, 20–21, ferritic nitrocarburizing 245
35, 127–129, 127(F), 128(F), 161, 203(F), fluidized–bed nitriding 117(F), 118(T)
206 forging dies 154
American Gas Co. 3 gears 159, 160(F)
Ammonia. See also Gas nitriding ion ferritic nitrocarburizing 236–237, 237(T)
decomposition of 3, 23–27, 50–51, 77–78, ion nitriding 136–137
77(F), 117–118, 185–186, 220 microhardness testing 177, 245–248, 246(F),
leaks in retorts 44–46 247(F), 248(F)
safety precautions 46–47 temperature factor 69(T), 137(T), 237(T)
Applications 153–161 Case formation 31–37. See also Compound
Aubert, Pierre 6 zone (layer)
Automotive crankshafts, gaseous ferritic Central Alloy Steel Corp. 7
nitrocarburizing of 223–229 Compound zone (layer)
in automotive crankshafts 223, 224(F)
carbon, influence of 32
B control of 65–69, 82–83
dual phase formation 4(F), 31(F), 32(F),
Bell, Thomas 67, 201 33(F), 65(F), 66(F), 141(F)
Berghaus, Bernhard 9, 72 early studies 7
British standard nitriding steels 5, 5(T) in extrusion dies 83(F), 155–156
and Floe process 8–9, 34, 66, 67(F)
in forging dies 154
C in gears 159
kinetic studies 142–150
Carbon, and compound zone 32 and Melonite process 203–204, 204(F)
Carbon steels and sputtering 142–150
AISI 1015, nitrocarburized microstructure tests 66
35(F) thickness 32–34, 35, 36, 118(T), 142–150,
AMS 6470, nitride networking 40(F) 143(F), 144(F), 145(F), 146(F), 147(F),
AMS 6470, nitrided case 181(F), 182(F) 148(F), 149(F), 205(F), 220–221
hardness 35, 203(F) and two stage process 66, 67(F)

Contaminants. See Surface contaminants Etching 177–179

Copper Examination methods 167–183
deoxidized, for retorts 44
for selective nitriding 163–164
Core hardness F
automotive crankshafts 227
extrusion dies 155(F) Ferritic nitrocarburizing. See also Ferritic
forging dies 153, 154(F) oxynitrocarburizing; Gaseous ferritic
gears 159 nitrocarburizing; Ion ferritic nitrocar-
high-speed steel cutters 157 burizing; Salt bath ferritic
and preheat treatment 28–29 nitrocarburizing
Corner effect. See Nitride networking advantages 193–194, 198–199
Costs, various surface treatments 198(F) corrosion 248–249
Crankshafts, gaseous ferritic nitrocarburiz- costs 198–199, 198(F)
ing of 223–229 distortion 249, 250(F)
Cutters, high-speed steel 156–157 early history 195–198
Cutting fluids. See Surface contaminants equipment 231–239
hardness testing 245–248
process evaluation 245–250
D surface preparation 241–243
training 199–200
Dawes, Cyril 195 Ferritic oxynitrocarburizing 234, 234(F),
Decarburization. See Surface contaminants 238–239, 238(F), 239(F)
Degussa 54, 55, 196, 201 Fingerprints. See Surface contaminants
Deposition, by ion nitriding 76–77, 157 Firth Brown Steels 5
Die casting, and gaseous ferritic Fixtures. See Trays and fixtures
nitrocarburizing 222 Floe, Carl F. 9, 34, 66
Dies Floe process 8–9, 26. See also Two-stage
drawing, wear behavior 207(F) process
extrusion 81(F), 83(F), 155–156, 155(F), Fluidized-bed nitriding
156(F), 157(F), 165(F) furnaces 111–118, 112(F), 113(F), 114(F),
forging 153–155, 154(F), 213(F) 115(F)
Diffusion zone (layer) 4(F), 31–32, 31(F), gas dissociation 117–118
32(F), 35, 65(F), 203(F), 204–205 oxynitriding 116
Dilution process 3, 5, 67–68 Fry, Adolph 4–5, 6, 7, 8, 14, 20, 127
Distortion Furnaces
control of 18–20, 120 fluidized-bed nitriding 111–118, 112(F),
and ferritic nitrocarburizing 194, 249, 250(F) 113(F), 114(F), 115(F)
in gears 158–159 gas nitriding 8(F), 23(F), 39–51, 39(F), 41(F),
growth 19–20, 19(F), 27–28, 27(F), 120, 45(F)
121–122, 158–159 gaseous ferritic nitrocarburizing 224–227,
postmachining 122–123 226(F), 233
shape 18, 120, 121, 249 ion ferritic nitrocarburizing 233–237, 233(F)
size 18, 119–120, 249 ion nitriding 10(F), 84, 85(F), 89–109,
stock removal 122 89(F), 95(F), 95(T), 99(F), 107(F),
Double-stage process. See Floe process; 109(F)
Two-stage process salt bath ferritic nitrocarburizing 231, 232(F)
salt bath nitriding 55–63, 56(F)
E
G
Electroplating 163, 164
Enameling, for retorts 44 Gas ionization. See Ion nitriding
Equilibrium diagrams Gas nitriding. See also Nitriding
iron-carbon 17(F), 194(F) case crushing 187
iron-nitrogen 2(F) discoloration 186
Equipment. See Furnaces; Oxygen probes; exfoliation 186–187
Programmable logic controllers; furnace design 8(F), 23(F), 39–51, 39(F),
Sensors; Thermocouples; Trays and 41(F), 45(F)
fixtures furnace heating 47–49

