Stainless Steels for Design Engineers

Copyright © 2008 ASM International®. All rights reserved. Stainless Steels for Design Engineers (#05231G)

www.asminternational.org

STAINLESS STEELS FOR DESIGN ENGINEERS

MICHAEL MCGUIRE

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

Copyright © 2008 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 2008 Great care is taken in the compilation and production of this book, but it should be made clear that NO WARRANTIES, EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE GIVEN IN CONNECTION WITH THIS PUBLICATION. Although this information is believed to be accurate by ASM, ASM cannot guarantee that favorable results will be obtained from the use of this publication alone. This publication is intended for use by persons having technical skill, at their sole discretion and risk. Since the conditions of product or material use are outside of ASM’s control, ASM assumes no liability or obligation in connection with any use of this information. No claim of any kind, whether as to products or information in this publication, and whether or not based on negligence, shall be greater in amount than the purchase price of this product or publication in respect of which damages are claimed. THE REMEDY HEREBY PROVIDED SHALL BE THE EXCLUSIVE AND SOLE REMEDY OF BUYER, AND IN NO EVENT SHALL EITHER PARTY BE LIABLE FOR SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES WHETHER OR NOT CAUSED BY OR RESULTING FROM THE NEGLIGENCE OF SUCH PARTY. As with any material, evaluation of the material under end-use conditions prior to specification is essential. Therefore, specific testing under actual conditions is recommended. Nothing contained in this book shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters patent, copyright, or trademark, and nothing contained in this book shall be construed as a defense against any alleged infringement of letters patent, copyright, or trademark, or as a defense against liability for such infringement. Comments, criticisms, and suggestions are invited, and should be forwarded to ASM International. Prepared under the direction of the ASM International Technical Book Committee (2007–2008), Lichun L. Chen, Chair. ASM International staff who worked on this project include Scott Henry, Senior Manager of Product and Service Development; Steven R. Lampman, Technical Editor; Eileen De Guire, Associate Editor; Ann Britton, Editorial Assistant; Bonnie Sanders, Manager of Production; Madrid Tramble, Senior Production Coordinator; Diane Grubbs, Production Coordinator; Patty Conti, Production Coordinator; and Kathryn Muldoon, Production Assistant Library of Congress Control Number: 2008934669 ISBN-13: 978-0-87170-717-8 ISBN-10: 0-87170-717-9 SAN: 204-7586 ASM International® Materials Park, OH 44073-0002 www.asminternational.org Printed in the United States of America

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

METALLURGY

Chapter 1

Metallurgy........................................................................................................1 CORROSION AND OXIDATION

Chapter 2

Corrosion Theory............................................................................................11

Chapter 3

Corrosion Kinetics..........................................................................................19

Chapter 4

Corrosion Types..............................................................................................27

Chapter 5

Oxidation.......................................................................................................57 STAINLESS STEEL ALLOYS

Chapter 6

Austenitic Stainless Steels ..............................................................................69

Chapter 7

Duplex Stainless Steels...................................................................................91

Chapter 8

Ferritic Stainless Steels .................................................................................109

Chapter 9

Martensitic Stainless Steels ..........................................................................123

Chapter 10

Precipitation-Hardening Stainless Steels ......................................................137

PROCESSING

Chapter 11

Casting Alloys...............................................................................................147

Chapter 12

Melting, Casting, and Hot Processing...........................................................155

Chapter 13

Thermal Processing ......................................................................................161

Chapter 14

Forming........................................................................................................173

Chapter 15

Machining ....................................................................................................181

Chapter 16

Surface Finishing ..........................................................................................193

Chapter 17

Welding........................................................................................................201

Chapter 18

Architecture and Construction.....................................................................213

Chapter 19

Automotive and Transportation Applications................................................225

Chapter 20

Commercial and Residential Applications ....................................................233

Chapter 21

Marine Systems Applications........................................................................243

Chapter 22

Petroleum Industry Applications ..................................................................247

Chapter 23

Chemical and Process Industry Applications ................................................257

Chapter 24

Pulp-and-Paper Industry Applications ..........................................................265

APPLICATIONS

APPENDIXES

Appendix 1

Compositions...............................................................................................269

Appendix 2

Physical and Mechanical Properties of Select Alloys....................................279

Appendix 3

Introduction to Thermo-Calc and Instructions for Accessing Free Demonstration ....................................................................281

Index .................................................................................................................................285

iv

Preface The rate of growth of stainless steel has outpaced that of other metals and alloys, and by 2010 may surpass aluminum as the second most widely used metal after carbon steel. The 2007 world production of stainless steel was approximately 30,000,000 tons and has nearly doubled in the last ten years. This growth is occurring at the same time that the production of stainless steel continues to become more consolidated. One result of this is a more widespread need to understand stainless steel with fewer resources to provide that information. The concurrent technical evolution in stainless steel and increasing volatility of raw material prices has made it more important for the engineers and designers who use stainless steel to make sound technical judgments about which stainless steels to use and how to use them. This book provides design engineers with an up-to-date source of information at a level useful for both metallurgists and other engineers and technicians. It seeks to bridge the gap between the internet where much current, but raw information is available and scholarly books and journals that provide theory that is difficult to put into practice. The content of the book is selected for utility for the user of stainless steel. The first section gives elementary metallurgy and identification of constituents of stainless, the effects of alloying elements and a significant section on corrosion. A second section is oriented toward processes important to users of stainless steel. The third section is about each family of stainless alloys and includes the most recent additions that have come to the market. The fourth section deals in some depth with the major applications for stainless steel. This last part is presented without the promotional bias which is found in many steel producers’, alloy producers’, and trade associations’ literature. While a number of steel producers have provided assistance to the author, there has been no attempt to unfairly bias information in their favor. To the contrary, those producers responsible for generating factual, useful data for the user community are those who should benefit the most by books such as this. The author is particularly indebted to Allegheny Ludlum and John Grubb, and his many colleagues who assisted him, for technical assistance throughout the writing and to Carnegie Mellon University for their support. The author also wishes to thank Professor Sridhar Seetharaman at Carnegie Mellon University for his help in writing the corrosion chapter and others who helped: Roy Matway of CMU, Vittorio Boneschi of Centro-Inox; Paul Mason of ThermoCalc; Bob Drab of Schmolz Bichenbach; Elisabeth Torsner and Chuck Turack Outukumpu, USA; Scott Balliett of Latrobe Steel; Jim Halliday and Fred Deuschle of Contrarian Metals Resources; Professors Tony DeArdo of Pitt and Gerhard Welsch of CWRU; the staffs of Centro-Inox, Euro-Inox, SSNA, The Nickel Institute; and the editorial staff at ASM International, Scott Henry, Eileen DeGuire, Charlie Moosbrugger and Steve Lampman. I would also like to thank the many members of my forum at Eng-tips.com who have contributed much collective knowledge and perspective to this book.

ASM International is the society for materials

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

Product code

Stainless Steels for Design Engineers

#05231G

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

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

Email [email protected]

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

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

ASM International is the society for materials

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

Product code

Stainless Steels for Design Engineers

#05231G

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

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

Email [email protected]

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

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

Stainless Steels for Design Engineers Michael F. McGuire, p 1-10 DOI: 10.1361/ssde2008p001

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 1

Metallurgy Summary COMPARED TO ALLOY STEELS, stainless steels are chemically complex. The large number of alloying elements makes possible a larger range of possible phases or basic crystal structures. The large amount of the alloying elements makes the deviation from the behavior of pure iron greater; consequently, the calculations that predict which phases will exist are more difficult. The three basic phases of stainless steels are ferrite, austenite, and martensite. The wide variety of alloys that exist is based on: • Combinations of these phases • Altering the composition of these phases • Adding secondary phases for particular purposes Metallurgy, as discussed in this chapter, focuses on phases normally encountered in stainless steels and their characteristics. In subsequent chapters on types of stainless steel, there are more detailed treatments of the alloys made of these phases and their properties.

Introduction Most widely used alloy systems, such as carbon steels, alloy steels, and aluminum alloys, are relatively dilute solutions of several elements in the parent matrix. Carbon and alloy steels, with very few exceptions, are principally of the magnetic body-centered cubic (bcc) phase or a slightly distorted version of it. Aluminum alloys share the face-centered cubic (fcc) structure of pure aluminum. A given structure, which can have a certain range of compositions, is what is meant by a phase, just as a gas or liquid is a phase. In solid metals, there can be a number of

phases coexisting simultaneously. Stainless steel is an exceptional alloy system in that it is not a dilute solution. Alloy steels may contain several percent of alloying elements, such as carbon, manganese, nickel, molybdenum, chromium, and silicon, in addition to the impurities sulfur, oxygen, and phosphorus. Alloy steels typically contain very small amounts of titanium, niobium, and aluminum. The total amount* of these alloying elements seldom exceeds 5%. The same is true for most aluminum alloys. In contrast, stainless steels contain no less than about 11% chromium alone. Most stainless alloys have manganese, silicon, carbon, and nickel in thermodynamically meaningful amounts as well as large concentrations of nickel and/or molybdenum. The result of the large number of alloying elements in relatively high concentrations is that stainless steel can have many stable phases concurrently. In almost every case, having phases other than the principal one or two phases for which the alloy was designed is undesirable because of the possibility of undesirable variations in mechanical or corrosion performance. The producer of stainless steel controls the chemical composition and thermomechanical processing, so that when the processor or end user receives the product it is usually in the correct condition. However, subsequent processing or service conditions may alter the carefully established phase structure. Therefore, it is necessary to discuss the phases that can exist in stainless steel and the conditions under which they form so that the enlightened user will know which phases to avoid and how to avoid them. It is possible to use thermodynamics to calculate which phases may exist at a given tempera* All compositions are given in weight percent unless stated otherwise.

2 / Stainless Steels for Design Engineers

ture for a given composition. It is not remotely feasible, however, to give an adequate treatment of the thermodynamics required to do this. The topic alone requires a book. The necessary knowledge has been embedded in proprietary computer programs that will be used instead.

Thermodynamics of Stainless Steel Pure metals, from a practical viewpoint, are either liquid or solid depending on temperature, with the possibility of some trivial small gas vapor pressure. A law of thermodynamics is that the number of possible condensed (i.e., solid) phases equals the number of elemental constituents plus one. The solid has a crystallographic structure that may vary with temperature. Many metals have a less-dense bcc structure at high temperature and transform to a denser fcc structure at lower temperatures. Iron does this. Iron has the curious characteristic of transforming from fcc back to the low-density bcc at still lower temperatures. This is a result of the unpaired 3d orbital electrons (those that give rise to ferromagnetism) that are not given up as valence electrons, causing repulsive forces between atoms and requiring a more widely spaced structure. All thermodynamic properties are based on interatomic attractions. In metals, the metal atoms give up valence electrons to the entire mass. These electrons are of varying energy states and highly mobile. They are responsible for the ability of metals to conduct heat and electricity well. The attraction, the strength of the bond, is proportional to the charge difference and distance. The attraction determines such macroscopic properties as melting temperature, density, and elastic modulus. In this book, the main concern with thermodynamics is predicting which phases are present both at equilibrium and in the quite frequent metastable state. The prediction involves calculating the free energy of the various possible phases. The phase with the lowest energy is most favored, but others may have free energies that permit them to exist. The difference between these two is that the equilibrium state, that of the lowest free energy, may require atomic rearrangements to occur for equilibrium compositions to be reached on an atomic scale. If diffusion is too sluggish for these rearrangements to take place, the structure may retain the prior metastable structure indefinitely. This is

not a small, pedantic point. Most stainless steels are used in the metastable condition. For example, the common alloy 304 (also called 18-8) is normally used in the fully austenitic condition. It would “rather” be partly ferritic, but the substitutional diffusion of chromium in austenite that is required to form a ferrite phase of a separate composition is so slow that it cannot occur in terrestrial time frames. However, if energy is applied by mechanical shear, the austenite can transform without diffusion to the lower freeenergy martensite phase, a quasi-bcc structure of lower free energy. The calculation of which phases exist under equilibrium conditions proves to be extraordinarily difficult in complicated alloy systems. This is because thermodynamic values can be measured accurately only in the liquid state, so the values for the solid state are extrapolations. Also, the interaction between elements is very important in nondilute alloys such as stainless steel. Consequently, most published phase diagrams are experimentally derived. To determine which phases exist at a given composition and temperature, a sample is made, equilibrated at the appropriate temperature, and quenched to room temperature. It is assumed that the characteristic equilibrium phases have been frozen and are then identified by various techniques for structure, composition, and the like. This important work is obviously tedious and susceptible to experimental error and applies only to specific compositions. Any “what if” extrapolation to a different alloy composition carries the risk of error. A practical tool has been developed that permits phase diagrams to be calculated for arbitrary compositions. These are computer simulated, mathematical models that can perform the complex thermodynamic calculations. To do this with accuracy requires databases of thermodynamic values. These values must be derived from computer analysis of experimental phase equilibrium diagrams. They are expensive to derive and validate, and only a few exist. Hence, they are proprietary. In Appendix 3, a license to one such program, Thermo-Calc, can be found. The version has a reduced three-element capability but uses the same proprietary thermodynamic database of the full version. The program allows determination of which phases can exist for any composition and temperature. Whether the phases will form depends also on kinetic factors. First, however, it is good to become familiar with the principal phases found in stainless steel.

Chapter 1: Metallurgy / 3

Phases Ferrite The basis of stainless alloys is, of course, iron. Iron, as stated, solidifies as a bcc alloy before transforming to the denser fcc austenite at lower temperatures. At still lower temperatures, it reverts to the bcc structure. It is accurate to surmise from this that the free energy of both structures is close. Alloying elements that promote one structure over the other can therefore change which one predominates. The element that produces the ability to form the passive film that makes stainless corrosion resistant, chromium, has the characteristic of stabilizing the bcc structure. As chromium is added to iron, the temperature range over which austenite is stable grows smaller until, at about 12% chromium, ferrite is stable at all temperatures. This is, coincidentally, the approximate level of chromium needed to keep alloys from rusting under ambient conditions, but this effect is not related to whether the structure is bcc or fcc. The iron-chromium phase diagram (Fig. 1) shows the composition and temperature regions where ferrite (a), martensite (α' ), austenite (γ), and sigma phase (σ) are stable. While chromium is the principal ferrite-promoting alloying element, other elements have similar effects, but none produces the quality of stainlessness. Silicon, aluminum, molybdenum, tungsten, niobium, and titanium all favor ferrite. Carbon, nitrogen, manganese, nickel, and copper do not and expand the temperature range over which austenite exists. Elements that are insoluble in iron at austenite-forming temperatures, such as the impurities phosphorus, sulfur,

Fig. 1

The iron chromium phase diagram. Courtesy of Thermo-Calc Software

and oxygen, have no influence on which phase is favored. Again, it must be emphasized that the influence of an alloying element on structure has zero bearing on its influence on corrosion resistance. The elements that promote ferrite over austenite also have the effect, at still lower temperatures, of promoting intermetallic compounds generally composed of iron, chromium, and some of those alloying elements. These are discussed separately. Metals are effective solvents in both the liquid and solid states. An important part of steelmaking is refining the molten metal to remove the undesired impurities dissolved in it. The normal technique is to add elements that react selectively with the targeted impurities to form an immiscible reactant that can become part of the slag and physically separated from the refined alloy. This is done for the primary impurities oxygen and sulfur. A third common impurity, phosphorus, is not so easily removed and must be excluded from raw materials to be kept under control. In stainless steel, carbon and nitrogen can be detrimental impurities. Both are quite soluble in molten iron-chromium alloys and are fairly soluble in ferrite at high temperatures. This solubility decreases exponentially with temperature so that it is essentially zero at room temperature. These elements have small atomic sizes compared to iron and chromium and, when dissolved, squeeze into interstitial sites within the bcc matrix. Such interstitial solute atoms profoundly distort the structure. They are much more soluble in the fcc structure, which, while denser, has roomier interstitial spaces, so they stabilize that structure. To preserve the ferrite structure, carbon and nitrogen must be eliminated. There are additional reasons to eliminate carbon and nitrogen. During cooling as these elements become less and less soluble, they must precipitate. The most thermodynamically favorable form in which they can precipitate is as a compound of chromium, with which they are very reactive. This occurs at the grain boundaries, where nucleation is favored, and depletes those regions of chromium, rendering them less corrosion resistant. A second effect is a loss of toughness due to these precipitates. The diffusion rates of carbon and nitrogen in ferrite are too high to prevent this precipitation by quenching. Modern refining methods can reduce carbon plus nitrogen to under 0.020%, but even this is

4 / Stainless Steels for Design Engineers

too high. So, to avoid the detrimental effects of chromium carbide and nitride formation in ferrite, other benign carbides and nitrides such as those of titanium or niobium are allowed to form preferentially. This approach is called stabilization and is used for most ferritic alloys today. The older approach, as characterized by alloy 430, is to permit chromium carbides and nitrides to form but then to perform a subcritical anneal to rehomogenize the chromium and coarsen the precipitates so that they have only a small negative effect on mechanical properties. Hydrogen and boron are other elements that can be interstitially dissolved in ferrite. Boron is normally found at levels of around 5 to 10 ppm. At higher levels, boron substitutes for carbon in carbides. Hydrogen is soluble to several parts per million by weight. It does not cause hydrogen embrittlement in annealed ferrite. If the ferrite is cold worked, the solubility of hydrogen increases as the defect structure accommodates hydrogen atoms. In this condition, ferrite may be embrittled by hydrogen, especially if it enters the metal through corrosion processes like pitting. This is one explanation of, and the most likely explanation for, stress corrosion cracking. While hydrogen is easily removed by argon oxygen decarburization (AOD), assuming absolutely dry blowing gases and additions are used, it can be picked up during pickling, welding, or annealing as well as by corrosion. All stainless alloys rely on having a uniform level of chromium and the other element, molybdenum, which assists in corrosion resistance, distributed throughout the matrix. If there are locally low levels of these elements, localized resistance to corrosion is reduced, and localized corrosion can occur. This can occur by the precipitation of any phase that is richer in chromium or other corrosion-resisting elements. Because chromium is a reactive element, its success depends to a great degree on maintaining the homogeneity required for proper corrosion-resistant performance. Incorrect thermal processing is the main way homogeneity can be lost. Stabilizing makes it much easier to keep chromium from segregating in ferritic alloys. A by-product of stabilization with titanium is that oxygen and sulfur are also eliminated as compounds of titanium along with carbon and nitrogen. These impurity elements would otherwise also precipitate as compounds containing some chromium, potentially depleting chromium in the vicinity of their precipitation.