Index / 253
gas circulation 40–41, 41(F) Ion ferritic nitrocarburizing

gas dissociation 50–51, 185–186 case depth 236–237, 237(T)
insulation 42–43 corrosion resistance 234, 234(F), 238–239,
leak prevention 44–46 239(F)
orange peel effect 187 early history 197–198
oxygen probe control 51 furnaces 233–237, 233(F)
process control and instrumentation 49–51 gas consumption 236
retort construction and maintenance 43–44 oxynitrocarburizing 234, 234(F), 238–239,
safety precautions 46–47 238(F), 239(F)
sensors 51 process control 234–235
stop-off procedures 163–164 surface cleanliness 243
surface contaminants 185, 186–187 trade names 202(T)
temperature control 40, 49–50 Ion nitriding
troubleshooting 185–187 advantages 85–86, 107–108
Gaseous ferritic nitrocarburizing of blind holes and cavities 75, 102–103,
applications 222–223 103(F), 165(F)
of automotive crankshafts 223–229 case depths 136–137
early history 195–196 cathode and anode 93–94
furnaces 224–227, 226(F), 233 cold-wall continuous dc system 84, 85(F),
gaseous supply 221 89–94, 89(F), 95(T), 100(F), 190, 190(F)
process development 219–220 compound zone 34, 68, 82–83, 142–150,
process principles 220–221 143(F), 144(F), 145(F), 146(F), 148(F),
safety considerations 223 149(F)
surface cleanliness 241–242 deposition techniques 76–77
surface properties 221–222, 227, 228(F) diffusion techniques 76
trade names 202(T), 220 early developments 9–11, 71–72
Gears 21, 81(F), 158–159, 158(F), 160(F), equipment 89–109
and ferritic nitrocarburizing 197–198
187, 222
furnace loading 105
General Electric Co. 10, 72
furnaces 10(F), 84, 85(F), 89–109, 89(F),
Glass, for retorts 43–44
95(F), 95(T), 99(F), 107(F), 109(F)
Glow discharge process. See Ion nitriding
of gears 158–159, 160(F)
Growth. See Distortion
glow discharge characteristics 73(F), 74–75,
74(F), 79(F), 96(F), 106(F), 189(F)
heating elements 91, 97(F), 103–104
H hollow cathode effect 102–103, 103(F)
hot-wall pulsed dc system 84, 85(F), 94–101,
Hardness. See Core hardness; Hardness 95(F), 95(T), 97(F), 99(F), 100(F), 191,
testing; Surface hardness 191(F)
Hardness testing of maraging steels 161
of automotive crankshafts 227, 228(F) masking 105–106, 164–165, 165(F)
hardness profiles 168, 168(F), 169(T), 228(F) nitride networking 80–81, 80(F), 81(F)
macrohardness 167 oxynitriding 86–87, 87(F)
microhardness 167, 177, 245–248, 246(F), plasma generation 84, 85(F), 90–91, 98–101
247(F), 248(F) process control 75–76, 84, 90, 105, 139–142
sample preparation 168–177 process gas flow 92, 139–142
Harris, F.E. 69 process gas ratios 77–78
Heat input requirement 48–49 process gases 83
High-speed steels process principles 72–73, 73(F), 77–80
compositions 126(T) of pure irons 160
cutters 156–157 sputter cleaning 104–105, 104(F), 164–165
Homerberg, V.O. 8 sputtering 142–150
Houghton Durferrit 202, 203 of stainless steels 129–136
stop-off procedures 105–106, 164–165, 165(F)
surface degradation 81
surface reactions 78–80
I surface stability 80
thermocouples 91–92
Imperial Chemical Industries 54, 195 troubleshooting 188–191
Insulation, for gas furnaces 42–43 vacuum pumps 92–93, 93(F), 94(F)
Interstitial diffusion 24, 25(F) work cooling 101–102

Irons N
cast, salt bath ferritic nitrocarburized 214,
215(F), 216(F)
ferritic nodular, salt bath nitrided 57(F) Nitralloy steels 5, 14, 20–21, 125, 126(T)
nitrided case formation 31–37, 33(F) Nitrex Metal, Inc. 68
pure 159–160 Nitride networking 5, 16, 34, 34(F), 40(F),
80–81, 80(F), 81(F), 154–155
Nitrided case. See also Case depth; Surface
hardness
J alloying elements, effect of 128(F)
examination methods 167–183
Jack, D.H. 67 on forging die 154(F)
Jeffries, Zay 6 structure of 4(F), 32(F), 35(F), 65(F), 181(F),
Jones, Claude 10, 72 182(F), 183(F)
Joseph Lucas Ltd. 195, 219 Nitriding. See also Floe process; Fluidized-bed
nitriding; Gas nitriding; Ion nitriding;
Salt bath nitriding; Selective nitriding
K advantages 13–22, 19(F)
applications 153–161
Ketcham, W.J. 6–7 current status of technology 11
Kinetic studies, of compound zone formation examination methods 167–183
142–150 furnace equipment and control systems
Klockner Ionen GmbH 9, 72 39–51
Kolene Corp. 55, 207, 208 historical background 3–11, 13–14
Krupp Steel Works 4, 5, 6, 14, 127 metallurgical considerations 1–2
microstructures 31–37, 32(F), 35(F), 181(F),
182(F), 183(F)
L process principles 23–29
process requirements 2–3, 14–21
Leyland Vehicles Ltd. 224 steels for 125–137, 153–161
Light microscopy. See Microscopic troubleshooting 185–191
examination U.S. vs. German processes 5–6
Lightfoot, B.J. 67 Nitriding potential 26, 67
Ludlum Steel Co. 8 Nitrocarburizing. See Ferritic
nitrocarburizing
Nitrogen diffusion 24–27, 33–34, 139–142
M See also Solubility limit
Nu-Tride process 207, 208–217
Machlet, Adolph 3–4, 6, 13, 14, 67
Maraging steels 160–161, 183(F)
Martin, Stuart 10, 72 O
Masking. See Selective nitriding
Mass spectrometry, for process gas control Oils. See Surface contaminants
analysis 140–141 Optical light microscopy. See Microscopic
Massachusetts Institute of Technology 8, 9, 66 examination
McQuaid, H.W. 6–7 Oxidation, resistance to 21
Melonite process 55, 202, 203–207 Oxygen probes 51
Metallographic examination. See Oxynitriding 86–87, 87(F), 116. See also
Microscopic examination Ferritic oxynitrocarburizing
Microhardness. See Hardness testing
Microscopic examination
of automotive crankshafts 227, 228(F) P
etching 177–180
microscope selection 180–181 Paint residue. See Surface contaminants
nitrided case microstructures 181(F), 182(F), Paints, stop-off 164–165
183(F) Paschen curves 74, 74(F), 96, 96(F)
sample preparation 168–177 Phase transformations 17(F), 18, 18(F), 24,
Microstructures, nitrided iron and steel 31–37, 26–27, 28–29
32(F), 35(F), 181(F), 182(F), 183(F) Photo spectrometry, for process gas control
Molding resins 171(T), 172(T) analysis 139–140, 140(F)

Index / 255
Piston rods, oxynitrided 87(F) wear resistance 206–207, 207(F), 215–216,