The bcc structure of ferrite allows more rapid diffusion than does the fcc structure of austenite. This is true for both the interstitial diffusion of the elements helium, boron, carbon, nitrogen, and oxygen and the substitutional diffusion of all other elements. The rate of diffusion of all elements, both interstitial and substitutional, in ferrite is about two or three orders of magnitude higher than in austenite. The practical implication of this is that precipitation reactions generally cannot be suppressed by quenching in ferrite if they involve interstitial elements, whereas they can be in austenite. Intermetallic phases can form more rapidly in ferrite. This becomes an issue only when total chromium plus molybdenum exceeds about 20%, above which the sigma phase appears. This is thus only an issue for superferritic (high chromium content) alloys or for the ferrite phase of duplex (ferrite-austenite) alloys. The mechanical properties of the ferrite phase are discussed extensively in Chapter 8, “Ferritic Stainless Steels.” Here, it is only necessary to note that ferrite in stainless steel closely resembles low-carbon steel in mechanical behavior. It shares the following characteristics: • A toughness transition that occurs around room temperature • Notch sensitivity • A yield point phenomenon • Pronounced crystallographic anisotropy of mechanical properties • High stacking fault energies and low workhardening rates These issues are dealt with in the same way as in carbon steel when these characteristics become an issue. The first two are controlled by reduction of interstitial levels and refining of grain size. The yield point is eliminated by slight elongation by temper rolling or elimination of interstitial carbon and nitrogen, whose interaction with dislocations causes the yield point. The anisotropy is either utilized to advantage by maximizing it, as in the case of deepdrawing alloys, or minimized by refining grain size and randomizing grain orientation by special thermomechanical processing. Ferrite has a greater thermal conductivity and lower thermal expansion than austenite. Its strength decreases with temperature more than that of austenite, but the good match in thermal expansion between the ferrite and its oxide still makes it an excellent high-temperature

Chapter 1: Metallurgy / 5

material. Ferrite has very nearly the same corrosion resistance as austenite, but since ferrite can hold no nitrogen in solution, it cannot benefit from this element. In duplex alloys, the ferrite is generally the more corrosion resistant phase because it is richer in chromium and molybdenum. Austenite The second major constituent phase of the stainless steel alloy system is austenite. Austenite has an fcc atomic structure. The fcc structure is common in many transition metals to the right of iron in the periodic table. As stated, the fcc structure should be considered normal for metals well below their melting temperature as it is a denser structure. The presence of the bcc structure relates to the unpaired 3d electrons, which provide ferromagnetism. Adding elements to iron that causes pairing of the 3d electrons diminishes ferromagnetism and promotes the fcc structure. Nickel and manganese are the most prominent alloying elements that do this, but the interstitials carbon and nitrogen are the most powerful austenite stabilizers on a percentage

Fig. 2

basis. Their use is limited by their solubility and their tendency to form precipitating compounds with chromium. Manganese acts largely through its ability to promote nitrogen solubility. Superaustenitic stainless steels, such as S34565, use 4 to 6 % manganese to permit nitrogen levels of 0.4 to 0.6% to be achieved, resulting in higher pitting corrosion resistance. Since all stainless steels contain principally iron and chromium, the addition of a substantial amount of austenitizing elements is necessary to transform the structure to austenite. As a rule of thumb, iron alloys require about 17% chromium and 11% nickel (or its equivalents) to remain austenitic at room temperature. One percent nickel can be replaced by about 2% manganese as long as nitrogen is present to maintain the same phase stability. The omnipresent carbon and nitrogen have an effect 30 times that of nickel, so even in the small amounts in which they are normally present, they have a significant effect. These stabilizing factors are mapped in the Schaeffler diagram of Fig. 2 (Ref 1), whose purpose is to predict the phase makeup of weld metal. Since welds solidify relatively rapidly, no carbides or intermetallic phases

Schaeffler-Delong constitution diagram showing phases present in as-solidified stainless steels at room emperature as a function of composition demonstrating carbon and nitrogen contributions to nickel effects. Adapted from A.L. Schaeffler, Constitution Diagram for Stainless Steel Weld Metal, Met. Prog., Vol 56, Nov 1949, p 680–688; and W.T. Delong, A Modified Phases Diagram for Stainless Steel Weld Metals, Met. Prog., Vol 77, Feb 1960, p 98

6 / Stainless Steels for Design Engineers

form, and only ferrite, austenite, and martensite will be present. Thus, they provide useful information about the compositional effects on phase development in nonequilibrium situations. The nickel equivalent (vertical axis) summarizes how nitrogen, carbon, and other elements combine to create a nickel-like effect. The horizontal axis does the same for chromium and those elements that have a similar effect. In most common stainless steels, austenite is normally present in the metastable state, for example, the retained austenite in alloy steels. Those with carbon above 0.02% would eventually break down into austenite plus carbides, and those with less than about 30% chromium plus nickel will form martensite if deformed sufficiently. But in the annealed state, the austenite in standard austenitic stainless steels will remain indefinitely as fully austenitic without precipitates unless heated above 400 °C (750 °C) for protracted periods of time or deformed extensively. Interstitial elements are much more soluble in austenite than in ferrite. Of these, only nitrogen is considered a beneficial alloying element. It both strengthens and improves the pitting corrosion resistance of austenite. Carbon has a parallel effect, but its tendency to form chromium carbides limits its use and in fact leads to its minimization in most alloys. Before the AOD was developed and carbon levels in stainless steels were higher, austenitic stainless steels were sometimes stabilized by titanium or niobium to counter the effects of carbon. Both carbon and nitrogen stabilize the austenite phase, permitting lower levels of nickel to be used in austenitic alloys. Interstitial atoms of carbon and nitrogen distort the fcc lattice, causing it to expand about 1% linearly per 1 wt% of solute (Fig. 3) (Ref 2). This produces solid solution hardening of the austenite. The work hardening of austenite is increased by nitrogen. A third interstitial solute, hydrogen, produces the same effect but to a lesser degree. Austenite is not embrittled by hydrogen to the extent ferrite or martensite is, but hydrogen does raise its flow stress and hardness while lowering its work-hardening rate. Sulfur and oxygen are considered impurities because they form inclusions, usually chrome/ manganese silicates and sulfides. If present in sufficient amounts, sulfur and oxygen precipitate as primary inclusions before or during solidification. In most austenitic stainless alloys, the remainder of these elements are near saturation in the as-solidified ferrite at very high temperatures

and then frozen in a state of supersaturation in the austenite when it forms on cooling. The sulfur and oxygen then precipitate during cooling or subsequent hot working as isolated inclusions. The interface between these inclusions and the matrix is the locus of corrosion pit initiation, quite probably because of chromium depletion occurring during and as a result of inclusion growth. When an alloy solidifies as austenite, sulfur immediately segregates to the grain boundaries because of its low solubility in austenite, and it forms a low-strength film with a low melting temperature. This causes poor hot workability and hot cracking of welds. The diffusion rates in austenite are quite low compared to ferrite, so even interstitial elements cannot move quickly enough to precipitate below about 400 °C (750 °F). This permits carbon and nitrogen to exist in very high degrees of supersaturation if introduced below this temperature, as is done by various proprietary processes. The low diffusion rates restrict such colossally supersaturated zones to thin surface layers, but they can reach phenomenal hardness of over Rc 70. The austenite structure does not discourage the formation of intermetallic compounds such as sigma, but it does, fortunately, make their formation very sluggish, as seen in Fig. 4. The difference of three orders of magnitude for carbide formation reflects the difference between the diffusion of carbon and that of substitutional elements. The formation of sigma in ferrite is about 100 times faster than in austenite. Sigma is almost never seen in commercial 316 alloys.

Fig. 3

Lattice expansions due to carbon. Source: Ref 2

Chapter 1: Metallurgy / 7

The mechanical properties of austenite are quite different from those of ferrite. Austenite is characterized by: • Low stacking fault energies leading to high work-hardening rates • Good toughness even at very low temperatures • Low notch sensitivity • Lack of a sharp elastic limit • Good high-temperature strength • Fairly isotropic mechanical properties While there is not a great deal of difference in the yield strengths of austenitic and ferritic alloys of similar alloy levels, austenitic alloys are more ductile, have high work-hardening rates, and therefore have higher tensile strengths. Austenite can be cold worked to extremely high strengths, around a maximum of 2000 MPa (290 ksi). Chapter 3, “Austenitic Stainless Steels,” gives a more thorough and quantitative treatment of the mechanical properties of austenite. In duplex stainless steels, a secondary austenite, γ2, can form from ferrite below 650 °C (1200 °F). At this temperature, it has the same composition as the ferrite from which it forms and is called type 1. In the 650 to 800 °C (1200 to 1470 °F) range, a range that can be encountered in the heat-affected zone (HAZ) at γ/δ boundaries during welding, another type forms. This so-called secondary austenite, γ2, type 2, is somewhat enriched in nickel over the ferrite from which it forms but poorer in nitrogen than the primary austenite, giving it poorer corrosion resistance. Secondary austenite can also coform with sigma as γ/δ grain boundaries are depleted of chromium. This secondary austenite is called type 3 and is also poor in chromium.

Fig. 4

Precipitation kinetics in 316 stainless steel. Source: Ref 3

The physical properties of austenite compared to ferrite include lower thermal and electrical conductivity and greater thermal expansion. It is also, of course, nonmagnetic. Martensite Martensite is a phase that forms from the diffusionless shear of austenite to a distorted cubic or hexagonal structure. This transformation can occur spontaneously on cooling or isothermally with externally applied deformation. It is essentially ferrite that has been formed with a supersaturation of carbon. The resulting structure is very fine and highly faulted, making it quite hard. As in carbon steel, the hardness of the martensite increases dramatically with interstitial content because of the huge strain interstitial elements impose on the bcc lattice, distorting it into tetragonality. Martensite in stainless steels is restricted to alloy levels at which austenite can form at higher temperatures but at which the austenite is unstable at ambient temperatures. This gives martensite a fairly narrow composition range. The lowest alloy level is that of the basic 12% chromium steels with 0.1 to 0.2% carbon. The most highly alloyed martensites are found in the precipitation-hardening grades. Thus, martensitic stainless steels are inherently limited in corrosion resistance to a level no better than a 17 or 18% chromium alloy and often barely qualify as stainless after the chromium tied up as chromium carbide is recognized as not contributing to the corrosion resistance. The as-formed martensite to the degree it has significant carbon content is hard and requires tempering to give it adequate toughness. The tempering reaction is the precipitation of carbon in the form of carbides with the concurrent loss of internal strain in the martensite lattice. The complexities of tempering require its discussion in detail to be found in Chapter 3, “Martensitic Stainless Steels.” It is worth noting, however, that all tempering involves carbide formation, thus losing some corrosion-fighting chromium. There are two forms of martensite, the ε, epsilon, and the α', alpha prime. Epsilon is formed in steels with low stacking fault energy, which are primarily the leaner austenitic alloys. Thus, it forms at cryogenic temperatures or by cold working. It appears in martensitic alloys of the precipitation-hardening type. It is nonmagnetic, has a hexagonal close-packed (hcp) structure,

8 / Stainless Steels for Design Engineers

and is very difficult to identify microscopically. The a' martensite is the familiar magnetic variety known in alloy steels that forms both by quenching and by deformation. The mechanical properties of stainless martensite are parallel to those of alloy steels. The high quantity of alloying elements in stainless give an extreme depth of hardening, so there is no concern with ancillary phases such as bainite. The physical properties are very close to those of ferrite of the same composition. Intermetallic Phases The number of phases that can coexist in an alloy is proportional to the number of alloying elements in the alloy. Table 1 lists data on the more common precipitates found in stainless steel. It is not surprising that stainless steel with iron, chromium, nickel, manganese, silicon, and often molybdenum, titanium, and niobium should have numerous ancillary phases. Intermetallic phases are normally hard and brittle. They can render the bulk alloy brittle when they form along grain boundaries. The other concern arising from intermetallic phase formation is the depletion from the surrounding matrix of Table 1 steels

Precipitated phases found in stainless

Precipitate

Structure

Parameter, A

Composition

NbC

fcc(a)

a = 4.47

NbC

NbN

fcc

a = 4.40

NbN

TiC

fcc

a = 4.33

TiC

TiN

fcc

a = 4.24

TiN

Z-phase

Tetragonal

a = 3.037 c = 7.391

CrNbN

M23C6

fcc

a = 10.57–10.68

Cr16Fe5Mo2C (e.g.)

M23(C,B)6 fcc

a = 10.57–10.68

Cr23(C,B)6

M6C

a = 10.62–11.28

(FeCr)21Mo3C; Fe3Nb3C; M5SiC

Diamond cubic

M2N

Hexagonal

a = 2.8 c = 4.4

Cr2N

MN

Cubic

a = 4.13–4.18

CrN

Gamma prime

fcc

a = 3.59

Ni3(Al,Ti)

Sigma

Tetragonal

a = 8.80 c = 4.54

Fe, Ni, Cr, Mo

Laves phase

Hexagonal

a = 4.73 c = 7.72

Fe2Mo, Fe2Nb

Chi phase bcc(b)

a = 8.807–8.878

Fe36Cr12Mo10

G-phase

a = 11.2

Ni16Nb6Si7, Ni16Ti6Si7

fcc

(a) fcc, face-centered cubic. (b) bcc, body-centered cubic.

chromium or molybdenum, causing localized lower corrosion resistance. Intermetallic phases form by diffusion of substitutional alloying elements, which makes their precipitation slower than that of carbides, but they can form in a matter of minutes in alloy-rich grades. Deformation, which enhances substitutional diffusion, accelerates their formation. The principal intermetallic phases are described next. Alpha Prime. Not to be confused with martensite, alpha prime is an ordered ironchromium phase (i.e., iron and chromium atoms occupy specific, rather than random, sites on two intersecting superlattices). This structure is quite brittle. It forms at relatively low temperatures, between 300 and 525 °C (570 and 980 °F). Before its true nature was understood, its presence was known through its causing the phenomenon called 475 embrittlement, originally called 885 °F embrittlement. This is sometimes confused with temper embrittlement, which occurs in the same temperature range but is caused by phosphide precipitation on prior austenite grain boundaries of martensite. Alpha prime precipitation can cause 475 embrittlement in ferritic or duplex stainless steels and limits their use in this temperature range but not at higher temperatures, at which the phase dissolves. This phase forms at chromium contents as low as 15%, but fortunately it takes a relatively long time to form, on the order of hours, so it will not occur inadvertently during thermal processing such as welding or annealing. Sigma. Sigma is a brittle tetragonal phase richer in chromium and molybdenum than either the ferrite or austenite matrix around it. It forms preferentially at ferrite-austenite boundaries in the temperature range 600 to 1000 °C (1110 to 1470 °F) in alloys with more than about 18% chromium plus molybdenum. Its composition is sometimes given as (CrMo)35 (FeNi)65, but examination of the iron-chromium phase diagram shows that it is archetypically an equiatomic iron chromium compound. It is strongly promoted by silicon and suppressed by nitrogen. Stabilized alloy grades show more rapid sigma formation than unstabilized alloy grades (e.g., 347 versus 304). In unstabilized alloys the prior precipitation of carbides destabilizes austenite, leading to subsequent sigma formation. This makes alloys like 310H, essentially 25Cr-20Ni, especially prone to sigma formation. Sigma forms much more rapidly from ferrite than from austenite because of the 100-fold

Chapter 1: Metallurgy / 9

higher diffusion rate of alloy elements in ferrite. This makes it a much larger issue in superferritic and duplex alloys, which have high chromium and/or molybdenum levels. Chapter 7, “Duplex Stainless Steels,” contains an indepth discussion of sigma. Chi. Chi, χ, is similar to sigma except it contains more molybdenum and less chromium and has a cubic structure. It can coexist with sigma and forms in the same temperature range. It also precipitates at ferrite-austenite boundaries and has the same deleterious effects. Laves Phase. The laves phase has the structure A2B where A is iron or chromium and B is molybdenum, niobium, titanium, or silicon. It forms at 550 to 650 °C (1020 to 1200 °F) over the course of hours. Thus, although its effect would be deleterious, it seldom becomes a practical problem. It is possible for it to form at temperatures below sigma and above alpha prime, but the long times for formation make it rare. Carbides, Nitrides, Precipitation Hardening, and Inclusions Carbon and nitrogen are very important in all steels, but they take on a special significance in stainless steel because chromium, the essential alloying element of stainless steel, reacts more vigorously with carbon and nitrogen than iron does. Except for its role in hardening martensite and strengthening austenite at high temperatures, carbon is almost universally a detrimental impurity from a corrosion point of view and is minimized. Its beneficial effect on corrosion resistance when it is in solution is negligible because so little of it can be held in solution. Nitrogen has a lesser tendency to form compounds with chromium, so it is considered a beneficial alloying element in austenite but not in ferrite, in which it has essentially zero solubility. Common carbide and nitride precipitating phases are also listed in Table 1. Carbides. M23C6 is the main carbide found in stainless steel. Its structure is orthorhombic, and it contains both iron and chromium. It can form at any temperature at which the host austenite or ferrite becomes saturated with carbon. It is mainly chromium carbide, but iron can substitute for chromium up to about 50%. Other elements, such as tungsten, vanadium, and molybdenum, can also dissolve in this carbide. The ratio of chromium to iron in the carbide increases with time and temperature, as chromium diffusion permits, up to a maximum of 4 or 5 to 1.

The precipitation of the carbide from ferrite occurs at grain boundaries, is extremely rapid, and cannot be suppressed by quenching. Less than 20 ppm carbon content is required to prevent its precipitation from ferrite, although up to 50 ppm can be effectively kept in solution by very vigorous quenching. From austenite, carbide precipitation occurs below about 900 °C (1650 °F) for carbon levels under 0.10% and at 650 °C (1200 °F) for carbon levels below 0.03%. For practical purposes, precipitation ceases below 500 °C (930 °F) due to the slowing diffusion of carbon. While carbon is essentially insoluble in austenite at room temperature, quenching can easily preserve up to 0.10% in supersaturation, as is commonly seen in type 301 stainless. The carbide precipitation occurs first at grain boundaries. The chromium that combines with the carbon comes from the matrix in the immediate vicinity and therefore decreases the chromium content of that region, giving rise to the phenomenon of sensitization, which comes from the original phrase “sensitization to intergranular corrosion.” Nickel and molybdenum decrease the solubility of carbon and thus accelerate the precipitation. Nitrogen retards precipitation. Cold work accelerates precipitation. The carbide has a hardness of about Rc 72. This makes the phase a useful constituent in wear resistance in martensitic alloys. In higher carbon grades such as the martensitic stainless alloys, additional, more carbonrich, carbides may form. These include M7C3 and M3C. The latter carbide forms during the low-temperature tempering of martensite, while the former precipitates at higher temperatures. Stabilizing carbides are those that are formed by the intentional addition of elements such as titanium and niobium. These elements form carbides of the type MC (metal carbide). The carbon in these compounds may be replaced by nitrogen or, in the case of titanium, sulfur. These carbides form preferentially over chromium carbides and thus prevent sensitization. They precipitate in both the liquid and solid states. In the solid state, the precipitate normally forms within grains. The Ti(CN) appears as a cube of gold TiN surrounded by gray TiC. The Nb(C,N) is less regularly shaped. They affect mechanical properties in ferrite both by their influence on recrystallization and by their ability to act as nucleation sites for brittle fracture Nitrides. At low levels, nitrogen can substitute for carbon in M23C6. At higher nitrogen

10 / Stainless Steels for Design Engineers

levels, Cr2N can form. This can occur in duplex alloys if they are heated to a solution annealing temperature at which the alloy has high solubility for nitrogen. Cooling from these temperatures can cause the excess nitrogen to precipitate as needles of Cr2N. Another nitride CrN can form in HAZs of welds. Precipitation-Hardening Phases. Phases that have a very similar lattice match to the parent phase can precipitate coherently, that is, without changing the continuity of the crystal lattice. In these cases, the slight mismatch causes a strain that can significantly restrict dislocation movement and thereby strengthen the matrix. One such precipitate is gamma prime, an intermetallic, ordered, fcc phase with the composition Ni3(AlTi). Copper forms the epsilon phase, essentially pure copper, which causes precipitation hardening. The secondary hardening of martensite due to the precipitation of molybdenum nitride or carbide is also a precipitation-hardening reaction. Inclusions. Inclusions are principally oxides and sulfides that form in the melt (type I), at the end of solidification (type II), or in the solid (type III). Type I inclusions are the largest and are globular. Except when they are deliberately kept to improve machinability, they are physically removed by various steelmaking practices. Type II inclusions form in interdendritic spaces as the solubility of oxygen and sulfur drop on solidification. Type III inclusions precipitate the remaining oxygen and sulfur, up to 100 ppm for normal manganese-silicon killed stainless steels, in the solid state either on preexisting inclusions or as micron-size particles. Inclusions are mainly oxides and sulfides of silicon and manganese. If more reactive elements, such as aluminum or titanium, are present, their oxides and sulfides can also be present. Sulfides and oxysulfides can be beneficial for machining as solid-state lubricants and chip breakers. Otherwise, their presence is detrimen-

tal as inclusions have been shown to be the initiation sites for corrosion pits, which have been linked to both their sulfur ions disrupting the passive layer and their chromium content causing slight local chromium depletion.

Properties of Stainless Steels Physical and mechanical properties of representative stainless steel alloys are summarized in Appendix 2. Properties are also discussed in chapters specific to each alloy family. The reader is referred to primary sources, such as company web sites, such as Ref 4 and 5.

REFERENCES

1. D.J. Kotecki, Welding of Stainless Steels, Welding, Brazing, and Soldering, Vol 6, ASM Handbook, ASM International, 1993, p 677–707 2. G.E. Totten, M. Narazaki, R.R. Blackwood, and L.M. Jarvis, Failures Related to Heat Treating Operations, Vol 11 ASM Handbook, ASM International, 2002, p 192–223 3. High Performance Stainless Steels, Reference Book Series 11 021, Nickel Development Institute, p 16 4. ASM Handbook, Vol 1, Properties and Selection, ASM International, 1990 5. ASM Speciality Handbook, Stainless Steels, ASM International, 1996

SELECTED REFERENCES

• D.J. Kotecki and T.A. Siewert, WRC 1992 Constitution Diagram, Welding Journal, Vol 5, 1992, p 171s–178s

Stainless Steels for Design Engineers Michael F. McGuire, p 11-18 DOI: 10.1361/ssde2008p011

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 2

Corrosion Theory Summary

Electrochemical Reactions

THIS CHAPTER INTRODUCES THE fundamentals of electrochemical theory as it pertains to corrosion. Topics covered include an overview of electrochemical reactions, Faraday’s law, the Nernst equation, galvanic versus electrochemical cells, and Pourbaix diagrams. The examples provided relate these fundamentals to the corrosion resistance of stainless steels.

In electrochemical reactions, charge is transferred across interfaces of species of different chemistry. Consider, for example, the reaction:

Introduction Corrosion—the environmental degradation of materials through electrochemical reactions—is a key subject for more or less all classes of alloys that fall within the broad definition of stainless steels because these alloys were developed with the intention of preventing corrosion. This chapter aims first to provide an introduction to the fundamentals of electrochemical theory as it pertains to corrosion. Thermodynamics are presented in light of electrochemical potentials as opposed to purely chemical ones. Chapter 3 introduces the formal terms needed to describe electrode reaction kinetics. Chapter 4 describes the various forms of corrosion and how they are related to alloy metallurgy, chemistry, and structure. Chapter 5 focuses on oxidation. For an in-depth study of electrochemical kinetics and electroanalytical methods, Ref 1 is recommended. For a broader study of corrosion, the reader is referred to texts by Jones (Ref 2), Uhlig and Revie (Ref 3), and Fontana (Ref 4) and to ASM Handbook, Volume 13A (Ref 5).

2Fe (s) + O 2 (g) + 2H 2 O → 2Fe 2+ + 4 OH −

(Eq 1)

An inspection of this reaction suggests that three phases must be present for the reaction to proceed: an ion-conducting phase (water-based solution), a metallic phase (iron), and a gas phase O2(g). Second, electrons have been transferred from the metallic phase, iron to O2 + H2O. Figure 1(a) shows the arrangement of an experimental setup in which Reaction 1 could proceed. On the left, iron is allowed to dissolve according to: 2Fe (s) → 2Fe 2+ + 4 e −

(Eq 2)

resulting in Fe2+ ions that dissolve in the water-based solution and electrons that are carried to the right side, where they participate in the reaction: O 2 + 2H 2 O + 4 e − → 4 OH −

(Eq 3)

Inside the water-based solution, ions (Fe2+, OH⫺, H+, or any others) migrate, thereby constituting a so-called ionic current. This current together with Reactions 2 and 3 and the transport of electrons from left to right form a closed circuit called an electrochemical cell. The cell is made up of four parts: the two electrodes where the charge transfer Reactions 2 and 3 take place

12 / Stainless Steels for Design Engineers

often are described as half cells, for example, Fe / O 2 / OH − and Fe / Fe 2+ .

Faraday’s Law If the cell in Fig. 1(a) was allowed to proceed and thermodynamics favored to proceed according to the direction in Reaction 1, then a current i will flow from the anode to the cathode, and the amount of charge passed per unit time as a result of this current will be linked to the amount of iron dissolved per unit time or the amount of oxygen reacted per unit time by virtue of Eq 2 and 3. This is given by Faraday’s law: nNF = it

(Eq 4)

Here, i * t is the charge passed (in coulombs); N is the moles of consumed/produced specie (e.g., moles consumed iron in Reaction 2); n is the ratio of electrons to consumed/produced species, which in the case of Reaction 2 will be 2; and F is Faraday’s constant, which is essentially the charge in coulombs corresponding to 1 mole of electrons. Fig. 1

Schematic illustration of (a) a differential aeration cell involving iron dissolution and (b) the same cell with a variable resistor and voltmeter

(the anode and cathode, respectively), an electrolyte, and an electron pathway. It should be noted that electrodes are interfaces that require several phases to be in contact. Oxidation, Reaction 2, occurs at the anode and reduction, Reaction 3, occurs at the cathode. The electrolyte is the medium through which the ions migrate; in the case of corrosion reactions, this is most commonly a water-based solution, but at high temperatures it could be a solid oxide. The final constituent of the electrochemical cell is a pathway through which electrons can migrate from the anode to the cathode. As a shorthand notation, electrochemical cells are written by separating components within a phase by a comma and separating phases by a slash; gaseous species are written next to their conducting electrode. For example, the cell described in Fig. 1(a) would be recorded as Fe / O 2 / OH − , Fe 2+ / Fe . This cell is an example of a differential aeration corrosion cell, which is discussed later. Processes at a single electrode

The Nernst Equation Electrochemical reactions require a transfer of charge; hence, there is a coupling between chemical and electrical energy. Consider the hypothetical setup in Fig. 1(a) with the addition of a variable resistor and a voltmeter, resulting in the arrangement shown in Fig. 1(b). Thermodynamically, the Gibbs free energy of the cell is that of Reaction 1: ΔG = ΔH − T ΔS

(a ) (a ) + RT ln 4

= ΔG

0

OH −

Fe 2+

aH O PO 2

2

(Eq 5)

2

where ⌬G is the Gibbs free energy, H is the enthalpy, S is the entropy, R is the gas constant, and T is the absolute temperature. If this is negative, the reaction would be expected to proceed spontaneously as written in Reaction 1. Let us assume that this is the case. The thermal heat produced by the system can be divided into two parts: the thermal heat produced by the cell Qt and the heat produced at the resistor QRes. QRes

Chapter 2: Corrosion Theory / 13

in this case is heat, but in essence it represents the available energy or work, which in the case of a resistance is given by the product of charge passed times potential difference. If the resistance approached infinity ( R→ ), Reaction 1 would proceed through infinitesimal steps and can be considered thermodynamically reversible. In this case, the thermal heat produced by the cell is minimized and according to thermodynamics is given as Qt = Qrev = T⌬S1. On the other hand, the net work gained QRes is maximized and constitutes the rest of the free energy: QRes = ΔG = ΔH − T ΔS

(Eq 6)

As mentioned, the energy dissipated through the resistance is charge passed times potential difference, and in this case the potential difference is the reversible potential difference E; thus, in an absolute sense: ΔG = nFE

(Eq 7)

Here, n is the number of electrons passed per atom of iron reacted, and F = 96,485 C per mole electrons, is Faraday’s constant. The reversible potential difference E represents the potential difference between the two electrode reactions (cathode and anode), and as such they are associated with Reaction 1 rather than a physical cell. The potential difference is referred to as the electromotive force (emf) of the cell. It is also referred to as the open circuit potential because it is the potential measured by the voltmeter in Fig. 1(b) when a negligible current flows. It is defined here as Erxn. By convention, this potential is positive for a spontaneous reaction (as opposed to the chemical free energy, which is negative); hence, Eq 7 becomes: ΔG = − nFErxn

(Eq 8)

and if all elements have unit activities: 0 ΔG 0 = − nFErxn

(Eq 9)

Equation 8 is the Nernst equation. By virtue of Eq 8 and 9 and the expression for Gibbs free energy of a reaction (e.g., Eq 5), an expression for Erxn is obtained: RT ⎛ p1 1 p2 2 p3 3 ... ⎞ ln ⎜ ⎟ nF ⎝ r1β1 r2β2 r3β3 ... ⎠ α

0 Erxn = Erxn −

α

α

(Eq 10)

Here, pi and ri are the concentrations of reactant and products, respectively, and αi and βi are the numbers that are needed to balance the reaction stoichiometrically. In the case of Reaction 1, Eq 10 would be: 0 Erxn = Erxn −

(

) ( ) ⎞⎟

4 ⎛ RT ⎜ aOH− aFe 2+ ln 4F ⎜ aH O PO 2 2 ⎝

2

⎟ ⎠

(Eq 11)

If the emf according to Eq 11 is positive, this means that the free energy is negative (according to the Nernst equation); hence, the net reaction is thermodynamically favored as it is written in Reaction 1. By inspection of Eq 11, it can be seen that it is the difference between two hypothetical half reactions, ( Erxn = EO /OH− − EFe 2+ /Fe ) 2 defined as: EO

= EO 0

2

/OH −

2

/OH −

)

(

4 ⎛ ⎞ RT ⎜ aOH− ⎟ − ln 4 F ⎜ aH O PO ⎟ ⎝ 2 2⎠

(Eq 12)

which corresponds to the reduction Reaction 3 and: EFe 2+ /Fe

)

4 RT ⎛ ( aFe ⎞ ⎟ ln ⎜ = EFe 2+ /Fe − 4 F ⎜⎝ aFe 2+ ⎟⎠ 0

(Eq 13)

which corresponds to the reverse of Reaction 2, that is, if it was a reduction reaction. The potentials as written in Eq 12 and 13 are called reduction potentials, and because Erxn = EO /OH− − EFe 2+ /Fe has to be positive for the 2 reaction to be thermodynamically favored as written in Eq 1, the reduction potential E O 2 /OH− has to be larger than E Fe 2+ /Fe. If it was not, then Reaction 1 would proceed in the reverse direction, which means that the electrode Reactions 2 and 3 would be reversed and thus so would the anode and cathode of the cell. It is useful to list reduction potentials for halfcell reactions, just as it is useful to list free energy data. However, half-cell potentials (like any electrical potentials) cannot be measured in an absolute sense; only potential differences can be measured. ( Erxn = EO /OH – − EFe 2+ /Fe can be 2 measured because it is a difference.) Therefore, half-cell potentials are measured with respect to a reference electrode. Reference electrodes are constructed such that they have a stable potential; this is discussed further in Chapter 3. A common reference electrode in aqueous solutions is

14 / Stainless Steels for Design Engineers

the normal hydrogen electrode (NHE), also known as the standard hydrogen electrode (SHE), with a potential set (arbitrarily) as zero at all temperatures. The NHE is schematically shown in Fig. 2. In shorthand notation, it is: Pt / H 2 (a = 1)/ H + (a = 1) , and the half-cell reaction is: 2 H + + 2 e− = H 2

(Eq 14)

Table 1 (Ref 6) lists half-cell reduction stan0 dard potentials ( EOx/Re) versus NHE that are a result of the emf of the following types of cells (for Reaction 2, as an example): Pt / H 2 (a = 1)/ H + (a = 1), Fe 2+ (a = 1) / Fe

Galvanic versus Electrochemical Cells When reactions in a cell occur spontaneously in the direction dictated by the open-circuit potential of a cell that is positive ( Erxn > 0) , a current flows as shown in Fig. 3(a). This is the case in environmentally caused electrochemical corrosion reactions. It also is the case in fuel cells and batteries (under discharge), in which the current is used as electricity. These types of cells are called galvanic cells, in which chemical energy is converted to electrical energy. Most of the discussion in the following chapters concerns these types of cells. In electrolytic cells (Fig. 3b), an imposed electrical potential counters the “natural” cell potential to drive a reaction in a desired direction. These types of cells are used for many metallurgical processes, such as electroplating, electrorefining and electroextraction (e.g., the Hall-Heroult aluminum smelting cell), and for other applications, such as charging batteries. In the case of corrosion, the principle is used for protection against corrosion. In electrolytic cells, electrical energy is converted to chemical energy.

Table 1 Standard half-cell reduction potentials versus the normal hydrogen electrode Reaction

Fig. 2

The normal hydrogen electrode (NHE)

Fig. 3

Schematic of (a) galvanic cell and (b) electrolysis cell

Standard half-cell reduction potential vs. NHE(a) (V)

Fe 3+ + e− = Fe 2 +

0.771

O 2 + 2H 2 O + 4e − = 4 OH – (pH = 14)

0.401

2 H + + 2 e− = H 2

0.000

Ni 2 + + 2 e− = Ni

–0.250

Fe 2+ + 2e− = Fe

–0.447

Cr 3+ + 3e− = Cr

–0.744

2 H 2 O + 2 e− = H 2 + 2 OH − (pH = 14 )

–0.828

(a) NHE, normal hydrogen electrode. Source: Ref 6

Chapter 2: Corrosion Theory / 15

Corrosion Tendency The tendency to corrode, that is, whether a system consisting of anode, cathode, and electrolyte can react thermodynamically, is determined by evaluating Erxn. If this is positive, then there is thermodynamically a possibility for corrosion. The rate of corrosion, which is in most cases determined by corrosion kinetics, is discussed in Chapter 3. Consider, for example, a case of iron in aerated water. Figure 1 (with electrode Reactions 2 and 3) can be viewed as an idealized equivalent cell for this situation. It should be noted, however, that the locations of anode(s) and cathode(s) on the iron surface cannot be identified with ease. At room temperature (298 K), 1 atm oxygen partial pressure, and using Table 1, Eq 12 can be written by assuming unit activity for water and unit activity coefficient for OH−: EO

2 / OH

–

= 0.401 + 0.059 pOH = 0.401 + 0.059(14 − pH ) = 1..227 − 0.059 pH V (vs. NHE)

EFe 2+ /Fe = −0.447 − 0.0295 log(aFe 2+) = –0.624 V (vs. NHE)

(Eq 15)

(Eq 16)

Figure 4(a) shows a schematic plot of the two reduction potentials (Eq 15 and 16) versus pH. Because a spontaneous reaction requires Erxn to be positive, if the only pertinent reactions were Eq 2 and 3, this means that corrosion (due to iron dissolution to Fe2+ and oxygen reduction) is possible when the line representing EO2/OH− (Eq 15) lies above the line representing EFe / Fe (Eq 16). This is indicated by the region shaded in gray in Fig. 4(a). Hydrogen reduction is another possible cathode reaction in water: 2+

2 H + + 2 e− → H 2

Here, the following definition of pH has been used: pH = –log CH+, pOH = –log COH− and pH + pOH = 14. Similarly, the iron dissolution Reaction 2 will have a reduction potential accord-

Fig. 4

ing to Eq 13, which, assuming a Fe2+ activity of 10−6 (this is an arbitrary value but is usually taken to represent a low ion concentration), becomes at room temperature (using Table 1 for the standard potential):

(Eq 17)

and its reduction potential is (using the definition of pH): EH+ / H = EH0 + / H − 2

2

RT PH2 ln nF aH2 +

= 0 − 0.059 pH V vs. NHE

Reduction potential versus pH for iron and (a) oxygen gas reduction and (b) hydrogen ion reduction

(Eq 18)

16 / Stainless Steels for Design Engineers

Figure 4(b) shows the condition in which corrosion under deaerated conditions (due to iron dissolution to Fe2+ and hydrogen ion reduction) is possible as a gray shaded region. In Fig. 4(a) and (b), the regions where iron is stable are denoted as immunity (corresponding to immunity from corrosion). When comparing these two figures, it is noteworthy that hydrogen ions are able to cause corrosion only under relatively low pH conditions, whereas oxygen gas is able to corrode iron in the entire pH range.

The Construction of Pourbaix Diagrams Figures 4(a) and (b) are types of phase diagrams that show the stable phases in an area bounded by pH and potential. In reality, several electrochemical and chemical reactions need to be considered when constructing these types of diagrams. Each reaction is represented by a line. In the case of iron, the following chemical reactions will have to be considered (the pH dependency of these reactions is listed next to them [Ref 7] and since they are not electrochemical, they are evaluated from the equilibrium constants): Fe 2 + + 2 H 2 O = Fe(OH )2 + 2 H + ,

(

pH = 6.65 − 0.5 log aFe 2 +

Fig. 5

)

(Eq 19a)

Fe(OH)2 = HFeO 2− +H + ,

(

pH = 14.30 + log aHFeO− 2

)

(Eq 19b)

Fe 3+ + 3H 2 O = Fe(OH)3 +3H + , pH = 1.613 – (1/3) log(a Fe3+ )

(Eq 19c)

Since these are independent of potential, they will appear as vertical lines (see lines 19a to 19c in Fig. 5a). The following pH-independent electrochemical reactions need to be considered, and they will result in horizontal lines (Fig. 5a): Fe 2+ + 2 e− → Fe,

(

EFe 2+ /Fe = −0.447 + 0.0295 log aFe 2+

)

HFeO 2− + H 2 O = Fe ( OH 3 + 2e− ,

(

)

(Eq 20a)

EFe(OH) / HFeO2– = −0.810 − 0.0591 log aHFeO– 3

2

) (Eq 20b)

The following electrochemical reactions will depend on pH and thus will be sloped depending on this dependence (Fig. 5a).

Pourbaix diagram for iron. (a) Schematic matching Eq 19 to 21 in text to lines. (b) Actual complete diagram. Source: Ref 7

Chapter 2: Corrosion Theory / 17

)

Fe + 2H 2 O = Fe ( OH 2 + 2H + + 2e− , EFe(OH)

2 / Fe

= −0.0470 − 0.0591 pH

(Eq 21a)

Fe + 2H 2 O = HFeO −2 + 3H + + 2e− , EHFeO- /Fe = 0.495 − 0.0886 pH 2

(

+ 0.0295 log aHFeO– 2

)

(Eq 21b)

)

Fe 2+ + 3H 2 O = Fe ( OH 3 + 3H + + e− , EFe(OH)

3 / Fe

2+

= 1.057 − 0.1773 pH

( )

− 0.0591 log aFe 2+

(Eq 21c)

Fe(OH)2 + H 2 O = Fe(OH)3 + H + + e− , EFe(OH)

3 /Fe(OH)2

= 0.271 − 0.0591 pH

Fig. 6

Pourbaix diagram for chromium in water. Source: Ref 8

(Eq 21d)

For the pH-dependent reactions (chemical and electrochemical), one can readily label the regions depending on what iron species increasing pH favors. If iron would be an anode and the tendency to corrode were to be evaluated, then the reduction potential for a possible cathode reaction would be placed on this diagram. If this point were to be, for example, in A in Fig. 5(a), this means that the reduction potential for this assumed cathode lies below any reduction potential of iron, and hence under these conditions iron is immune (since Erxn is negative). In fact any Fe2+ ions present could plate as iron. On the other hand, if the reduction potential of the assumed cathode reaction were to lie in point B, then there is a tendency to dissolve iron to Fe2+ since Erxn is positive. Finally, if the reduction potential of the assumed cathode was at point C, corrosion would occur, resulting in Fe(OH)3, but when oxides or hydroxides are formed there is a possibility that this product could form a solid protective layer that kinetically hinders further corrosion. These types of diagrams are called Pourbaix diagrams. Figure 5(b) shows the Pourbaix diagram for iron overlaid with the common cathode reactions in water, Eq 15 and 18 (Ref 8). The ionic activity was previously arbitrarily set at 10–6, but from the Pourbaix diagram it can be seen that changes in ion activity do not have dramatic effects on the boundaries. It can be seen that both the oxygen gas reduction reaction and hydrogen ion

reduction are able to cause corrosion through the entire pH region. Unfortunately, Fe-OH corrosion products are generally not passivating. Iron or carbon steel alloys are therefore not particularly corrosion resistant in water solutions. Figure 6 shows the Pourbaix diagram for chromium (Ref 8). While chromium oxidizes even more readily than iron, it forms Cr2O3 over a significantly large region that is of relevance to pH values in water solutions. Since Cr2O3 is protective, it prevents further corrosion. When chromium is added to iron as an alloying element, it corrodes selectively due to its low reduction potential, but this means that it also protects the iron alloy due to the properties of Cr2O3. This is the basic design principle behind iron-chromium-based stainless steels.

REFERENCES

1. A.J. Bard and L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, 2nd ed., Wiley, 2001 2. D.A. Jones, Principles and Prevention of Corrosion, 2nd ed., Prentice Hall, 1996 3. H.H. Uhlig and R.W. Revie, Corrosion and Corrosion Control: An Introduction to Corrosion Science and Engineering, 3rd ed., Wiley, 1985 4. M.G. Fontana, Corrosion Engineering, 3rd ed., McGraw-Hill, 1986

18 / Stainless Steels for Design Engineers

5. ASM Handbook, Vol 13A, Corrosion: Fundamentals, Testing, and Protection, S.D. Cramer and B.S. Covino Jr., Ed., ASM International, 2003 6. Handbook of Chemistry and Physics, 71st ed., CRC Press, 1991 7. D.A. Jones, Principles and Prevention of Corrosion, 2nd ed., Prentice Hall, 1996, p. 59

8. S.A. Bradford, Corrosion Control, 2nd ed., CASTI Publishing, Inc., 2001, p 41 SELECTED REFERENCE

• M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, NACE, 1974

Stainless Steels for Design Engineers Michael F. McGuire, p 19-25 DOI: 10.1361/ssde2008p019

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 3

Corrosion Kinetics Summary

the penetration due to dissolution of element i becomes:

CORROSION INVOLVES chemical reactions with equilibrium that is known through thermodynamics. In practice, the rate at which corrosion reactions occur is the most important consideration. This chapter deals with corrosion kinetics, which allows us to understand the rates of corrosion.

Introduction Consider the differential aeration cell discussed in the Chapter 2 on corrosion theory, Fe/ O2/OH–, Fe2+/Fe. If the thermodynamic conditions favor electrochemical corrosion of iron, that is, Erxn = EO / OH – EFe / Fe is positive, then a net corrosion current i will flow, resulting in iron dissolution and consumption of oxygen gas according to the net reaction, 2Fe (S) + O2 + – 2H2O → 2Fe2+ + 4OH . The magnitude of this current will determine the rate or iron dissolution according to Faraday’s law, which was introduced in Chapter 2: nNF = it. Because n = 2 and F = 95,485 C per mole electrons, the moles of dissolved iron are given as a function of time as N = i* t/(2* 95,485). Practically, this can be readily converted to lost mass m, which in the case of iron loss becomes m = N* MFe = MFe* i * t/(2* 95,485), or thickness reduction r, which in the case of iron becomes r = MFe* i* t/(2* 95,485* A* ␳Fe). Here, MFe and ␳Fe are molar mass and density of iron, respectively. It is often the thickness loss (referred to as penetration per unit time) that is useful; therefore, i/A is often replaced by j, which is defined as current density and has the units amperes/square meters. A general equation of –

r= j

t ⋅ Mi ni ⋅ F ⋅ ρ

(Eq 1)

The penetration rates for iron and various stainless steels are listed in Table 1 (Ref 1) in units of mils (0.001 in.) per year, or mpy. In the case of alloys, the ratio of Mi /ni is computed as an equivalent weight (EW) according to:

EW =

1 fn ∑ Mi i i

(Eq 2)

2+

2

where fi, ni, and Mi are the weight fraction, valence, and molar mass of element i, respectively. As stated, the amount of corroded (dissolved) iron is determined by the current i, and the magnitude of this current is determined by

Table 1 Penetration rates for a current of 1 µA/cm2 (mpy) Alloy

Fe 304 321 309 316 430 446 20Cb3

Element/oxidation state

Fe/2 Fe/2,Cr/3,Ni/2 Fe/2,Cr/3,Ni/2 Fe/2,Cr/3,Ni/2 Fe/2,Cr/3,Ni/2,Mo/3 Fe/2,Cr/3 Fe/2,Cr/3 Fe/2,Cr/3,Mo/3,Cu/1

(a) Equivalent weight. Source: Ref 1

Density, g/cm3

EW(a)

Penetration rate, mpy

7.87 7.9 7.9 7.9 8.0 7.7 7.6 7.97

27.92 25.12 25.13 24.62 25.50 25.30 24.22 23.98

0.46 0.41 0.41 0.41 0.41 0.42 0.41 0.39

20 / Stainless Steels for Design Engineers

EFe 2+ /Fe = EFe0 2+ /Fe −

Fig. 1

Schematic illustration of a differential aeration cell involving iron dissolution. Kinetic steps: (1) electrode reactions, (2) ion conduction, (3) electron conduction

the corrosion potential. The corrosion potential is determined by the reaction potential (which was discussed in Chapter 2) and the kinetics of the various steps involved in completing the electrochemical circuit depicted in Fig. 1. These involve: (a) electrode reactions at the cathode and anode, (b) conduction of ions in the electrolyte, and (c) conduction of electrons from the anode to the cathode. The conduction of electrons is generally not a problem in stainless steels because the corroding metal (iron) and scale (Cr2O3) provide an easy path for electrons. The other two kinetic processes are discussed briefly in this chapter.