Plasma nitriding. See Ion nitriding 217(F)
Plasma-assisted chemical vapor deposition Salt bath nitriding
76–77, 157 advantages 54
Plasma-assisted nitrocarburizing. See Ion bath aging 55–58
ferritic nitrocarburizing bath maintenance 61–62, 188
Preheat treatment 28–29 bath replacement 58
Process chambers. See Retorts bath testing and analysis 58–61
Programmable logic controllers 39, 39(F), 49, early developments 9, 54
76, 101, 139–140, 140(F), 235, 235(F) equipment 55, 56(F), 58, 63
furnace types 56(F)
procedure 55–63
process types 54–55
Q–R safety precautions 62–63
salts 53–54, 58, 60–61, 187
QPQ process 55, 205(F), 207–208, 208(F), stop-off procedures 164
209–210, 209(F), 210(F), 211(F) surface contaminants 57, 59
Quench requirements 16–18 troubleshooting 187–188
Refractory firebrick, for retorts 44 Sample preparation
Resins, molding 171(T), 172(T) cutting 168–170, 170(F)
Retorts mounting 171–174, 171(T), 172(F), 172(T),
construction 43–44 173(T)
maintenance 44 polishing 176–177, 176(F)
sealing 44–46, 45(F) pregrinding 174–176, 174(F), 175(F)
Rocker arms, wear resistance 206–207, vapor degreasing 170, 179–180
207(F) Seals, and ammonia leaks 44–46, 45(F)
Ryzhov, N. 139 Selective nitriding 105–106, 163–165
Sensors 51
Sergeson, Robert 7–8
S Sintered steels 223
Society of Manufacturing Engineers (SME)
6, 14
Safety precautions
Solubility limit, nitrogen in iron 2, 139–142
for ammonia use 46–47 Sputter cleaning 104–105, 105(F), 243
etchants 179, 180(F) Sputtering, and ion nitriding 142–150
gaseous ferritic nitrocarburizing 223 Stainless steels
salt bath nitriding 62–63 corrosion resistance 36, 129, 134(F)
vapor degreasing 179–180 hardness 35, 129
Salt bath ferritic nitrocarburizing nitridability 129–137
corrosion resistance 208, 208(F), 209(F), 217 for retorts 43
dimensional stability 209 salt bath ferritic nitrocarburizing 214, 214(F)
early history 195–196, 196(F), 197(F) type 304, salt bath ferritic nitrocarburized
engineering properties 215–217 216(F)
fatigue strength 217, 218(T) type 316, salt bath ferritic nitrocarburized
furnaces 231, 232(F) 214(F)
low-cyanide 202–207 type 422, ion nitriding of 130–131, 132(F),
lubricity 216, 217(F) 134(F)
Melonite process 55, 202, 203–207 type 440A, ion nitriding of 131–134
metallurgical results 211–215, 216(F) type 440C, tempering curve 133(F)
Nu-Tride process 207, 208–217 type 630 (17–4 PH), ion nitriding of 135–136,
post treatment 207–208 136(F)
process control 210 Steels. See also Alloy steels; British standard
process parameters 201–202, 209–210 nitriding steels; Carbon steels; High-
QPQ process 205(F), 207–208, 208(F), speed steels; Maraging steels; Nitralloy
209–210, 209(F), 210(F), 211(F) steels; Sintered steels; Stainless steels;
surface cleanliness 241–242 Tool steels
surface hardness 206, 206(T) compositions 126(T)
Sursulf process 217 compound zone thickness 143(F), 144(F),
Tenoplus process 217 145(F), 146(F), 148(F), 149(F)
tensile strength 206, 206(T) nitrided case formation 31–37
trade names 202(T) selection 125–126