4 RT ⎛ ( aFe ) ⎞ ln ⎜ ⎟ 4 F ⎝ aFe 2+ ⎠

When a cell is not under open circuit (i.e., a net current passes through it), the cathode and anode potentials deviate from the half-cell potentials, and the electrode states are then defined as being polarized. The polarization is quantified as overpotentials ␩, which are defined by the deviation from the equilibrium half-cell potentials, that is, for the cathode, ηc = Ecathode – EO2 / OH and for the anode, ηa = Eanode – EFe 2 + Fe. Effectively, the overpotential reduces the activation energy for the electrode Reactions 3 and 4. In the case of a reduction reaction at a cathode, such as Reaction 3, the overpotential is negative, and driving an electrode toward a lower potential drives electrons from the electrode into the solution, resulting in a net cathodic current ic at this electrode. Similarly, at the anode the overpotential is positive, which results in electrons that are favored to be removed from the solution and transferred into the electrode, thus producing a net anodic current ia. If the magnitudes of cathode and anode polarization are large, as would be expected in a galvanic cell, the relation between each electrode current/current density and overpotential is given by the following equations (for a thorough derivation of the current overpotential equation, Ref 2 is recommended):

The Butler-Volmer Equation For the case study 2Fe (S) + O2 + 2H2O → 2Fe2+ + 4OH–, the cathode and anode reactions are: O 2 + 2H 2 O + 4e − → 4 OH −

jc =

(Eq 4)

The Nernst equation predicts an open circuit potential of Erxn = EO / OH – EFe / Fe where –

2+

2

EO

= EO 0

2

and

/OH –

2

/OH –

(

)

4 ⎛ ⎞ RT ⎜ aOH – ⎟ − ln 4 F ⎜ aH O PO ⎟ ⎝ 2 2⎠

⎛ (1 − α )nF ηa ⎞ ia = j0 ,a exp ⎜ ⎟⎠ Aa RT ⎝

(Eq 6)

and ja =

2 Fe ( s ) → 2 Fe 2 + + 4 e−

(Eq 5)

2

(Eq 3)

and

CO (0, t ) ⎛ αnF ηc ⎞ ic exp ⎜ − = j 0 ,c 2 Ac C O* RT ⎟⎠ ⎝

where the jo,i terms are the exchange current densities and represent the equally large cathode and anode currents at equilibrium (zero overpotential) at the electrodes. The exchange current densities are a measure of the electrocatalytic ability of the surface to promote/demote the electrode charge transfer reactions; as such, they can vary over many orders of magnitude depending on the surface chemistry and structure and on electrode reaction. The α-terms are fractions that define the amount to which the

Chapter 3: Corrosion Kinetics / 21

activation energies are lowered. They do not have to be the same for the anode and the cathode, but due to the uncertainty in evaluating them, they are often taken as 0.5. The concentration terms represent the ratios between the reactant concentration at the electrode/electrolyte interface and bulk, which could deviate from unity as a result of consumption/production of species at the interface. In an iron-based alloy, this ratio for the anode would be close to unity because the reactant is iron itself, and no concentration gradient would be expected as a result of the corrosion reactions. When a net corrosion current flows, icorr = ia = ic. If the cathode and anode areas are assumed to be equal, then jcorr = ja = jc and Eq 5 and 6 can be rewritten (using βc = 2.3RT/(αnF) and βa = 2.3RT/ ((1 ⫺ α)nF) as: ηc = β c log

j 0 ,c jcorr

+ β c log

CO (0, t ) 2

C O*

electrode, and the rate of cathode reaction will depend on how rapidly oxygen molecules diffuse to the electrode/electrolyte interface. As a limiting case, when the oxygen concentration is actually zero at the interface, the corrosion current can, through Faraday’s law, be coupled to the steady-state flux of diffusive oxygen supply through a boundary layer δ. This limiting case current is called the limiting current (iL or jL) and can be expressed as: jcorr = jl =

DO nFCO* 2

In a nonlimiting case, the corresponding equation would be:

jcorr =

DO nF (CO* − CO (0, t )) 2

2

2

δ

(Eq 7)

2

(Eq 11)

2

δ

(Eq 12)

Combining Eq 11 and 12, one obtains:

and CO (0, t ) j ηa = β a log corr j0 ,a

2

(Eq 8)

Tafel Regime: Electrode-Kinetics Control. If the electrode charge transfer reactions are rate limiting, the supply of oxygen to the reaction site would be rapid enough to maintain a concentration at the electrode close to that of the bulk. In this case, Eq 7 and 8 would both result in a linear dependence of the overpotentials versus log jcorr: ηc = β c log j0 ,c − β c log jcorr

CO *

= 1−

2

jcorr jl

(Eq 13)

Thus, Eq 7 can be written: ηc = β c log

j 0 ,c jcorr

⎛ j ⎞ + β c log ⎜ 1 − corr ⎟ ji ⎠ ⎝

(Eq 14)

The slower the diffusion (small d), the lower the limiting current and thus a larger contribution from the mass-transfer-dependent second term on the overpotential.

(Eq 9)

Migration and Ionic Diffusion and ηa = β a log jcorr − β a log j0 ,a

(Eq 10)

The ionic transport in the electrolyte phase, the flux of an ion i under an electric field φ across a distance L, can be shown to be:

Mass Transfer Control. In Eq 7, the term: CO (0, t ) 2

CO* 2

stands for the ratio of oxygen gas concentration at the electrode/electrolyte interface and the concentration in the bulk, sufficiently far away from the interface. If the electrode reaction kinetics are very fast, the depletion of oxygen will lead toward a zero oxygen concentration at the

J i = − Di

∂ci zi F Δφ − Dc ∂x RT i i L

(Eq 15)

In an electrolyte with many different ions, an ion current through an area A can be computed by multiplying Eq 15 with zi * A and summing the contribution from all ions: i = FA∑ zi Di +

∂Ci + ( x ) ∂x

+

F2A RT

22 / Stainless Steels for Design Engineers

× ∑ zi 2Ci Di Δφ / L

(Eq 16)

Because the first term is important only at the regions near the electrodes (where consumption/creation of species occur), the current in the majority region of the electrolyte can be estimated as:

ηc = β c log

F A ∑ zi 2Ci Di Δφ / L RT

(Eq 17)

Using Ohm’s law (R = U/i), the electrolyte resistance can be computed as: ⎛ F2A ⎞ Relectrolyte = L / ⎜ ∑ zi 2Ci Di ⎠⎟ ⎝ RT

(Eq 18)

jcorr

⎛ j ⎞ + β c log ⎜ 1 − corr ⎟ ji ⎠ ⎝

jcorr j0 ,a

Ecathode = EH0 + / H − 2

(Eq 22)

RT PH2 ln + ηc nF aH2 +

= 0 − 0.059pH + β c log

j0 ,c jcorr

⎛ j ⎞ + β c log ⎜ 1 − corr ⎟ ji ⎠ ⎝

Mixed Potential Theory and Polarization Diagrams Viewing the electrochemical cell as an electrical circuit, Kirchoff’s law can be used to design a so-called polarization diagram. Consider, as a case study, a steel corroding under deaerated conditions, in a water solution, as shown in Fig. 2. Assume that the pH is such that a passive layer does not form (see the discussion of Pourbaix diagrams in Chapter 2). The cathode and anode reactions, respectively, are: 2 H + + 2 e− → H 2

(Eq 19)

Fe ( s ) → Fe 2 + + 2 e−

(Eq 20)

(Eq 23)

( )

Eanode = −0.447 − 0.0295 log aFe 2+ + ηa = −0.624 + β a log

jcorr j0 ,a

Natural Seawater Fresh (tap) water adjusted with seawater Fresh (tap) water adjusted with seawater Fresh (tap) water adjusted with seawater Deionized water adjusted with fresh (tap) water

Ratio by volume

Resistivity, ohm-cm

... 28:1

25 500

68:l

1,000

950:1

3,000

21:10

(Eq 24)

A polarization diagram is now constructed by plotting the anode and cathode potentials versus log jcorr. Strictly speaking, to close the circuit,

Test solution resistivity

Test solution

(Eq 21)

Now, the potentials of anode and cathode when current is flowing are in each case the equilibrium potential plus overpotential, that is:

The resistivities of some test solutions are shown in Table 2.

Table 2

j 0 ,c

and ηa = β a log

2

i≅

and the respective equations describing the overpotentials will be:

10,000

Fig. 2

Schematic polarization diagram

Chapter 3: Corrosion Kinetics / 23

the potential drop across the electrolyte needs to be included, which simply equals icorr * Relectrolyte (the electrolyte resistance is evaluated from Eq 18); however, in many cases, this term can be neglected. A schematic polarization diagram is shown in Fig. 3. The anode polarization is linear with decade current as predicted by Eq 24 because the overpotential has only a Tafel regime and no mass transfer dependence. On the other hand, the cathode polarization deviates from the Tafel behavior as a result of the effect of the mass transfer (hydrogen ion supply), dependent on the limiting current in Eq 23. It is noteworthy that the cell shown in Fig. 1 does not have a macroscopic anode and cathode. Different microscopic regions on the surface are assumed to act as cathodes and anodes, and in the lack of more detailed knowledge, the cathode and anode areas are assumed to be equal. The overall mixed potential of the surface would be at a corrosion potential Ecorr, defined in Fig. 2. In effect, the corrosion current resulting from the cell depends on the equilibrium half-cell potentials (Ecathode and Eanode), the Tafel slopes (βc and βa), the exchange current densities (jo,c and jo,a), and any limiting current density (jl). Figure 3 shows schematically how decreasing any of the Tafel slopes and increasing an exchange current density increases the corrosion rate. The effect of the electrolyte resistance has been ignored; that is, corrosion current is where the two polarization curves intersect. Figure 4 shows the effect of increased mass transfer, which would result in an increase in the limiting current. In an active (nonpassive) alloy, this results in an increased corrosion current up to a point.

Fig. 3

Passivation Theory. In Chapter 2, it was identified through the Pourbaix diagrams that there were conditions under which an alloy could be passive. In the case of stainless steels, the range of pH and other conditions under which this would occur has been increased thanks to the chromium content, which readily forms a Cr2O3 scale. In general, a passive layer constituted of adsorbed molecules or thin oxide/hydroxide layers decreases the corrosion current. Researchers (Ref 3) have reported that the constituents of the passive film are alpha Cr2O3 and Cr(OH)3nH2O. The structure is reported to be a nanocrystalline spinel, epitaxial to the surface. The grain size may decrease with increasing chromium content. This protection by chromium requires a threshold level of 11 to 12% chromium. Effect on Polarization Diagrams. The polarization diagram for a passive alloy is quite different from those discussed for active alloys. A schematic of a typical polarization curve is shown in Fig. 5. When a passive alloy is anodically polarized, it initially behaves like an active alloy (i.e., with a Tafel slope, etc., as the passive layer is building up). The building up is actually a selective dissolution of iron, which causes a greater remaining surface concentration of chromium and other alloying elements. Once the passive layer is formed and offers protection against further dissolution, the potential-decade current relation drops to lower currents. This happens at potentials beyond the passivation potential Epp. At some high enough

Corrosion rate and the effect of (a) Tafel slope and (b) exchange current density

24 / Stainless Steels for Design Engineers

polarization level, the passive layer breaks down, and the metal becomes active again; this region is called the transpassive regime. The design of a structure involving a passive metal should aim at forming a corrosion cell in which the cathode polarization curve intersects the anodic one in the passive regime. Consider, for example, an alloy that exhibits the behavior shown in Fig. 6 in deaerated acidic solutions with different pH. If mass transfer limitations due to hydrogen ion supply are neglected, then the cathode polarization is given by:

Ecathode = EH0 + / H − 2

RT PH2 ln + ηc nF aH2 +

= 0 − 0.059pH + β c log

j0 ,c jcorr

This will result in a straight line as shown in Fig. 6, which will be shifted vertically depending on the pH. The dashed circles indicate the intersection between anode and cathode polarization curves that would yield the corrosion current. At a sufficiently high pH (= pH1), the alloy is clearly not optimal because intersection occurs in the active regime, and the passive properties are not utilized. This is what occurs when a reducing acid is too strong for a given stainless steel, such as with concentrated hydrochloric acid. At pH = pH2, on the other hand, a low-corrosion current is obtained as a result of intersection at the passive regime. This is the benevolent case when stainless steel is correctly matched to the environment, and low rates of uniform corrosion occur. Finally, at pH = pH3, the resulting corrosion current is again high as a result of intersection occurring at the transpassive region. This could occur with some stainless steels exposed to a very strong alkali solution. Using a similar argument, the readers can themselves deduce the effects of cathode exchange current density and Tafel slopes. In the discussion of active anode polarization, it was found that increasing the transport rate of cathode reactants through agitation, for example, would increase the corrosion rate up to a point but beyond that have no further effect (see Fig. 4). In the case of a passive/active behavior, the effect of mass

Fig. 4

Effect of increasing the limiting current by, for example, increased agitation in the electrolyte. Beyond the dashed line, increasing the limiting current would have no further effect

Fig. 5

Schematic of a passive anode polarization curve

(Eq 25)

Fig. 6

Effect of cathode polarization

Chapter 3: Corrosion Kinetics / 25

Fig. 7

Effect of mass transport

transport is somewhat different, as shown schematically in Fig. 7. Increasing mass transport, such that the limiting current increases, results initially in an increased corrosion current (e.g., increasing jl from 1 to 2). It should be noted that there are several intercepts possible (both in the active and passive regime), but assuming there are defects present, it is likely that there will be corrosion corresponding to the higher current. Increasing the limiting current beyond the knee corresponding to Epp, however, results in a drop in the current because now the only corrosion potential possible is at the intersection in the passive regime. This is the case for jL3. In the normal use of stainless steel, achieving passivity takes on several forms. What is often called passivation is actually a cleaning process in which contaminants, such as tramp iron, are removed from the surface. Dilute nitric acid is an excellent vehicle to achieve this. This medium has the additional benefit of forming a passive film on an active stainless surface. This

is the actual passivation; the iron removal is really a chemical cleaning operation, which happens to be called passivation. During the production of stainless steel, after a final anneal another version of passivation is carried out. The oxide from annealing in air is dissolved by a strong mixture of nitric and hydrofluoric acids, which does not allow passivation. This treatment, called pickling, removes by dissolution both the oxide layer and the chromium-depleted layer below the oxide formed during annealing. The depleted layer can extend a number of microns in depth and would seriously degrade corrosion resistance if not removed (Ref 4). This is then followed by a straight nitric acid immersion, which ensures complete passivity. This is the procedure that should be performed on the oxides formed during welding if full corrosion resistance is to be restored. Simply removing the oxide through mechanical means leaves a chromium-depleted layer that corrodes more readily than is expected of the alloy. REFERENCES

1. D.A. Jones, Principles and Prevention of Corrosion, 2nd ed., Prentice Hall, 1996 2. A.J. Bard and L.R. Faulkner, Electrochemical Methods: Fundamentals and Applications, 2nd ed., Wiley, 2001 3. M.P. Ryan et al., Critical Factors in Focalized Corrosion, Proc. Electrochem. Soc., Vol 150, 2003, p 583–594 4. J. Grubb and J. Maurer, “Corrosion of the Microstructure of a 6% Molybdenum Stainless Steel with Performance in a Highly Aggressive Test Medium,” paper 300 presented at Corrosion 95, NACE International, 1995

Stainless Steels for Design Engineers Michael F. McGuire, p 27-56 DOI: 10.1361/ssde2008p027

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 4

Corrosion Types Summary STAINLESS STEEL is unusual among alloy systems in that its corrosion resistance derives from the passivating ability of a minor constituent, chromium. Thus, while stainless steels can be made to be essentially immune to corrosion in many environments, it can also experience various debilitating forms of localized corrosion, which stem from the failure of this passive film. This chapter explores the behavior of stainless steel in media that promote uniform corrosion and the various mechanisms of localized corrosion, such as pitting and crevice corrosion.

plex in their behavior because the influence of processing and alloying variables changes the ability of this layer to form and remain stable in the face of aggressive environments. The behavior of stainless steel is further affected by its microstructural complexity. Stainless steel alloys may have many constituent elements and many thermodynamically possible phases, and none of these are necessarily uniform in their composition. Yet, it is the composition of the alloy in contact with the specific environment at any microscopic point that determines the corrosion resistance of that particular point.

Uniform Corrosion Introduction To most designers, the most recognized characteristic of stainless steel is corrosion resistance. Stainless, unlike noble metals such as gold, does not obtain its excellent corrosion resistance from inertness. Instead, it is the reactivity of chromium that allows the surface layer of corrosion product to become sufficiently adherent and impenetrable, which effectively stops further corrosion by isolating the base material from the environment. This resistance to corrosion is called passive behavior or passivity. Other metals, such as aluminum and titanium, form similar layers and also exhibit passivity. The important difference in the case of stainless steel is that chromium is still a minor constituent, never more than 30% by weight, sometimes little more than 10%. How much chromium there is and how uniformly it is distributed have a profound effect on corrosion resistance by virtue of its ability to concentrate into the surface film. Stainless steels are com-

When all parts of a corroding surface have equal access to the corroding atmosphere and the structure of the corroding metal is relatively uniform, a uniform thinning of the material is expected. Stainless steels are materials of choice because, by virtue of their passive behavior, they show very low rates of uniform corrosion in many environments. The metallurgy and processing of a particular grade are designed to provide passivity in a given environment. The environment can be too aggressive to allow passivity to be maintained either by being too reducing, as with some acid media, so that passivating species cannot form or by being too oxidizing so that the oxidized species that normally affect passivity are no longer stable. The former is called dissolution in the active state, while the latter is termed transpassive dissolution. Intelligent design and knowledge of the environmental variables for a stainless steel component ensure that the alloy is used in the passive state, at which uniform corrosion occurs at a very low rate.

28 / Stainless Steels for Design Engineers

Among the important media with which we encounter uniform, but acceptably controlled, corrosion in stainless steel are atmospheric and marine environments and chemical environments such as sulfuric acid, phosphoric acid, nitric acid, strong bases, and organic acids, such as acetic and formic. Pickling is an example of controlled, accelerated uniform corrosion. This is typically done with 10 to 20% hot sulfuric acid or a mixture of hydrofluoric and nitric acids. Environmental Variables Influencing Uniform Corrosion The corrosion of stainless steels is usually the result of contact with an electrolyte, allowing a complex set of partial electrochemical reactions, which may occur sequentially or concurrently. The corrosion rate depends on the current exchanged between the negative and positive electrode (anode and cathode). These may be on a macroscopic or microscopic level. The main consideration is normally ionic transpor tthrough the passive film, which after all is what makes stainless so effective against corrosion. The chemical parameters that influence the media with respect to uniform corrosion rate are the acidity and the oxidation-reduction (redox) potential of the electrolytic medium, both of which act through their influence on the stability of the passive film, rendering it active, pas-

sive, or transpassive; Fig. 1 illustrates the effect of redox potential on a solution. Certain anions have strong effects in media through their well-known, if not well understood, disruption of the passive film. Halides are well known for this effect, but sulfides are active. These anions seem to intervene in the adsorption of the hydroxyl ions. In acid media, these anions accelerate uniform corrosion, while in neutral media they may result in localized corrosion. Anions that form soluble complexes with elements in stainless, such as amines, formates, or acetates, can also disrupt the stability of the passive film and thus promote active corrosion. Of the physical variables of the environment, it should be obvious that temperature is paramount since all the reactions are thermally activated. Increasing temperature may speed the formation of the passive film when thermodynamic conditions are favorable, but in general one expects increasing temperature to increase corrosive attack. Access to passivating species, such as oxygen, is important in establishing and maintaining passivity. Increased diffusion of reacting species in the liquid will normally accelerate the partial reactions, but if passivity is stable, the rate-limiting transport through the passive film will not be affected. Therefore, increasing the flow rate of a corrosive fluid does not automatically accelerate corrosion. The reduction of concentration gradients can be beneficial against localized

Fig. 1 Reduction potential versus pH for iron and (a) oxygen gas reduction and (b) hydrogen ionreduction

Chapter 4: Corrosion Types / 29

corrosion, and flow can bring to the surface an increased supply of passivating species. Increased flow rate in a fluid medium is deleterious if it induces mechanical damage to the passive film by erosion, abrasion, or cavitation. These are complex mechanisms, but it should be apparent that the success of a stainless steel to a given flow condition will depend mainly on its ability to form and re-form its passive film. Somewhat counterintuitively, thinner passive films are more protective than thicker films among stainless alloys. The tenacity of the thin passive films on stainless (and titanium) make these alloys quite resistant to flow-accelerated corrosion, as contrasted to copper and aluminum alloys, which have soft, thick corrosion product films. Material Variables Stainless steels have a great variety of alloying elements and microstructures. As a generalization, we can say that corrosion resistance is a function of composition rather than structure. Then, we must quickly add the qualifiers to this statement. On an undisrupted, stress-free surface, local composition does quite precisely determine corrosion resistance. But, stainless steels are seldom homogeneous or at thermodynamic equilibrium. Impurities such as oxygen and sulfur are usually present, mainly as inclusions since they have diminishingly small solubility at room temperature. At high temperatures after solidification, as in welds, they can be present in supersaturation, ready to precipitate as inclusions that alter local composition. The tendency of carbon and nitrogen to form precipitates is controlled by diffusion rates, which if elevated by increasing temperature can cause debilitating, composition-altering precipitation. The even more slowly diffusing substitutional alloying elements, such as chromium, molybdenum, and nickel, have strong tendencies to form phases that disturb their uniformity in the austenite or ferrite matrix in which they are intended to work. So, any discussion of the influence of alloying element on corrosion resistance of a phase like austenite or ferrite must recognize that alloying elements exert their effect when they are in solution in that phase. The same element may under some conditions not be in solution and have a contrary effect. An example is molybdenum, which is obviously a great enhancer of corrosion resistance when in solid solution. When it precipitates as a con-

stituent of sigma phase, which it promotes, however, it combines with chromium. If this happens at relatively low temperatures, the surrounding matrix is depleted of both chromium and molybdenum, and the corrosion resistance in that region is diminished. Nitrogen also is effective when in solid solution in austenite but can precipitate as a chromium nitride under certain conditions and cause depletion of the remaining matrix. Local structure and composition are paramount. This becomes more important to localized corrosion, as discussed later, but it should be remembered in examining uniform corrosion because corrosion will cease to be uniform when composition becomes nonuniform. The compositional material variables that influence uniform corrosion are not exactly the same as those that will be seen to influence localized corrosion. The foremost element is, of course, chromium. Researchers (Ref 1) have reported that the constituents of the passive film are alpha Cr2O3 and Cr(OH)3nH2O. The structure is reported to be a nanocrystalline and epitaxial to the surface. The grain size may decrease with increasing chromium content. This protection by chromium requires a threshold level of 11 to 12% chromium. This threshold has been attributed most convincingly to the minimum chromium content that permits chromium atoms on surface sites to be linked by adsorbed oxygen atoms (Ref 2). In any event, the mechanism by which this thin, severalnanometer-thick, film forms is the subject of ongoing debate, but we do know that it is enriched in chromium, and that it is thinner for higher chromium alloys. The critical current density j, as measured during polarization, is also smaller as chromium content increases (Fig. 2). This is consistent with the lower dissolution of noncontributing elements required to achieve a critical surface chromium concentration. Increases in chromium can also be seen to lower the current density in the passive region. This is manifest in alloy performance as a reduction in the uniform corrosion rate in a given medium. From an electrochemical point of view, this is explained as a manifestation of the stability of the Cr(OH)3 nH2O. The role of molybdenum is less clear. The observed action of molybdenum is to greatly reduce the critical current density required for passivation. This is also seen as accelerating the formation of the passive films and as increasing the resistance of the alloy to depassivation at

30 / Stainless Steels for Design Engineers

Fig. 3 Fig. 2

Influence of alloying element on corrosion rate as explained by the effect on polarization.Source: Ref 6

Schematic illustration of polarization behavior for a passive alloy with and without pitting occurring

lower pH. The role of molybdenum is not to enrich in the passive film itself, although it can be found in the film. Its potency is far more than its presence can take into account. Pure molybdenum is itself not passive. Its action does not appear to be via a product of reaction. Instead, it seems to reduce the dissolution rate of elements other than iron, which would promote a surface richer in chromium (Ref 3). The action of molybdenum as an alloying element is complicated by the fact that molybdate ions are known to impede pit growth as a separate effect from their action within the alloy matrix (Ref 4). Copper has a similarly complicated effect, with copper ions gettering sulfide ions and redepositing as metallic copper (Ref 5). Nickel also lowers the critical current density for passivation without contributing directly to the passive film’s stability. This also may be the result of the stronger bond between nickel and chromium reducing the anodic dissolution rate of the alloy by permitting the anodic enriching of the surface by selection iron dissolution. Nickel does not actively help passive film formation and can actually hinder film stability in highly acidic/oxidizing environments. Nitrogen, however, appears to be more like molybdenum in its effect. While nickel and copper provide no benefit to the stability of the passive film once it is formed, both nitrogen and molybdenum do, and to a degree that cannot be explained by their presence in the film. This may then relate to their thermodynamic action within the alloy itself. Molybdenum and nitrogen act both to enhance the enrichment of

Fig. 4

Influence of alloying elements on uniform corrosion rate in 20% sodium chloride solution with carbon dioxide pressure of 20 MPa. Source: Ref 7

chromium in the passive layer and to decrease active dissolution of noniron alloying elements, thereby promoting both the formation and stability of the passive film. A summary of the known major alloying effects in acidic chloride media is shown in Fig. 3 in acidic chlorides. Alloying elements provide benefits in the part of the chart where they appear (Ref 6). From this, it can be seen that chromium, molybdenum, nickel, copper, and nitrogen all assist in the active region, while chromium, molybdenum, and nitrogen expand the region of passivity and diminish the corrosion current. An example of the influence of these alloying elements on the uniform corrosion rate of stainless steels in a sodium chloride/carbon dioxide environment is shown in Fig. 4 (Ref 7). Note the alloying composition is measured by a crevice corrosion index (CCI), which is discussed in the section Localized Corrosion.

Chapter 4: Corrosion Types / 31

Fig. 5

Corrosion table for stainless steels and titanium in sulfuric acid plus copper sulfate. Corrosion rate legend: 0, < 0.1 mm/yr (corrosion resistant); 1, 0.1–1 mm/yr (useful in certain circumstances); 2, > 1.0 mm/yr (material not recommended). Source: Ref 8; see source for interpretation of data. Courtesy of Outukumpu Stainless

Unfortunately, it cannot be assumed that this relationship is true for other environments, although other empirical relationships exist or can be generated. Because the influence of alloying element varies with environment, we need to discuss some of the more commonly encountered severe environments. Corrosion in Acids and Bases The examples discussed in Chapter 3, “Corrosion Kinetics,” refer mostly to corrosion in these aqueous solutions, in which the slow thinning rate of the chosen alloy can be determined through the mixed potential theory and polarization diagrams. In the case of stainless steels, the alloy chemistry is chosen such that the passive-active behavior favors corrosion in the passive regime. The corrosion rate of the various stainless steels in the myriad possible environments has been measured in probably all practical cases. These data can be obtained from a number of sources, such as the National Association of Corrosion Engineers (NACE) and ASM Handbook volumes. None is more accessible than the Web site of Outukumpu, which contains a “Steel Professional Tool,” a lookup table in which the corrosion rate of many stain-

less steels in a great number of environments can be obtained. Figure 5 shows an example of one such table. Many of the isocorrosion charts in this book are reprinted from this source, http://www.outokumpu.com/applications/documents/start.asp (Ref 8). These tables are supplemented by isocorrosion diagrams such as that shown in Fig. 6. These diagrams show constant corrosion behavior under varying environmental conditions such as temperature and solution composition. This information is available to guide the designer in selecting appropriate steels for various environments, and it is highly recommended that it be used. Free sites tend to promote proprietary alloys, as these charts suggest. The serious engineer will consult multiple sources and unbiased sources before making alloy decisions. The influence of alloying element is by no means the same in all environments. So, while it is useful and necessary to have these experimental data, it is also helpful to understand the peculiarities of some of the major alloy-environment pairings. Sulfuric Acid. Stainless steels require more than a minimum amount of alloying to resist sulfuric acid. Straight 16% chromium grades such as 430 fare poorly, while the nickel-containing

32 / Stainless Steels for Design Engineers

Fig. 6

Isocorrosion curves for 17-12-2.5 stainless steel and titanium in sulfuric acid plus copper sulfate.Source: Ref 8. Courtesy of Outukumpu Stainless

304 has more than an order of magnitude better corrosion rate in either dilute or concentrated sulfuric acid at ambient temperatures. Figure 7 (Ref 9) shows the isocorrosion rate curves for several common alloys. Alloying with molybdenum is also very effective, as is alloying with copper. If passivity cannot be established, increasing chromium content actually increases corrosion rate. The corrosion behavior of sulfuric acid varies greatly with concentration. At low concentrations, sulfuric is a classic reducing acid. It dissociates in water to create hydrated hydrogen ions (H3O+) that release hydrogen gas bubbles during the corrosion reaction. As the acid concentration increases, the solutions become more corrosive, and progressively more highly alloyed stainless steels are required to provide adequate corrosion resistance. At about 50% acid, only the most highly alloyed stainless alloys (alloy 20, AL-6XN, C-276, etc.) can provide acceptable corrosion rates, and even these alloys are restricted to use at near ambient temperatures. As acid concentration increases beyond 50%, the solution begins to show oxidizing behavior. At acid concentrations above 80%, nickel-molybdenum-copper-bearing stainless steels begin to exhibit useful corrosion resistance. In the 93 to 98% sulfuric acid concentration range, carbon steel can be used to hold sulfuric acid at ambient temperatures, although stainless steels provide better performance at elevated temperatures or if flow-erosion can occur. In the 96 to 100% sulfuric acid concentration range, at elevated temperatures, the oxidizing character is quite pronounced, and oxidation-

Fig. 7

Isocorrosion rates of various stainless steels in sulfuric acid. Source: Ref 9

resistant high-chromium (type 310S) and highsilicon (MECS ZeCor UNS S38815 and Sandvik SX S32615) stainless steels are frequently used, especially in sulfuric acid-manufacturing

Chapter 4: Corrosion Types / 33

Fig. 8

Fig. 9

Influence of alloying element on corrosion rate in contaminated sulfuric acid. Source: Ref 11

Isocorrosion curves for various alloys in sulfuric acid

equipment. Sulfuric acid-containing dissolved sulfur trioxide is called oleum, and such solutions are often identified as sulfuric acid of greater than 100% concentration. Highchromium stainless steels (i.e., type 310S) are among the very few materials that exhibit corrosion resistance in oleum. (See MTI Materials Selector Volume 3—Sulfuric Acid at www.mtiglobal.org for more information.) Aeration has a major influence on corrosion rates because oxygen stabilizes the passive film. Molybdenum-alloyed stainless has dramatically lower corrosion rates in aerated solutions than

non-molybdenum-bearing alloys. Their superiority in deaerated solutions is much less marked. Studies (Ref 10) have shown that in sulfuric acid molybdenum is highly enriched in the passive film, and when molybdenum is an alloy, chromium also is enriched. This is a manifestation of selective dissolution of other elements in the matrix. Oxidizing impurities, such as ferrous ions, act like aeration to diminish the corrosive attack, but reducing impurities such as halides have an extremely negative effect, as the corrosion tables will show. These effects are not linear and underscore the value of these tables. The uniform corrosion rate in contaminated sulfuric acid may be more important than in pure acid since this represents a potentially likely failure mode because contamination is a constant hazard. Figure 8 shows the corrosion rate of various alloys in sulfuric acid contaminated with chlorides and iron. These researchers (Ref 11) found that the resistance to attack correlated to the alloy content by the formula shown. Figures 9 and 10 show how isocorrosion rates vary with alloy and contamination level. Hydrochloric acid is very destructive of the passive film on stainless. An alloy like 304 is not suitable even in a deaerated 1% HCl solution at room temperature. Chromium additions

34 / Stainless Steels for Design Engineers

Fig. 12

Fig. 10

Isocorrosion curves for various alloys in sulfuric acid with chlorides

Fig. 11

Isocorrosion curves for various stainless steels in hydrochloric acid. Source: Ref 8. Courtesy of Outokumpu Stainless

are only modestly helpful, while nickel, copper, and molybdenum are more beneficial. Stainless steels are not good materials for contact with hydrochloric acid. Figures 11 (Ref 8) and 12 (Ref 12) show how even the most highly alloyed grades can withstand only dilute concentrations and low temperatures. While a stainless steel vessel may not be intended to be used for hydrochloric acid, resistance to lesser amounts of chlorides is important because of the possibility that an acidic environment may be contaminated with chlorides. When this is a possibility, then proper alloy selection must guard against it. Figure 13 shows a correlation between alloy content and resistance to the general corrosion (GI) by sulfuric acid contaminated with hydrochloric acid (Ref 11).

Isocorrosion curves for austenitic AL-6XN (UNS N08367) and 904L (UNS N08394) stainless steels in hydrochloric acid. Source: Ref 12

Nitric acid is strongly oxidizing. This actually promotes the passive film formation; consequently, even low-chromium alloys remain passive at all concentrations at ambient temperature (see Fig. 14) (Ref 8). The addition of molybdenum, which is so generally helpful, is deleterious in this case because it forms soluble compounds. It is useful to keep carbon, silicon, and phosphorus as low as possible. Silicon is unusual in that normal levels (0.4 to 1.0 %) are worst, with very low (0.05%) or very high levels (4.0%) beneficial. The low levels of silicon contents of these alloys are useful for their action in minimizing grain boundary segregation, which is the usual locus of attack. High silicon levels contribute to a general protective silica surface layer in concentrated acid, which augments the true passive layer. This superiority appears above the azeotropic composition of about 67%, which is a common commercial concentration, as shown in Fig. 15 for high-silicon austenitic stainless steels (Ref 13). Phosphoric Acid. This oxidizing acid behaves more like sulfuric acid in that simple iron-chromium alloys have only moderate resistance to uniform corrosion in them, while alloying with molybdenum and copper produces major improvements. This can be seen in Fig. 16, in which alloys with increasing nickel (18-10) show clear benefits over a chromium-molybdenum alloy (18-2), and the added molybdenum in 317 (17-14-4) is better, while 904L with nickel, molybdenum, and copper is even better (Ref 8). In the commercial production of phosphoric acid, halide impurities may be present, in which

Chapter 4: Corrosion Types / 35

Fig. 13

Influence of alloy content on corrosion rate in hydrochloric acid

Fig. 14

Isocorrosion curve for nitric acid. Courtesy of Outokumpu Stainless

Fig. 15

Corrosion behavior of high-silicon alloys in concentrated nitric acid. Courtesy of Outokumpu Stainless

case alloys with higher molybdenum, chromium, copper, and nitrogen may be required. Organic Acids. The weakly dissociating organic acids are normally not aggressive against stainless steels. The exceptional dangerous environments are those that include high temperature and the presence of chloride contamination. It should be noted that in formic acid, which does dissociate more strongly, nickel is detrimental. This phenomenon is also seen in the production of urea via the intermediary ammonium carbamate. The difficulty lies in the hightemperature solubility of nickel complexes and is best addressed by the use of ferritic or duplex alloys. Alloying with molybdenum seems to provide the greatest resistance to uniform corrosion in strong organic acids, as illustrated in Fig. 17 (Ref 8). If halides are present in organic acids and liberated by contact with water, then pH and chloride concentration will govern the corrosive attack, which could then become nonuniform. Strong Bases. In strong bases, the stainless steels are generally quite resistant to uniform corrosion. Straight chromium (17%) alloys are usable at any concentration up to 50 °C. Adding molybdenum and nickel does little to further improve performance as the underlying resistance is due to chromium. Increasing chromium levels provide increased resistance. Attack when it does occur can be manifested as grain boundary attack. Figure 18 shows isocorrosion curves for sodium hydroxide (Ref 8).

36 / Stainless Steels for Design Engineers

Fig. 16

Isocorrosion curves in phosphoric acid: (a) 0.1 mm/yr for various stainless steels; (b) 0.1 mm/yr for titanium and 17-12-2.5 stainless steel. Courtesy of Outokumpu Stainless

Fig. 17

Isocorrosion curves in organic acids: (a) acetic acid; (b) formic acid. Source: Ref 8. Courtesy of Outokumpu Stainless

In the pulp-and-paper industry, chemical pulping is called the kraft or sulfate process. In the presence of sulfur, nickel can be quite detrimental, and ferritic or duplex alloys are preferred. This again is caused by the solubility of nickel complexes formed in the presence of sulfur-containing compounds. This can be seen in Fig. 19, which shows a 26-1 (chromiummolybdenum) alloy significantly outperforming higher alloys that contain nickel and molybdenum (Ref 12). Atmospheric Corrosion Fig. 18

Isocorrosion curves for various materials in sodium hydroxide. SCC, stress corrosion cracking. Courtesy of Outokumpu Stainless

Atmospheric corrosion is an example of uniform corrosion that occurs when a thin layer of water condenses on a metal surface and as such

Chapter 4: Corrosion Types / 37

Fig. 19

Corrosion rates of various alloys in simulated evaporator liquid. Source: Ref 12

depends on humidity, temperature, and other atmospheric conditions. The rate of corrosion measured as defined in Chapter 3, “Corrosion Kinetics,” as dissolution r of element i as r= j

t ⋅ Mi ni ⋅ F ⋅ρ

where j is current density, t is time in seconds, M is molar mass, n is valence, F is the faraday constant, and ρ is the density), and therefore has two contradicting effects of temperature. In general, temperature increases the exchange current density and transport properties and thus the kinetic rates involved in corrosion. On the other hand, increasing temperatures may reduce the concentration of dissolved oxygen in the electrolyte and eventually will dry the surface and thus limit the electrochemical corrosion due to the access to an electrolyte. In steels, the corrosion products are (a) an outermost layer of porous rust (FeOOH) characterized by low water content but easy access to oxygen and (b) an inner layer of magnetite (Fe2O3) in which pores are filled with water. The access of oxygen to the bare metal limits the cathode reduction reaction rate, and in relatively pure atmospheres, the corrosion rate is negligible due the protective nature of the oxide. However, sulfur dioxide impurities in the atmosphere react with water to form sulfuric acid, which tends to dissolve the protective oxide. In the case of stainless steels, the passive region is extended due to chromium, to a wide enough region in terms of pH (see Pourbaix diagrams in Chapter

2) that atmospheric corrosion can in effect be prevented. The most deleterious impurity in the atmosphere for stainless is the chloride ion. Chlorides are pervasive. Borne from oceans by normal climatological processes, they are found far inland. In many colder climates, they are also seen in high concentrations from road salts. Without washing or the natural rinsing by rain, surface chloride concentrations can become very high. Thus, the rules of corrosion of aerated aqueous solutions are followed by stainless with respect to atmospheric corrosion. The difficulty is accurately estimating the solution that constitutes the aqueous solution. Much experience has shown that if coastal and road salt effects are minimal, then 18% chromium alloys such as 304 experience such negligible visible corrosion that they can be used for exposed, unrinsed architectural purposes. If the same alloy is used in an unrinsed coastal environment, red rust stain will occur. This is the corrosion product from metastable and possibly stable pitting. In Japan, where coastal conditions prevail throughout, much research has been done that has shown that a pitting resistance equivalent number (PREN) of 25 is necessary for freedom from corrosion (i.e., zero pitting) (Ref 14). This contrasts to a requirement of about PREN 35 to resist pitting in seawater. Pitting is a form of localized corrosion, and PREN is an index to pitting resistance. These concepts are examined in the next section. Localized Corrosion Localized corrosion is in general more damaging from a structural integrity point of view than uniform corrosion since the corrosion current is limited to a small area and the penetration distance is large. Often, localized corrosion involves a large-area cathode and a small-area anode, which means that for a given corrosion current, the corrosion current density at the anode is very large. In localized corrosion, unlike uniform corrosion, the anode and cathode are clearly identifiable locations, and the reason that certain structural features assume the roles of cathode and anode can be used to categorize and exemplify different cases. Interestingly, the cathode and anodes, while identifiable, can vary across scales, that is, from distinct macroscopic components or parts to microstructural features. In Chapter 2, the tendency for corrosion was introduced as a positive value for an

38 / Stainless Steels for Design Engineers

electrochemical cell potential (Erxn) corresponding to a spontaneous electrochemical reaction forming a galvanic cell. Erxn is obtained as: 0 Erxn = Erxn −

α α α RT ⎛ p1 1 p2 2 p3 3 ... ⎞ ln ⎜ β β β ⎟ nF ⎝ r1 1 r2 2 r3 3 ... ⎠

(Eq 1)

Here, pi and ri are the concentrations of reactant and products, respectively, and αi and βi are the numbers that are needed to balance the reaction stoichiometrically. Any time that Erxn is positive, there is thermodynamically a tendency for an electrochemical reaction, in our case a corrosion reaction. The rate of corrosion, as discussed in Chapter 3, is dependent on the polarization behavior. Dissimilar Metals and Differential Aeration Cells The case of dissimilar metals and differential aeration cells is perhaps more important in active alloys than for stainless steels, which are generally passive, and occurs when two metals/alloys are in contact that have elements in them that are dissimilar in the electromotive force (emf) series (see Chapter 2) and there is an electrolyte present. For example, if nickel and iron pipes are connected and water flows though them containing some traces of Ni2+ ions, then: 0 Erxn = ENi − EFe0 2 +/ Fe − 2+ / Ni

RT ⎛ aFe 2 + ⎞ ln ⎜ ⎟ 2 F ⎝ aNi 2 + ⎠

(Eq 2)

In this case, the corrosion tendency is primarily caused by the first two terms on the right side of Eq 2, the dissimilarity in the standard half-cell reduction potentials: 0 ENi − EFe0 2 + / Fe = − 0.250 + 0.447 = 0.197 V 2+ / Ni

This tendency is caused by the galvanic dissimilarity between the metals. This is normally important for alloys joined to stainless that are themselves less noble. Less-noble alloys, such as carbon steel, can fail rapidly if coupled to stainless. A classic example is the use of carbon steel fasteners for joining stainless sheets. Different stainless steel alloys have minor differences when passive, but if the environment is such that one alloy is active while another is passive, then the galvanic differential could be harmfully large. In many cases, a situation

arises in which access of oxygen is not the same to different areas of a sample. In effect, this results in that the cathode reaction: O 2 + 2H 2 O + 4e − → 4 OH −

(Eq 3)

is limited from proceeding in some areas but not others. This gives rise to a differential aeration cell. For example, consider Fig. 20(a), in which a metal is partially immersed in water. Transport distance of oxygen increases with depth; the limiting current would then vary with depth, such as at locations 1 and 2, and cathode polarization curves as a result of this are schematically plotted in Fig. 20(b). Near the surface of the water, where oxygen is readily replenished, passivation is likely to be fast, and thus anodic iron dissolution is slow. This region assumes the role of the cathode, and reaction 1 occurs. Sufficiently far away from the surface, if there are regions where passivation is incomplete (e.g., surface defects or scratches) or has broken down as a result of, for example, Cl− (see section on pitting), repassivation does not readily occur since oxygen transport is too slow. These regions become anodes where the following reaction occurs: 2Fe (s) → 2Fe 2+ + 4e −

(Eq 4)

The distance at which this occurs is balanced by being large enough to limit the rate of oxygen transport but not too long to be strongly influenced by ion transport that is needed to complete the electrochemical cell. Resulting corrosion currents are shown in Fig. 20b. This type of degradation is called waterline corrosion. Crevice Corrosion. In stainless, the more significant occurrence of this type of cell occurs when a crevice, from whatever cause, exists, and reactions within the crevice or pit cause the accumulation of iron ions by: 2Fe (s) → 2Fe 2+ + 4e −

(Eq 5)

The regions adjacent to the drop that maintained their passive layer and have access to oxygen act as cathodes where the oxygen reduction reaction takes place: O 2 + 2H 2 O + 4e − → 4 OH −

(Eq 6)

This reaction maintains an alkali solution. As a result of the geometry, Fe2+ ions remain and

Chapter 4: Corrosion Types / 39

Fig. 20

Schematic illustration of (a) sample partially immersed in water; (b) resulting polarization behavior for two different passivating alloys (A and B polarization curves)

enrich in the water-filled pit; to maintain charge neutrality, Cl– migrates into the pit. This causes the following reaction:

)

Fe 2+ + 2H 2 O + 2Cl − → Fe ( OH 2 + 2HCl

(Eq 7)

which has several consequences: (a) Hydrochloric acid further acidifies the pit and increases the rate of iron dissolution since decreasing pH increases cathode half-cell potential, which increases corrosion rate (see polarization diagram construction in Chapter 3, “Corrosion Kinetics.”) (b) The formation of porous Fe(OH)2 further helps to isolate the pit, thereby separating anode and cathode regions in the differential aeration cell. (c) The presence of Cl– prevents repassivation. As a result of increased acidification, the dissolution rate becomes autocatalytic, and as a result the pit grows in depth. At the outside, the reaction:

)

2Fe ( OH 2 + O 2 + H 2 O → 2Fe ( OH

)

3

(Eq 8)

further consolidates the isolation of the pit and impedes the ingress of oxygen. Pitting Corrosion Pitting corrosion is the most intensely studied and debated form of corrosion of stainless steel.