Stop-off procedures 163–165 Tenoplus process 217

Stress relieving. See Distortion Thermocouples 49–50, 91–92
Sturges, Derek 10, 72 Thin-film deposition 76–77, 157
Sulfinuz process 195, 201 Timken Detroit Axle Co. 6
Surface contaminants Tool steels
and case formation 36–37 compositions 126(T)
in ferritic nitrocarburizing 241–243 D2, microstructure 216(F)
in gas nitriding 185, 186–187 extrusion dies 155–156, 155(F), 156(F)
in salt bath nitriding 57, 59 forging dies 153–155, 154(F)
Surface hardness. See also Case depth; H13, microstructure 182(F), 216(F)
Nitrided case salt bath ferritic nitrocarburizing 211, 213(F),
and alloying elements 4–5, 7–8, 35, 127–129, 216(T)
127(F), 160, 161, 203(F), 206 salt bath nitriding 188
automotive crankshafts 227 Townsend discharge 74, 74(F), 96(F)
diffusion techniques compared 16(F) Transition zone 4(F), 31(F)
early results 14 Trays and fixtures
high-speed steel cutters 156–157 maintenance 44
maraging steels 161 for salt bath ferritic nitrocarburizing 197(F)
Melonite process 203–204, 203(F), 204(F), Troubleshooting 185–191
206, 206(T) Tufftride process 54, 55, 196, 201, 203
pure irons 160 Two-stage process 34, 66, 67(F). See also Floe
stainless steels 161 process
values 20–21
Sursulf process 217
U–Z
T University of Aachen 72
Vapor degreasing 36, 170, 179–180
Temperature Walsted, J.P. 8
control 49–50, 154–155 Wehnheldt, Dr. 9, 71, 72
factors 69(T), 137(T), 237(T) White layer. See Compound zone (layer)
requirements 14–16

David Pye - Practical Nitriding and Ferritic Nitrocarburizing (ASM)

Uploaded by

Copyright:

Available Formats

David Pye - Practical Nitriding and Ferritic Nitrocarburizing (ASM)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

David Pye - Practical Nitriding and Ferritic Nitrocarburizing (ASM)

Uploaded by

Copyright:

Available Formats

FM_Nitriding.

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

© 2003 ASM International. All Rights Reserved. www.asminternational.org

© 2003 ASM International. All Rights Reserved. www.asminternational.org

First printing, December 2003

Library of Congress Cataloging-in-Publication Data

Pye, David, 1939–

Printed in the United States of America

© 2003 ASM International. All Rights Reserved. www.asminternational.org

© 2003 ASM International. All Rights Reserved. www.asminternational.org

Can Plain Carbon Steel Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . 35

© 2003 ASM International. All Rights Reserved. www.asminternational.org

What Happens in the Ion Nitriding Process . . . . . . . . . . . . . . . . . . . . 77

Cold-Wall Continuous dc Plasma Nitriding . . . . . . . . . . . . . . . . . . . . . 89

Heating Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Size Distortion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Steel Selection Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

© 2003 ASM International. All Rights Reserved. www.asminternational.org

Can Stainless Steels Be Nitrided? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

© 2003 ASM International. All Rights Reserved. www.asminternational.org

PART 2 Ferritic Nitrocarburizing

© 2003 ASM International. All Rights Reserved. www.asminternational.org

© 2003 ASM International. All Rights Reserved. www.asminternational.org

© 2003 ASM International. All Rights Reserved. www.asminternational.org

colleagues at DeHavilland Propellors with whom I worked from 1956 to

© 2003 ASM International. All Rights Reserved. www.asminternational.org

1908 followed by Dr. Adolph Fry’s patent in 1922, it was considered in

Up to now, many of the answers have been determined mostly by per-

© 2003 ASM International. All Rights Reserved. www.asminternational.org

also requires individual ingenuity and dogged determination, tempered

Practical Nitriding and Ferritic Nitrocarburizing Copyright © 2003 ASM International ®

THE NITRIDING PROCESS, first developed in the early 1900s, con-

Metallurgical Considerations and

2 / Practical Nitriding and Ferritic Nitrocarburizing

The solubility limit of nitrogen in iron is temperature dependent, and at

Chapter 1: An Introduction to Nitriding / 3

Control of the process parameters is necessary to ensure formation of

• Total surface area to be nitrided

The Pioneering Work of Machlet

4 / Practical Nitriding and Ferritic Nitrocarburizing

Diffusion zone consisting

Transition zone from

Fig. 2 Schematic of a typical nitrided case structure

Parallel Work in Europe

Chapter 1: An Introduction to Nitriding / 5

Table 1 British standard nitriding steels

6 / Practical Nitriding and Ferritic Nitrocarburizing

Machlet’s process was not widely accepted in the United States, as it

Developments in the United States

• Influence of temperature on both case formation and case depth

Chapter 1: An Introduction to Nitriding / 7

• Effects of the ammonia/hydrogen relationship and dilution by hydrogen

8 / Practical Nitriding and Ferritic Nitrocarburizing

Door lift mechanism

Fan drive motor

Air circulating Refractory

Fig. 3 Schematic of a simple ammonia gas nitriding furnace

the nascent nitrogen diffusion if present in significant quantities. More

Other Early Developments

Chapter 1: An Introduction to Nitriding / 9