Pitting corrosion is important to designers because it is corrosion under conditions at which corrosion may not have been anticipated. Thus, it is both a materials selection and an environmental control problem. Its consequences may be only cosmetic, such as on a building or appliance facade, or potentially catastrophic, such as if leaks of toxic materials were to result from perforation. Stainless steels are designed to be passive, and localized corrosion is the local loss of passivity. Whether the consequences are major or not, it is always undesirable, and good design allows it to be avoided. What do we know for certain about pitting? We know quite a lot, really. Experts now conclude that since the early 1970s the local chemistry of pitting has been understood (Ref15). The greatest contributions to this field have been electrochemical studies. The tools of electrochemistry have been especially successful in elucidating the mechanism involved in pit growth and pit stability (Ref 16). The local environment within pits has been sufficiently measured and correlated with cavity geometry that some experts can say, “In a sense, all pitting is crevice corrosion” (Ref 15). This is to say that the electrochemistry of cavities such as pits and crevices is quite similar and has been well modeled. These same tools, however, have been much less successful in clarifying the mechanism of pit initiation, which is still the subject

40 / Stainless Steels for Design Engineers

Inclusions. The question of what causes the initial dissolution that causes both stable and metastable pits focuses on inclusions, which most authorities (Ref 18) have concluded are associated in some way with pit initiation. In the absence of inclusions, metastable pitting events are not noted, and the potential at which pitting occurs is the beginning of the transpassive regime. What are the typical inclusions in stainless steel? Inclusions in steel are normally the residue of normal deoxidation and desulfurization taken during steel refining usually done in an argon oxygen decarburization (AOD). After removal of the carbon, the subsequent objective is to remove or render less harmful the dissolved oxygen and sulfur, which if left in solution would later precipitate as low-meltingpoint iron compounds that would make the steel fragile and unworkable at high temperatures. Inclusions in stainless steel are typically oxides and sulfides. A key point to understand when considering inclusions as initiation sites for pitting is that inclusions are not simply inert debris but precipitates that are seeking thermodynamic equilibrium with the steel in which they have previously been dissolved. The reactions in stainless steel differ thermodynamically from those in carbon steel because of the presence of high chromium concentrations. This lowers the activity of oxygen and sulfur, making them more soluble, as Table 1 indicates (Ref 19). It also alters the efficiency of deoxidizing elements. Aluminum is a powerful deoxidant in carbon steel but is less effective in stainless, while titanium becomes a stronger deoxidizer in stainless. Their effect on sulfur is similar to that on oxygen. The bottom line is that oxygen and sulfur are generally removed by silicon/manganese deoxidation, but that this process occurs in both the liquid and solid states. That it carries over significantly into the solid state means that diffusion has a major role in determining if equilib-

of debate, possibly indicating that the root causes are more metallurgical than electrochemical. Figure 21 depicts a polarization curve for stainless steel in a chloride-containing solution. Pitting occurs in the zone in which passivity is expected. As potential increases, small spikes in corrosion current occur. These spikes measure local dissolution, called metastable pitting. Some such sites complete their dissolution and repassivate, while others continue to grow as stable pits. The potential at which stable pitting occurs is the pitting potential, while metastable pitting can occur at much lower potentials. Pitting events, stable or not, cause the generation of iron ions and local pH reduction. To the extent these remain concentrated in a small volume, they will affect subsequent events. The dissolution during metastable pitting is located at the matrix-inclusion interface. Different researchers assume dissolution of the inclusion, while others assume dissolution of the matrix. The dissolution parameters, as measured by current transients, depend on variables not of the inclusion chemistry but of the matrix composition, notably molybdenum and nitrogen levels (Ref 17), which is in keeping with the reduction in dissolution of the matrix that these alloying elements confer.

Fig. 21

Schematic of a passive anode polarization curve

Table 1 Typical values of activities and activity coefficients in liquid steels: activities in the 1 mass % solution: ai = fi . %i Metal

Carbon steel, 1600 °C Stainless steel, 1600 °C

%i fi ai %i fi ai

Al

C

Mn

P

S

Si

Ti

H

N

O

... 1.05 ... ... 3.6 ...

0.05 1.06 0.053 0.05 0.49 0.025

0.45 1.0 0.45 0.45 1.0 0.45

0.02 1.1 0.022 0.02 0.32 0.006

0.01 1.0 0.01 0.01 0.66 0.007

0.3 1.15 0.345 0.3 1.24 0.372

0.05 0.93 0.046 0.05 9.4 0.47

... 1.0 ... ... 0.93 ...

... 0.97 ... ... 0.17 ...

... 0.85 ... ... 0.21 ...

Cr

... ... ... 18 0.97 17.5

Ni

... ... ... 8 1.0 8.0

Chapter 4: Corrosion Types / 41

rium reactions occur and whether they go to completion. We will see that they do not. Oxide inclusions also are common. They are formed as the products of the reactions of silicon and manganese with dissolved oxygen. The thermodynamics of the reactions determine at any time how much oxygen can be dissolved in the steel at equilibrium. That equilibrium is easily achieved in the molten state, in which diffusion is very rapid, but achieved more slowly once the material has solidified. The inclusions in the solid state grow by the diffusion of oxygen to inclusion sites, where it precipitates as an oxide of silicon or manganese to the extent that these are locally present or of chromium when its local concentration (or more properly, its activity) makes it more favorable. These oxides are often the nucleation sites for manganese sulfide inclusions. Sulfur is a very surface active impurity that assists in weld penetration in stainless by virtue of its effect on weld pool circulation. Otherwise, it is a detrimental impurity, forming low-melting oxysulfides that diminish hot workability. Manganese is a strong sulfide former, and it is the main agent used to tie up sulfur. Manganese sulfide precipitates as an inclusion as a function of manganese and sulfur concentrations and temperature. Inclusions form not only in the molten metal but also in the solidified metal. The solubility, which is high in the liquid state, decreases on solidification, as seen in Fig. 22. Only resulfurized free-machining stainless steels have sufficient sulfur to precipitate manganese sulfide in

Fig. 22

Pseudo-binary-phase diagram for iron and sulfur at 1.8% manganese and 18% chromium

the liquid. At high sulfur and manganese concentrations, some manganese sulfides can precipitate during solidification interdendritically, while normal alloys with less than 100 ppm of sulfur form their inclusions after solidification. The distinction is important because precipitation in the liquid state permits rapid diffusion, which results in the most thermodynamically favorable species, manganese sulfide, to form. It may, and often does, nucleate on a preexisting inclusion, such as silicate present from the deoxidation process. In austenitic steels, manganese is generally present at a level of around 1.5% as a deoxidant and as a substitute for some nickel. An inclusion formed in the molten metal does not cause alloy depletion around it. One that forms or grows in the solid state does cause depletion of the elements that are precipitating, causing its growth. If manganese is lowered to very low levels, the supersaturation of sulfides is pushed to a lower temperature, at which lower diffusion rates hinder or prevent the precipitation. Thus, low-manganese alloys can be free of manganese sulfide inclusions even at somewhat high sulfur levels. Such alloys have elevated resistance to pit initiation. Lower manganese levels also thermodynamically reduce the chromium sulfide coprecipitation in inclusions, lowering chromium depletion around manganese sulfide/chromium sulfide inclusions. Elements more effective than silicon and manganese are now in use for deoxidation and desulfurization. These include aluminum, calcium, cerium, and other rare earth metals (REMs), and titanium. The action of calcium is notable. In a well-deoxidized and well-stirred melt and with a basic slag, calcium dissolved in the metal will react with dissolved sulfur to form calcium sulfide, which will be incorporated into the slag phase. Aluminum, while a potent deoxidizer, is less effective directly in desulfurization, but it can act indirectly by reducing a small amount of Ca2+ in the slag, allowing the formation of calcium sulfide. Titanium can sequester some sulfur as titanium carbosulfide precipitates. The greatest amount of sulfur removal is obtained by the addition of cerium or other REMs, usually in the form of the alloy mischmetal. These reactive elements typically form oxysulfide particles in the melt that may be trapped in the slag before metal solidification. Oxygen is normally dissolved in solidifying stainless steel, also at amounts in the neighborhood of 100 ppm depending on deoxidation

42 / Stainless Steels for Design Engineers

methods. Inclusions based on oxygen and sulfur formed in the liquid or during solidification are relatively large, greater than 1μ. As the alloy cools after solidification, precipitation continues since sulfur and oxygen are decreasingly soluble with temperature, to virtually nil at room temperature. This causes existing inclusions to grow and new ones to nucleate. This precipitation is similar to that which carbon undergoes in stainless, except carbon is generally not supersaturated until below 1200 °C at the highest in most alloys, whereas sulfur and oxygen are normally near saturation even at freezing or almost always when the solidifying ferrite transforms to austenite. Thus, inclusions grow via diffusion of oxygen and sulfur, which, as interstitials, diffuse much more rapidly than the silicon or manganese with which they have the greatest thermodynamic affinity. But precipitate they must, even if the silicon and manganese in the vicinity of their inclusion are exhausted. Thus, inclusions can grow with chromium substituting for either silicon or manganese as the precipitating partner for oxygen and sulfur. The inclusion growth necessarily depletes the surrounding region of reactants, silicon, manganese, and chromium (Ref 20). Inclusions thus formed are nonequilibrium in nature, and thermal cycles of steel production are rarely sufficient for the equilibrium to be attained. The chromium enrichment of such inclusions and corresponding chromium depletion of surrounding regions has been measured (Ref 21) and corresponds to the depletion seen next to chromium carbide precipitates at grain boundaries in sensitized alloys. These zones are altered in size and shape by thermomechanical processing in wrought alloys but exist fairly undistorted in welds. Hot rolling and cold rolling followed by annealing elongate manganese sulfide inclusions and flatten them, allowing depleted zones around the inclusion in the reduced dimension to be more rapidly homogenized during annealing. Thus, wrought material has better pitting resistance than cast or welded material. Inclusions that precipitate from the liquid, as is more the case for alloys solidifying in an austenitic mode, are at equilibrium with the surrounding matrix by virtue of the faster diffusion in liquids, do little to diminish the chromium content around them, and have a small effect on lowering pitting resistance. Pitting resistance is still affected to a degree by alloy depletion due to solidification segregation. However, if the alloy solidifies in a ferritic mode (FA, i.e., ferrite forming first on

solidification as opposed to austenite first, AF), as is almost always the case with commercial alloys, more sulfide precipitation happens in the solid state, pitting resistance is lowered proportionately to the sulfur level (Ref 22), and there is little negative effect from solidification segregation, as is shown in Fig. 23 and 24 (Ref 23). Solidification can also occur in a mixed ferriticaustenitic mode, in which case each microstructural component behaves according to the chart. The ratio of chromium and chromium-like elements molybdenum and silicon to nickel and nickel-like elements carbon, nitrogen, manganese determines the mode of solidification. It

Fig. 23

Influence of sulfur level on pitting resistance of unannealed welds for different solidification modes. Source: Ref 23

Fig. 24 Influence of sulfur level on pitting resistance of welds without homogenizing anneal. FA, ferrite forming first on solidification as opposed to austenite first, AF. Source: Ref 23

Chapter 4: Corrosion Types / 43

can also be altered by freezing rate. Faster cooling favors austenitic solidification. Long-term annealing of welds has shown that sufficient time and temperature to achieve some rehomogenization the alloy result in better pitting resistance (Ref 24), approaching that of the wrought alloy. Examination of the decreasing solubility of sulfur in stainless in Fig. 22 indicates that the precipitation of sulfides that cause chromium depletion occurs in delta ferrite on freezing when sulfur exceeds 0.007% and in austenite when sulfur exceeds 0.003%. Oxygen behaves in a parallel manner and is usually present in sufficient quantities, about 0.01% in manganese/silicon deoxidized steels, to cause the same phenomenon. This fundamentally is due to the high ratios of the diffusivities of oxygen and sulfur to chromium, which are about 10,000 and 680, respectively. Whenever fast-diffusing elements such as oxygen, sulfur, carbon, and nitrogen, which have a strong affinity for chromium and a solubility that decreases strongly with temperature, are present in steel, their precipitation will result in some degree of chromium depletion around the precipitation site because chromium diffuses too slowly to be replenished. The low chromium around inclusions is a sufficient condition for the local dissolution measured as metastable pitting, and if the depletion zone shape and size are favorable, then stable pitting would ensue. Certain other types of inclusions/precipitates are less harmful in this regard. Titanium, for instance, which is often added to form carbides and nitrides, also forms sulfides and oxides more strongly than manganese and therefore does so at higher temperatures. Such precipitates have a much lower tendency to allow chromium to join in the precipitation since the higher the temperature of precipitation the more that diffusion allows the more favorable reaction to occur. Rare earths also behave the same way. Metastable pitting is diminished by the presence of these elements. The initiation of pitting is also affected by stress and inclusion orientation (Ref 25), which the researchers correlated to the dimensions of the inclusion-derived cavity being able to sustain a sufficiently low pH due to iron dissolution to maintain stable pitting. The influence of stress was to cause cracking at otherwise unfavorably shaped inclusions, which then provided a crevice capable of sustaining stable pitting. This will be relevant to later discussions of

stress corrosion cracking (SCC). There have been numerous proposed mechanisms for the breakdown of a passive film in chloride-containing media; these have been summarized in other publications (Ref 23). These hypotheses deal with how a passive film on a homogeneous surface could break down. They include: • Adsorption of chloride ions • Penetration of the passive film by chloride ions • Film breakdown by electrostriction • Formation of stable metallic chlorides • Coalescence of cationic vacancies • Random localized thinning of the passive film • Local variations in the composition of the corrosive medium By and large, these mechanisms presuppose a stainless steel surface that is homogeneously passive and try to explain the observed inhomogeneous behavior of the passive film. However, since it is clear that the surface is not homogeneous, especially with regard to the passive film, these hypotheses are not necessary to explain the behavior of everyday stainless steels, which unfortunately have abundant inclusions and chemical inhomogeneities capable of locally diminishing the integrity of the passive film. More research in understanding the exact nature of the inhomogeneity of stainless steel surfaces is necessary for a complete understanding of pit nucleation and therefore prevention. Pitting Resistance. Pitting has been extensively correlated with environment and compositional variables. The most well-known and useful correlations are between the PREN and the critical pitting temperature (CPT) and by extension to the pitting potential. For austenitic alloys: PREN = % Cr + 3.3 % Mo + 30 % N

(Eq 9)

For ferritic alloys, which hold no nitrogen in solution: PREN = % Cr + 3.3 % Mo

(Eq 10)

For duplex alloys, which have two phases, neither of which matches the bulk composition: PREN = % Cr + 3.3 % Mo + 16 % N

(Eq 11)

These equations are useful, if approximate, and their correlation is shown in Fig. 25 (Ref 26) They do not include tungsten, which, if

44 / Stainless Steels for Design Engineers

Fig. 25

Variation of critical pitting temperature with pitting resistance equivalent number (PREN) of austenitic steels in water plus 6% FeCle. Source: Ref 26

Fig. 26

Differential variation of critical pitting temperature of several stainless steel alloys for unwelded wrought and welded material. Source: Ref 13

present, has half the effectiveness of molybdenum. They neglect carbon, which seldom varies enough to have a visible effect but has been shown when in colossal supersaturation to have a factor of about 10, not unlike nitrogen, another interstitial that it resembles in solution thermodynamically (Ref 27). It also does not include the negative influence of elements such as sulfur. Likewise, the equations cannot deal with inhomogeneity issues, so welded alloys have different CPTs for the same PREN (Fig. 26) (Ref 13). These equations are all-other-thingsbeing-equal equations and are useful for gross

alloy behavior predictions. It is noteworthy that the elements copper and nickel, which are beneficial against uniform corrosion and which slow the growth of pits by this same action, do not contribute to increasing the resistance to the onset of pitting. This is another manifestation of pitting initiated by the local stability of the passive film, which is primarily a function of local chromium content. Molybdenum and nickel thus seem to bolster local chromium content in the passive film. Nitrogen seems to act by concentrating at the passive film-alloy interface rather than by buffering the solution by ammonia formation, which has been proposed (Ref 36). Research on very pure sputtered films of iron-chromium alloys have demonstrated that both titanium and niobium in solution diminish active dissolution, assist repassivation, and improve pitting resistance (Ref 29). In most practical cases, these elements are not found in solution because of their affinity for oxygen, sulfur, carbon, and nitrogen, with which they form compounds. It should also be noted that the critical PREN values vary with crystallographic structure. Ferritic alloys require somewhat lower PREN values to exhibit similar pitting resistance as austenitic alloys of somewhat higher PREN. While pitting is of great theoretical and practical interest, there are significant problems in actually conducting good tests. Monitoring of the electrochemical potential during the test is considered mandatory by most researchers. How is a metallic sample suspended in a solution without creating any crevices and without exposure at the liquid-gas interface? The development of the flooded gasket technique (used in ASTM G150) was a milestone, but it also has some problems—most notably the potential for dilution of the test solution, especially during prolonged testing. FeCl3 testing benefits from the fact that the solution creates a reproducible positive potential. While the PREN approximates the pitting resistance of an alloy, there is a standard test by which the CPT is measured. Pitting in a given medium capable of causing pitting does not occur below a temperature that is characteristic of the medium and the material, with the myriad exceptions of stress state, surface finish, microstructure, etc. The most commonly used test media are the unacidified 10% FeCl3, which is used in the ASTM G 48 practice B, and the 3.5% NaCl solution of the ASTM G 150. The latter, if modified to 0.1N NaCl, allows the ECPT, the

Chapter 4: Corrosion Types / 45

electrochemical pitting potential, of lower alloys such as 304 to be measured (Ref 30). Crevice Corrosion In the case of pitting, the geometry that makes up the pit is essential in creating the differential aeration cell and to cause the autocatalytic dissolution process. In many cases, a geometry that retains and acidifies water is already present in crevices in different types of structures such as gaskets, under faulted coatings, under bolt or screw heads, etc. Crevice corrosion occurs because zones have restricted access of reactants and restricted exit of corrosion products. It is especially the inhibition of the cathodic reaction inside the crevice by the dearth of oxygen, which sets up a more aggressive environment within the crevice than without. The interior reactions become increasingly anodic, and the aggressiveness of the environment can reach a threshold at which active corrosion occurs, while the situation exterior to the crevice is safely passive. Crevice corrosion occurs at lower temperature than pitting in the same environments, so it is a greater danger in that sense. The relationship between the alloy content, given as the crevice corrosion resistance equivalent number (CCREN), and critical crevice corrosion temperature (CCT), shown in Fig. 27 (Ref 11), is similar to that of PREN (PI) to CPT except for the molybdenum factor being more important:

Fig. 27

Cl = %Cr + 4.1%Mo + 27%N

(Eq 12)

Since a crevice has a preexisting favorable geometry for pit growth, any pitting event, metastable or stable, can initiate ongoing crevice corrosion. Crevices are thus incubators for corrosion triggered by metastable pitting events. The dissolution of iron during passivation itself as well as the differential oxygen cell created by the crevice contribute to the process. It is logical to think that alloying the elements that contribute to lowering the critical current density for passivation and the uniform corrosion rate, such as nickel, would reduce the creation of the reactants that start the crevice corrosion process, but this presumed effect is not strong enough to be reflected in this actual behavior Eq 12 represents, although it is generally acknowledged that austenitic steels perform better than ferritic steels in the absence of molybdenum. Materials are characterized as having a critical depassivation pH. If crevice conditions are such that the reactions over time allow the pH to be reduced to this level, then active corrosion will begin within the crevice. Thus, passive film stability seems to be the critical factor rather than corrosion rate after initiation. Preventing Crevice Corrosion. The countermeasures against crevice corrosion are cathodic protection, design, maintenance, and, of course, alloy selection. Designing to avoid crevices should include maximizing the volume of unavoidable crevices, engineering flow to enhance transport in and out of crevices, and

Variation of critical crevice corrosion temperature with alloy content

46 / Stainless Steels for Design Engineers

avoiding stagnation. Any maintenance or design procedure that prevents formation of deposits is beneficial. Welds are particularly vulnerable surface sites, so any combination of welds and crevices or crevices caused by poor weld geometry must be avoided. S32205 is a benchmark alloy of sorts. It has just sufficient alloying to resist pitting in seawater, but it is susceptible to crevice corrosion. As a practical matter, crevices are almost impossible to eliminate. Threaded fasteners and joints represent severe crevices and should be avoided in aggressive environments if possible. Gasketed joints are another severe crevice location, and their usage should be curtailed to the minimum practical extent. In these situations, judicious use of very expensive, highly corrosion resistant materials is justified. The use of smooth welded joints is thus generally preferred. In a more general consideration, deposition and fouling create crevice sites, and design and operational controls to preclude the formation of deposits and the prompt removal of sludge and the like are necessary. But in some situations, such as marine exposures, biofouling will create crevice sites. This fouling may be macroscopic, such as from shellfish and barnacles, or it may be microscopic. Microscopic biofouling causes the special form of crevice corrosion called microbiologically influenced corrosion (MIC) discussed in a separate section). Sensitization/Grain Boundary Corrosion The maintenance of a passive layer in a wide range of pH conditions in stainless steels is dependent on the alloying elements, primarily

Fig. 28

Schematic illustration of sensitization due to chromium-rich precipitates that deplete adjacent regions of chromium. GB, grain boundary

chromium. In the various grades of stainless steels, there are many intermetallic phases that are thermodynamically stable but kinetically slow to precipitate that are enriched in chromium. An example of such a phase is chromium carbide (Fe, Cr)23C6. These phases tend to form at grain boundaries where nucleation is favored, resulting in a depletion of chromium in the adjacent regions, as shown in Fig. 28. Thus, the chromiumdepleted regions near the grain boundaries are sensitized in that they behave as active anodes compared to the larger interior of the grains that are still passive. In an aerated corrosive environment, the smaller chromium-depleted nonpassive anodes dissolve, whereas the larger cathodes reduce oxygen, resulting in a localized corrosion along grain boundaries. Any heat-treating or welding procedure of stainless steels should thus be tailored to avoid sensitization. When a stainless steel is heat treated, there is a risk that the unwanted phases may form, depending on the time-temperature history and precipitation kinetics of the unwanted phase. Figure 29 shows schematically the temperature versus time due to welding and the resulting sensitization. Figure 29 shows a TTT (timetemperature-transformation) curve for precipitation of the unwanted phase. Near the weld (A), the time spent in the temperature region where precipitation occurs is too short, whereas far away from the weld (C) the temperature experienced is too low. At location B, there is, however, a risk for sensitization. Austenitic. Sensitization can occur at any temperature at which carbon is supersaturated in an alloy. Current austenitic stainless steels have carbon levels of under 0.10% normally

Fig. 29

Schematic illustration of how a heat treatment relates to sensitization due to precipitation kinetics. TTT, time-temperature-transformation

Chapter 4: Corrosion Types / 47

and under 0.03% for low-carbon L grades. Thus, normal grades sensitize below around 800 °C. The supersaturation increases with decreasing temperature, but below about 500 °C diffusion of carbon is too slow for carbon to move to grain boundaries and cause the damaging combination with chromium that causes sensitization. Low-carbon grades avoid sensitization because they are not sufficiently supersaturated at temperatures at which carbon is mobile enough to diffuse to grain boundaries. Ferritic. Another situation exists in ferritic stainless steels, in which carbon is much less soluble but is much more mobile. Annealing over 900 °C can put enough carbon in solution to cause sensitization even at the lowest carbon levels attainable in an AOD and even at the fastest possible quench rates. The damaging chromium depletion caused by this very rapid precipitation can be undone by a simple rehomogenization anneal of the remaining chromium. This is theoretically possible with austenitic alloys also, but the diffusion rates of chromium in austenite as so slow that it is impractical in most real cases. Duplex steels have a subtle near immunity to carbide sensitization. While they are typically low carbon anyway, the carbides that do form do so at ferrite-austenite grain boundaries. Here, chromium is consumed from the chromium-rich ferrite phase, leaving the austenite intact. Their large grain boundary area keeps carbide concentration per unit area low, and the fast diffusion in the ferrite keeps austenite from becoming depleted. However, the rapid formation of intermetallic phases at the ferrite-austenite interfaces can lead to a rapid loss of corrosion resistance and a severe loss of toughness if exposure to temperatures within the intermetallic precipitation range is not controlled. Martensitic steels are quenched as austenite to and through the Ms temperature without time for carbon to precipitate in austenite. The carbon in the martensite can precipitate and cause sensitization if reheated to the 300 to 700 °C region. Fortunately, heating to above 700 °C rehomogenizes the chromium and eliminates sensitization. Effect of Alloying. Besides determining basic phase structure, alloying plays a role in susceptibility to sensitization. Those elements that reduce the tendency of chromium carbides to form also reduce the susceptibility to sensitization. This is a purely thermodynamic effect. Molybdenum, silicon, and nickel promote carbide formation by increasing the thermodynamic

activity of carbon and make alloys more susceptible. Nitrogen lowers the tendency for carbide formation and slows sensitization. Nonthermodynamic effects are those of austenite grain size and prior cold work. Decreasing grain size and therefore increasing grain boundary surface area decreases the amount of precipitate per unit area of grain boundary and therefore the amount of chromium depletion per unit area. Cold work accelerates diffusion and makes precipitation more rapid, thus aggravating sensitization. The thermodynamic affinity tool can be used to prevent chromium carbide formation in another way. Introducing alloying elements that combine with carbon more strongly and rapidly than chromium can exhaust the supply of carbon available to precipitate as chromium carbide. There are a number of candidate elements, zirconium, vanadium, tantalum, niobium, and titanium, most prominently. Of these, the diffusivity and affinity for carbon of niobium and titanium make them the best for this purpose. Each forms stable carbides at much higher temperatures than chromium, starving chromium of sufficient carbon to form damaging precipitates. The caveat with titanium is that it forms oxides, sulfides, and nitrides preferentially to carbides. Therefore, sufficient quantities must be used to accommodate the prior formation of these phases. Niobium tends more toward carbide than nitride formation but is a weaker carbide former than titanium. The solubility products of these precipitation reactions are: log [Ti ][C ] = 2.97 −

6780 T

(Eq 13)

9350 T

(Eq 14)

log[ Nb][C ] = 4.55 −

These equations follow the form of the general equation for precipitation reactions: log[ M ][ X ] = A − H / RT

(Eq 15)

in which A is a constant, H is the heat of dissolution, R is the gas constant, and T is the absolute temperature. If the amount of titanium or niobium is stoichiometrically sufficient, no carbon will form chromium carbides under equilibrium conditions. It is possible to defeat the stabilization reactions by quenching the alloys from temperatures at which titanium carbide or niobium carbide is dissociated. If free carbon is

48 / Stainless Steels for Design Engineers

left free in the matrix by quenching, then on reheating it may form carbides with the most locally accessible favorable element, such as chromium, rather than the most thermodynamically favorable element, which would be titanium or niobium. This can occur when a stabilized alloy such as 321 is welded. A zone away from the weld may experience a high enough temperature to put carbon into solution and then cool just rapidly enough to not form only the equilibrium titanium carbide but also Cr23C6 at grain boundaries, causing the type of sensitization called knife-line attack. This problem has nearly ceased to exist as modern 321 has low levels of carbon and nitrogen for economic reasons; this effectively precludes this chromium carbide precipitation in most cases. Welding. Many of the most severe problems of sensitization arise when stainless steels are welded to carbon or low-alloy steels. In these situations, construction code rules usually require that the carbon steel component be given a stress relief annealing (SRA) treatment. Such SRA treatments are typically in the sensitization temperature range for austenitic stainless steels. Use of low-carbon or stabilized grades is necessary in such cases. Even then, use of the lowest allowable temperature SRA treatment for the shortest allowable time is preferred.

Corrosion Combined with Fatigue or Fracture Environmentally induced failure occurs when brittle failure under tensile mechanical loading occurs at a lower stress when a material is subjected to a corrosive environment than what would happen in a noncorrosive environment. This introduces us to what is perhaps the most controversial technical subject in all of stainless steel research, SCC.

layer occur, and region II, where the protective layer is not fully developed, suggesting an appreciable electrochemical effect. The latter is a zone that exists in alloys that have zones of chromium depletion. Stress corrosion cracking has always been among the most controversial subjects among metallurgists and electrochemists. The debate centers on whether the critical mechanism is dissolution or fracture, and if a fracture, by what mechanism. Is the cracking zone locally softened, locally hardened, transformed, to a more brittle phase or embrittled by hydrogen? As of this writing, there is no general agreement on which type mechanism is the fundamental cause, but there is room for convergence. Obviously, elements of many may come into play. It is likely, as in most prolonged arguments, that no hypothesis is completely correct. We will try to fairly set out what is known and agreed on as fact and then present researchers’ views in an unbiased manner, but since we concern ourselves only with stainless steel, no attempt is made to address an all-encompassing theory. Crack Initiation. In stainless steels, cracks can be seen to initiate at surface defects and irregularities. In stainless steel, it must be agreed by all that the preponderant initiation site is a corrosion pit or, in some cases, a crevice. Intergranular corrosion sites, as are seen in sensitized material, can also provide the conditions for SCC initiation. The interrelationship between pits and SCC cracks has been studied (Ref 25). Stress lowers the anodic potential at which pitting occurs and permits metastable pits

Stress Corrosion Cracking The key cause for SCC is the cooperating effects of tensile stress and a corrosive environment. Such cases can be identified in most alloy systems, and even pure metals, which were thought to be more or less immune, also have had cases of SCC reported. In passive metals, two sensitive potential regions for the occurrence of SCC have been identified and are shown in Fig. 30: region I, where pitting and breakdown of the passive

Fig. 30

Zones of susceptibility to stress corrosion cracking

Chapter 4: Corrosion Types / 49

to become stable via the generation of cracks. Cracks, once formed, presumably have favorable geometry to duplicate pit internal chemical reactions and must be considered to be described by the models that apply to pits and crevices. The stress at which SCC initiates has a threshold, which has been reported as between 25 and 50% of the yield strength in austenitic stainless steel. The temperature at which SCC is initiated ranges from ambient to under 100 °C for martensitic materials, while austenitic alloys begin their sensitivity above room temperature and increase in susceptibility with increasing temperature. The ferritic steels, while considered nearly immune to SCC, have their maximum susceptibility in the same range as martensitic steels. In environments of mixed chlorides and sulfides, however, SCC can occur in all types of stainless at room temperature. This has been seen in the SCC of austenitic stainless steel in swimming pool environments, in which chloride ions can condense on the stressed steel and cause pitting and SCC. Cracks propagate very slowly below specific certain stress intensity levels, but once that intensity is reached, they have a plateau rate that is fairly constant until the stress level at which catastrophic failure occurs at very high propagation rates. Rates of crack propagation are exponentially increased by increasing temperature. The crack propagation rate has been seen across a range of alloys to be linearly proportional to the average current density that alloy experiences when its surface is strained, indicating that reactions at the crack tip are strain sensitive, and overall rate limiting, but not necessarily the mechanism of cracking. Crack growth is discontinuous with individual steps of growth many times the average rate, which is similar to that seen with gaseous hydrogen embrittlement (HE). The crack growth gives off acoustic emissions as cracking steps occur. These steps of growth are brittle and are seen as facets on fractographs with cleavages corresponding to crystallographic planes. The crack facets match with high perfection, showing almost no evidence of plastic deformation or dissolution. The propagation path may be intergranular or transgranular. Grain boundary propagation in stainless steels usually corresponds to conditions under which grain boundaries are less corrosion resistant because of either material or

environmental variables. The most common example is that of sensitized 304 in high-temperature water or caustic media. The relevance of this to the normal case of stainless steels must be questioned since, by definition, the sensitized grain boundaries themselves can be depleted of chromium to a degree they are not stainless and have a much less stable austenitic structure, having their martensite start temperature Ms, raised by the loss of chromium. Material Variables. Martensitic stainless steels and martensitic precipitation hardened stainless steels are quite susceptible to SCC. This susceptibility increases with hardness, yield strength, and embrittling heat treatments. They will crack at threshold stresses equal to 50% of yield strength. These alloys can be tempered at sufficiently high temperatures that they become soft and tough enough to have very good resistance. Ferritic stainless steels of low and medium chromium are generally not susceptible to SCC. Ferritic alloys, which can have a martensitic structure, should be considered martensitic for SCC purposes. If purely ferritic alloys are alloyed with copper, molybdenum, and nickel, they can become susceptible. The presence of α'( or high-temperature embrittlement also increases susceptibility, as does cold work. Despite the controversy surrounding the mechanism of SCC in austenitic stainless steels, there is almost complete agreement that SCC of body-centered cubic (bcc) stainless steels, martensitic, ferritic, and pH is simply a manifestation of HE, with hydrogen provided by either anodic (e.g., active corrosion within a pit) or cathodic reactions. Duplex stainless steels have low susceptibility to SCC. Their dual-phase microstructure ensures that under conditions that crack austenite, ferrite remains as a crack-arresting phase, while under conditions that cause SCC in highly alloyed ferrite, the austenite is a crack arrester. A second explanation of the resistance of duplex alloys to SCC is that their two phases have different corrosion potentials, and that the mixed potential that arises because they are in intimate contact is outside the potential range for SCC on either phase. This fits with the resistance to SCC of wrought alloys with a lamellar structure and the lesser resistance of cast alloys that lack that structure. Austenitic stainless steels are the type of stainless steel generally associated with SCC, and they vary in their degree of susceptibility to

50 / Stainless Steels for Design Engineers

SCC. All other things being equal, alloying elements that delay or prevent localized corrosion do the same to delay SCC. This is simply the delay of initiation. However, if pitting can be delayed indefinitely, then SCC can also, assuming, of course, more harmful localized corrosion, such as that due to intergranular chromium depletion, is not occurring. Molybdenum, which we already know helps prevent pitting and crevice corrosion, also increases the threshold stress for SCC, as shown in Fig. 31 (Ref 31). But, if metastable or stable pitting is occurring, the threshold stress has been reached, and the temperature is sufficient, then SCC will proceed. It is mitigated by material variables such as cold work and by alloying elements that increase austenite stability. Many publications cite nickel as beneficial in enhancing resistance to SCC, often referring to the data from Fig. 32. However, its role seems mainly to be as an austenite stabilizer and as a retarder of active corrosion. The minimum in the curve corresponds to the nickel level at which the structure is entirely austenitic, but least stably so. Lower nickel levels produce better immunity through the duplex structure, while higher levels promote austenite stability and correspond to alloys having more alloying elements, such as chromium and molybdenum,

Fig. 31

Influence of molybdenum on resistance to stress corrosion cracking (SCC) in austenitic steels

which are probably the greater cause of resistance to SCC. Environmental Variables. There are three key types of environments in which SCC occurs in stainless: • Chloride-containing solutions • Caustic solutions • Polythionate and thiosulfate solutions The cases of polythionate and thiosulfate solutions are industrially important but can be adequately explained as simply the stress-assisted intergranular corrosion of sensitized material. Hot caustic solutions are aggressive against stainless steels. Certain combinations of concentrations, temperature, impurity, and dissolved oxygen can cause SCC as well as other undesirable corrosive attack. Resistance to general corrosion is proportional to nickel content, but ferritics and duplex alloys are less prone to SCC. 304 has been reported to have no meaningful threshold stress for SCC in hot caustic solutions, leading one to question whether such a failure should even be classified with SCC of the typical chloride-induced type or belong with the previous polythionate and thiosulfate solutions. Chloride containing environments are the main ones that induce SCC. Water can cause

Fig. 32

Variation of resistance to stress corrosion cracking with nickel (and other) content and structure

Chapter 4: Corrosion Types / 51

SCC at sufficiently high temperatures (i.e., above 100 °C) if there are even very low combined concentrations of chloride (greater than 0.1 ppm) and oxygen (greater than 0.1 ppm) dissolved (see Fig. 33) (Ref 32). Failure times decrease exponentially with decreasing chloride content. Crack growth rate increases by a factor of ten with each 30 °C rise in temperature. Decreasing pH lowers the temperature at which SCC occurs in a given time. Mechanisms. There have been many mechanisms proposed for SCC in stainless steel. We focus only on those that address the failure in chloride-containing media, the main concern for users of stainless steel. The models that have found some support are: • Slip dissolution • Adsorption-enhanced plasticity • Adsorption-induced brittleness • Hydrogen embrittlement Slip dissolution (anodic dissolution) was the earliest proposed model for SCC. It simply proposes that at a crack tip a passive film forms, and after time it fractures by an unspecified mechanism. The fresh active surface may or may not repassivate, after which the process repeats itself. The strength of this model is that it actually does describe what is happening. The crack

Fig. 33

does advance discontinuously, and after each advance there is fresh surface, which comes into equilibrium with the solution within the crack. So, any experiment, such as that shown in Fig. 34 (Ref 33), that tests crack propagation against electrochemical events will absolutely support this model. It is axiomatic that films must rupture and reform as cracks advance discontinuously. The weakness of this model is that it does not provide a mechanism for brittle fracture, and the very brittle features of transgranular SCC fracture surfaces do not show any supporting evidence of dissolution. Research (Ref 34) showing that metal dissolution at the crack tip is isotropic rather than crystallographically oriented make dissolution models incapable of being reconciled with the crystallographic fracture surface facets. Adsorption-induced brittleness, also known as stress-sorption, looks to the parallels between liquid metal embrittlement and SCC to explain the mechanism of SCC as the action of adsorbed species weakening atomic bonds on the crack tip surface. If the action is on the surface, however, the mechanism cannot produce the observed discontinuous, brittle cracks which characterize SCC. Only in alloys such as Fe-3Si are steps small enough to make this mechanism plausible.

Variation of susceptibility to stress corrosion cracking (SCC) with media oxygen and chloride content for 304 stainless steel. Source: Ref 32

52 / Stainless Steels for Design Engineers

Fig. 34

Crack propagation rates of various metals plotted versus current density. Source: Ref 33

Adsorption-enhanced plasticity/hydrogen embrittlement encompasses a number of models that observe that adsorbed species enter the lattice in the vicinity of the crack tip and then cause failure by one of several mechanisms: • Dealloying and porosity • Adsorption-induced brittleness • Coalescence of voids formed by cross slip enhanced by the adsorbed species Since hydrogen is the only species that is produced in quantity and is capable of diffusing into the lattice, HE is implicit in all these models. All of the models have support in that they have some experimental observations that show that the phenomena they propose as causal

actually take place, but none is specific enough to have been tested by critical experiments to prove or disprove it. It has been demonstrated that hydrogen is absorbed into the material at the crack tip. The main question is whether it causes damage by creating porosity, altering dislocation mobility, or causing lattice decohesion. There is support for each. It has been observed that where SCC occurs there is a large concentration of vacancies. This has led to speculation that porosity is a weakening mechanism responsible for SCC (Ref 35). It has been proposed and supported by calculations that hydrogen lowers the energy required or vacancy formation. The lowest energy

Chapter 4: Corrosion Types / 53

configuration is calculated as two hydrogen atoms per vacancy. This pairing of hydrogen solute atoms to dislocations is very reasonable given the major distortion the interstitial hydrogen causes to the lattice, so there is no basis to challenge the enhanced vacancy formation. Whether the effect is large enough to cause failures has not been demonstrated. The largest measurable effect of hydrogen has been a slight acceleration of stress relaxation in martensite. The relevance of hydrogen-induced vacancy agglomeration as the principal cause of failure must be considered questionable until some further critical experiments link the vacancies to the observed instances of failure quantitatively, and more important, to show how this mechanism could account for the temperature and stress dependence observed. The major influence of vacancy formation due to hydrogen may be to enhance the volume expansion due to hydrogen. Adsorption-induced brittleness, also known as stress sorption, looks to the parallels between liquid metal embrittlement and SCC to explain the mechanism of SCC as the action of adsorbed species weakening atomic bonds on the crack tip surface. If the action is on the surface, however, the mechanism cannot produce the observed discontinuous, brittle cracks that characterize SCC. Only in alloys such as Fe3Si are steps small enough to make this mechanism plausible. In stainless steels, there seems to be nothing to support this proposed mechanism. Adsorption-enhanced plasticity has become known recently as HELP or hydrogen-enhanced localized plasticity. The underlying mechanism at work in this model is the hydrogen-induced shielding between microstructural defects. This has been observed distinctly in single crystals of austenitic stainless alloys. The Cottrell atmosphere of hydrogen around dislocations causes mutual repulsion, causing strain to be localized on certain slip systems. This has been observed to occur and has caused deformation to become concentrated in Luders bands in austenitic alloys, which of course do not show such behavior without hydrogen (Ref 36). This also produces ε-martensite in austenitic alloys, which would be considered stable without hydrogen and deformation. This theory encounters a problem, however, with the fact that the same studies showed that hydrogen actually strengthens the matrix by solid solution hardening. It acts in much the same way as carbon and nitrogen do, as shown

Fig. 35

Stress-strain curve for single crystals of stable austenitic stainless steel with and without hydrogen. Source: Ref 36

in Fig. 35 (Ref 36). All these interstitials strain the lattice and therefore harden in proportion to their atomic size. Hydrogen, as the smallest of them, has about half the distorting effect and half the hardening effect. But, its small size makes it mobile at ambient temperatures, so it can diffuse to sites where it can alter mechanical properties. But, while hydrogen causes dislocation motion and lower work hardening, it does not weaken austenite, so this theory by itself cannot account for the role of hydrogen in SCC and, by inference, in HE in the more general case. The quandary of hydrogen finally having been shown to have a clear effect on mechanical properties but having that not account for either SCC or HE may be put to rest by the additional observations of hydrogen’s role as a lattice distorter (Ref 37). While not formalized as a proposed hypothesis for SCC, the role of hydrogen as a generator of very high stresses has been pointed out as a factor that cannot be neglected when evaluating other proposed mechanisms. Hydrogen has been shown to distort the lattice in proportion to its concentration. The effect is not small, accounting for about 1% strain per 0.1% concentration by weight, as shown in Fig. 36 (Ref 38). At hydrogen levels of over 1000 ppm, which are thought to exist around growing SCC crack tips, there could therefore be hydrogen concentration gradients capable of producing additional tri- or biaxial stresses on the matrix ahead of the crack tip that may approach the yield stress and account for some or all of the difference between the normal fracture toughness KI and the KISCC, that for SCC. This also precludes the necessity of hypothesizing hydrogen-induced phase changes, although

54 / Stainless Steels for Design Engineers

Fig. 36

Dilation of austenite due to hydrogen in solution. Source: Ref 38

were they to exist, they would result in the same lattice expansion. In both cases, the failure would occur at a region ahead of the crack tip and beyond the highest hydrogen concentration, which is what is observed to occur. The growth of this stress over time with increasing hydrogen-producing corrosion would account for the observed kinetics, locus, and stress dependence of SCC. If nothing else, the main models for SCC and the experimental results on which they are based should be reexamined in view of the fact that the stresses induced by hydrogen are not negligible and, in fact, may account for much of the observed SCC behavior of stainless steels. The next few years may finally see the resolution of the lengthy debate over the causes of SCC. If it comes, it will be from critical experiments, which can quantitatively differentiate among the above effects and measure the contribution of each. Hydrogen Embrittlement Like SCC, there has been debate about HE that has produced more heat than light. This in-

volved distinguishing among the same mechanisms, namely: • Decohesion • Enhanced local plasticity • Adsorption embrittlement • Void coalescence The identification of the operative mechanism for HE involves again distinguishing what role each of the above contributes to HE in a given situation since all are known to be real metallurgical phenomena. The main difference between HE and SCC in stainless steel is that HE is limited to ferrite, which is hardened by cold work or alloying, and martensite. Austenite is somewhat diminished in ductility by hydrogen, but not subject to the completely brittle, discontinuous cracking of bcc stainless. The observations that make a given model plausible as a mechanism for SCC lack traction for the same materials in HE. It hard to envision enhanced plasticity involved in the completely brittle fracture of high-strength martensitic stainless steels, whereas void coalescence by vacancy creation seems more likely to account for the observed behavior.

Chapter 4: Corrosion Types / 55

The resolution of mechanism here also must account for the contribution of hydrogen-induced stress as well as hydrogen effects on mechanical processes, especially since the observed susceptibility to HE is proportional hardness, therefore to the amount of hydrogen a given material can hold both in normal interstitial solution and the amount it can trap at lattice defects (Ref 39), especially the dislocations within the plastic zone at the crack tip, which provide enhanced hydrogen solubility where it can aggravate the applied crack opening with a wedge effect from hydrogen dilation. Corrosion Fatigue Just like SCC, corrosion fatigue causes brittle failure under a combined environment of corrosion and a tensile stress component. The stress, however, is cyclic and in a test of stress versus number of cycles (S vs. N), failure will occur at a lower N under the corrosive environment. The cracks are transgranular, and the collaborative effect of corrosion and fatigue is that corrosion accelerates the plastic deformation that accompanies the evolution of extrusions and intrusions. In corrosion fatigue, an obvious pit corrosion site may not be necessary because of the combined action of cyclic stresses and the environment. However, an initiation site that is the weakest link in a combined mechanical and metallurgical sense will be the initiation point after which conditions that may not cause SCC can help propagate fatigue cracking at lower stresses than would be expected in more benign environments and in environments that may not cause SCC or pitting under static loads. The importance of the environmental interaction is reflected in the sensitivity to frequency of stress application. High-frequency loading gives less time for corrosive attack and brings crack propagation rates down closer to those in air. In some materials, crack propagation rates are elevated above those in air at all stress levels, while in others a threshold stress intensity must be reached before an acceleration is noted. Some materials show a combination of both. The first case seems to be merely fatigue assisted by corrosion, while the last two seem to indicate an SCC–type behavior. The same uncertainties that cloud our understanding of SCC necessarily disguise the precise mechanism of corrosion fatigue, which must be viewed as a combination of SCC and fatigue.

Biocorrosion and Microbiologically Induced Corrosion There are many cases for which biological organisms contribute to initiating or enhancing rates of corrosion. This can occur in natural environments such as ground or seawater as well as domestic and industrial environments such as the nuclear and chemical processing industries, for example. This is called biocorrosion or MIC, microbiologically induced corrosion. The bacteria that are known to influence corrosion can be sorted as aerobic bacteria that lie in aerated water and anaerobic bacteria. Among the anaerobic bacteria that are known to [16] (Ref 40) affect stainless steels can be counted: Desulfibrio and Desulfotomaculum. Both of these are so-called sulfate-reducing bacteria (SRB), which means that they promote the reaction: SO 24− → S2− + 4 O

(Eq 16)

which in turn accelerates the cathode reaction: 2H = + 2e − → 2H

(Eq 17)

Aerobic bacteria flourish under oxygen (Ref 40). Examples are the iron-oxidizing Gallionella and Sphaerotilus, which increase the anode dissolution reaction: Fe → Fe 2+ = 2e −

(Eq 18)

by converting the ferrous iron-ion product (Fe2+) to less soluble ferric (Fe3+). Due to this, macroscopic so-called tubercules form that can cause crevice-type shelters where differential aeration and pit initiation can occur. Countering MIC with biocides can cause problems in manganese-containing waters. Oxidizing biocides, such as ozone, chlorine, or peroxide, can cause manganese to be oxidized to manganese dioxide. The precipitated deposits of manganese dioxide can accelerate pitting corrosion even in low-chloride waters in which alloys such as 316 would otherwise be safe from pitting attack. Biocorrosion is most commonly encountered in ambient aqueous environments, which are the environments in which most microorganisms have evolved to thrive. So, it tends to be a problem for the medium-alloyed steels such as 304 and 316, which are used in these environments.

56 / Stainless Steels for Design Engineers

The more chemically or thermally hostile environments in which higher alloyed grades are used are also hostile to bioorganisms and thus minimize the problem. The development of microbiological consortia allow anaerobes to flourish under biofilms that form in an aerated environment. These represent a differential aeration cell that acts just like a severe crevice. Also, the action of microbes in raising the corrosion potential is key to understanding why natural seawater is so much more corrosive than sterile sodium chloride or synthetic seawater solutions. And, macrofouling organisms are important. They create crevices and sites where microfouling can start early. At the same time, they are sources of turbulence in flowing systems, and this turbulence can cause flow erosion in copper materials, making use of stainless steels more attractive. REFERENCES

1. M.P. Ryan et al., Critical Factors in Localized Corrosion, Proc. Electrochem Soc., Vol 150, 2003, p 284–294 2. K. Sieradski and R.C. Newman, J. Electrochem. Soc., Vol 133, 1986, p 1980 3. L. Brewer, Science, Vol 161, 1968, p 115 4. W.J. Tobler and S.Virtanen, Critical Factors in Localized Corrosion, Proc. Electrochem Soc., 2003, p 583–594 5. B. Baroux et al., Corros. Sci., Vol 47 (No. 5), 2005, p 1097–1117 6. http://www.alleghenyludlum.com/pages/ products/xq/asp/T.1/qx/productCategory. html 7. K. Kimura et al., High Cr Stainless OCTG with High Strength and Superior Corrosion Resistance, JFE Technical Report 7, Jan 2006 8. http://www.outokumpu.com/applications/ documents/start.asp 9. J.E. Truman, Corrosion: Metal/Environment Interaction, Vol 1, Newness-Butterworths, 1976, p 352 10. J.P. Audouard, Stainless Steels, Les Editions de physique, 1993, p 268 11. H. Mimura et al., Nippon Steel Tech. Report 90, July 2004, p 94–99 12. http://www.alleghenyludlum.com/ludlum/ Documents/AL-6XN_sourcebook.pdf 13. http://www.alleghenyludlum.com/ludlum/ Documents/al610_611.pdf 14. F. Tagashi et al., Kawasaki Technical Report 31, 1994

15. R.C. Newman, Corrosion, Dec 2001, p 1030–1041 16. N.J. Laycock and R.C. Newman, Corros. Sci., Vol 39, 1997, p 1771 17. Y. Kobyashi, S.Virtanen, and H. Bohni, Proc. Electrochem. Soc., 1999, p 533–540 18. Z. Szlarska-Smialowska, Pitting Corrosion of Metals, NACE, Houston, TX, 1986 19. E.T. Turkdogan, Fundamentals of Steelmaking, Institute of Materials, 1996 20. H.S. Kim and H. Lee, Met. Trans. A, Vol 32A, June 2001, p 1519 21. M.P. Ryan, D.E. Williams, et al., Nature, Vol 415, Feb 2002, p 770–777 22. A.J. Grekula et al., Corrosion, 40, 1984, p 569 23. Stainless Steels Les Editions de physique, 1993 24. N. Suutala and M.Kurkela, Stainless Steel ‘84, Metals Institute, 1985, p 240–247 25. T.Suter, E.G. Webb, H. Bohni, and R.C. Alkire, J. Electrochem. Soc., Vol 148 (No. 5), 2001, B174–B185 26. M.O. Spiedel, Stainless Steels ‘87, Institute of Metals, London, 1988, p 247–252 27. Y. Cao, F. Ernst, and G.M. Michal, Acta Mater., Vol 51, 2003, p 4171. 28. G. Lothongkum et al., Corros. Sci., Vol 48, 2006, p 137–153 29. S. Fujimoto , Sci. Technol. Adv. Mater., Vol 5, 2004, p 195–200 30. J.D. Fritz, J.F. Grubb, B.W. Parks, and C.P. Stinner, Stainless Steel World, KCI, P01488, 2001 31. M.O. Spiedel, Met. Trans. A, Vol 12A, 1981, p 779 32. A.J. Sedricks, Corrosion of Stainless Steels, Wiley, 1979, p 158 33. R.N. Parkins, Br. Corros. J., Vol 14, 1979, p 5 34. S. Tahtinen, H. Hahhinen, and T. Hakkarainen, Stainless ‘84, Metals Institute, 1985, p 143–148 35. M. Nagumo et al., Met. Trans. A, Vol 32A, Feb 2001, p 332 36. H. Hanninen et al., Hydrogen Effects on Materials Behavior, TMS, 2003, p 201–210 37. V.J. Gadgil, Scr. Metal., Vol 28, 1993, p 1489–1494 38. M. Hoelzel et al., Mater. Sci. Eng. A, Vol 384, 2004, p 255–261 39. B.G. Pound, Hydrogen Effects on Materials Behavior, TMS, 2003, p 93–103 40. S.C. Dexter, Microbiologically Influenced Corrosion, Corrosion: Fundamentals, Testing, and Protection, Vol 13A, ASM Handbooks, 2003, p 398–413

Stainless Steels for Design Engineers Michael F. McGuire, p 57-68 DOI: 10.1361/ssde2008p057

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 5

Oxidation Summary STAINLESS STEEL, often considered mainly as a corrosion-resisting material, plays an important role as a heat-resisting material. This is partly due to its ability to retain strength at higher temperatures at which many otherwise useful alloying systems, such as aluminum, copper, and even titanium, soften. Stainless steel retains strength and has excellent oxidation resistance from room temperature to nearly 1000 °C, at which other economical alternatives are lacking.

Introduction High-temperature oxidation is a form of environmental degradation of metals and alloys that results from the following chemical reaction in which metal atoms M react with gaseous oxidants: M (s) + 0.5 yX 2 ( g) → MX y

(Eq 1)

Due to the high temperatures involved, these reactions are generally rapid and thus are a concern for high-temperature applications such as components for power generation. The electronegative gaseous oxidant X could be sulfur, chlorine, etc., but the discussion here mainly is limited to oxidation by oxygen or water vapor (in the latter case, hydrogen would be added as a product in Eq 1. For a thorough study of oxidation, referred to Ref 1 to 3).

Thermodynamics of Oxidation As discussed in Chapter 2, Corrosion Theory, a reaction will be possible when the net free energy

is negative. In Eq 1, the free energy G is decreased by a lower nobility of the metal (or a higher activity a of a metallic alloying element), a lower temperature T, and a higher partial pressure P of the oxidizing gas according to: ⎛ aMeX ⎞ y ΔG = ΔG 0 + RT ln ⎜ ⎟ ⎜⎝ aM PX0.5 y ⎟⎠ 2

(Eq 2)

In the case of alloy oxidation, for which temperatures are high enough to form mixed oxides or spinels, the activities of the oxide species also need to be considered. The standard Gibbs free energy ΔG0 is often presented in Richardson-Jeffes (Gibbs free energy-temperature) diagrams such as the one shown in Fig. 1 (Ref 4). It is evident from Fig. 1 that the major alloying element in stainless steels, chromium, forms a thermodynamically significantly more stable oxide, Cr2O3, than those of the base alloy iron (FeO, Fe3O4, and Fe2O3) or the major ternary element nickel (NiO), and to a great extent, the chromium content determines the oxidation behavior of stainless steels. The Effect of Chromium. The oxidation of multicomponent alloys is a complex process from both thermodynamic and kinetic points of view. A range of oxides may form with various degrees of thermodynamic stabilities and stoichiometries (including complex ones with different cations), and there might be degrees of solubilities of oxides in one another. Kinetics of their growth is complex because metal solute diffusion in the metal phases varies, as do metal and oxygen ion mobilities in the different oxide phases. Birks, Meier, and Pettit distinguished between two basic types of behavior: (a) a noble matrix

58 / Stainless Steels for Design Engineers

Fig. 1

Standard Gibbs free energy of formation of some metal oxides as a function of temperature. Source: Ref 4

metal with less-noble alloying elements and (b) both matrix element and alloying elements are nonnoble. The concept of nobility is decided by the thermodynamic conditions of Eq 2; that is, an element for which the free energy defined by Eq 2 is negative is nonnoble. We first discuss the more common case (b), in which oxidation takes place under significantly oxidizing conditions, such as air. In such a situation, it can be seen from Fig. 1, that the matrix iron is nonnoble and so are many of the solutes (chromium, molybdenum, aluminum, silicon, manganese, etc.). A high-temperature Fe-Cr-O phase diagram is shown in Fig. 2. It can be seen here that Fe2O3 and Cr2O3 are soluble in each other, and that the

spinels Fe1.5Cr1.5O4 (with a solid solubility with Fe3O4) and FeCr2O4 form. The progressive change in oxidation behavior as chromium is added to iron has been described in the literature (Ref 1). At lower chromium contents and above a minimum temperature, an iron-chromium alloy would behave as pure iron, where FeO would form next to the metal, then gradually Fe3O4 and Fe2O3 would form toward the gas as oxygen potential increases. Isolated pockets of spinel may form within the FeO layer. The oxidation of iron proceeds predominantly due to the rapid ionic diffusion of Fe2+ cations on the FeO layer, which leads to growth of this layer. If chromium

Chapter 5: Oxidation / 59

Fig. 3

Fig. 2

The iron-chromium-oxygen phase diagram at 1300 °C. Source: Ref 5

in the base alloy is increased, the spinel pockets increase, and the mobility of Fe2+ decreases. As chromium content is increased further, a mixedspinel scale is formed. Iron diffusion through the mixed spinel is significant, and thus the scale is not yet protective. As chromium content is increased further, an outer layer of Cr2O3 is formed, and the oxidation behavior becomes similar to that of chromium. A chromium limit of roughly 20% is needed to achieve a permanent Cr2O3 scale. This amount decreases if nickel is added. While the thermodynamic driving force is important, the chromium content of the alloy can override because the supply of chromium to the interface becomes dominant. At chromium contents less than about 16 wt%, the oxidation rate is influenced by the rate of supply of chromium from the alloys beneath the oxide. Above 16%, the supply of chromium is fast enough that chromium gradients are low enough that instead transport in the oxide layer controls the rate. The rate of oxidation then follows a so-called parabolic law (this is explained in the next section), by which the mass change per unit area due to oxidation (incorporation of oxygen) is given by:

( m / A)

2

= km t

(Eq 3)

Parabolic rate constants for the growth of several oxides. Source: Ref 6

Here, m is the added mass, A is the area exposed to the oxidizing atmosphere, t is the time exposed, and km is the parabolic rate constant. The subscript “m” is added here to denote that the reaction is measured as added mass (it can also be defined for oxide thickness X). As shown in Fig. 3, the parabolic rate constant for the oxides of chromium, silicon, and aluminum are low compared to others, and this is the reason that these elements are used as alloying elements to reduce oxidation rates for alloys in high-temperature applications. The solubility of Fe3O4 in the spinel will eventually result in continued iron oxide formation, as the iron-chromium system is not an optimal basis for high-temperature oxidation resistance, although it might be an option from an economical standpoint compared to other alloy systems such as superalloys. The high mobility of both iron and manganese in the spinel structure is also an important factor. The major stainless steels used for oxidation resistance fall into two categories: the ferritic stainless steels and the austenitic. Table 1 lists some of the more significant alloys commonly encountered in applications for which oxidation resistance is paramount. The value of ferritic alloys (such as 409, 439, and 446) is that they are relatively inexpensive, and that they have a thermal expansion coefficient that is closer to that of the oxide than do austenitic alloys (such as 302B, 309, and 310). This gives them an advantage in cyclic oxidation applications even though their strength at high temperatures does not rival that of austenitic alloys. The ferritic stainless steels are the most widely used alloys based on their low cost, which has

60 / Stainless Steels for Design Engineers

Table 1

Oxidation-resisting grades of stainless steel in common use

UNS

Name

Composition, % C

S40900

409

0.08

S43935

11Cr-Cb(a) 12SR(a) 439

0.01 0.02 0.07

S44600 S30215 S30415 S38150 S30900 S31000

18Cr-Cb(a) 18SR(a) 4742(a) 446 302B 153MA 253MA 309 310

0.02 0.015 0.08 0.2 0.15 0.04–0.06 0.05–0.10 0.2 0.25

N

Cr

Ni

Mn

10.5–11.75

0.5

1

0.015 0.015 0.04

11.35 12 17.0–19.0

0.2 ... 0.5

0.25 ... 1

... ... ... 0.25 ... 0.12–0.18 0.14–0.20 ... ...

18 17.3 18 23.0–27.0 17.0–19.0 18.0–19.0 20.0–22.0 22.0–24.0 24.0–26.0

... 0.25 ... 0.6 8.0–10.0 9.0–10.0 10.0–12.0 12.0–15.0 19.0–22.0

0.3 0.3 0.7 1.5 2 0.8 0.8 2 2

Si

Ti

Nb

1

6X(C + N) to 1.10 1.3 ... ... 0.3 1 0.20 + 4X(C + N) to 0.75 0.45 0.25 ... 0.25 ... ... 1 ... 2.0–3.0 ... 1.0–2.0 ... 1.4–2.0 ... 0.75 ... 1 ...

Other

...

...

0.35 0.6 ...

... 1.2 Al ...

0.55 ... ... ... ... ... ... ... ...

... 1.7 Al 1.0 Al ... ... 0.04 Ce 0.04 Ce ... ...

Note: All compositions include Fe as balance. Single values are maximum, unless otherwise specific (a) Indicates typical analysis

made them the standard alloys for automotive exhaust systems.

Transient Oxidation The oxidation of a clean metal surface on exposure to an oxidizing environment will initially lead to all the nonnoble components of the alloy being oxidized together, forming mixed oxides having composition similar to the base alloy. In stainless steels, these initial oxides are typically Fe-Cr-Ni-Mn mixed oxides. This is called transient oxidation. As these oxides thicken, the partial pressure of oxygen at the scale-metal interface falls until only the most reactive element present in high concentration can be oxidized. For stainless steels, this means that a layer of Cr2O3 is eventually established in contact with the alloy.

scale-gas surface. Similarly, the electric charge can be carried by either n-type (electrons) or ptype (electron holes) electronic defects. The case will be determined by the equilibrium defect structure of the oxide, which depends on temperature and oxygen partial pressure. In the case of Cr2O3, chromium cations are the predominantly mobile defects (Fig. 4b) as a result of a very small degree of deviation from stoichiometry in the cation lattice, that is, Cr2-5O3, leading to metal deficiency. The contribution of chromia grain boundary diffusion is large and probably dominates the process at temperatures of interest. The defect can be described as an interaction with oxygen, at high oxygen potentials, through Kroger-Vink notations as: 3 3 O ( g) = VCr''' + OOx + 3h • 2 2 2 −3

The Electrochemical Nature of Oxidation Once an inner scale of Cr2O3 is formed, as shown in Fig. 4(a), the oxidizing gas is reduced at the gas-scale interface, and the chromium is oxidized at the metal-scale interface. The Cr2O3 scale serves as both electrolyte, through which ions are transported, and electron lead, through which electronic defects are transported. In principle, either or both metal or oxygen ions can migrate. If oxygen ion mobility dominates, then the scale would continue to grow at the oxide-metal interface, whereas if chromium ion mobility dominates, the oxide will grow at the

3

⎡VCr''' ⎤ p3 = K 2 pO3/ 4 → ⎡VCr''' ⎤ = K12 K 2 Po4 ⎣ ⎦ ⎣ ⎦ 2 2

(Eq 4)

The electron holes that form as charge-compensating defects serve as the “electron lead” in the electrochemical cell in Fig. 4b. The free energy of Reaction 4 determines the concentration of mobile defects and thus the diffusion coefficient and electrochemical mobility (Be) of the cation according to: ⎛ ΔGm ⎞ DCr 3+ = ⎡⎣VCr 3+ ⎤⎦ γΛ 2 υ ⋅ exp ⎜ − ⎝ RT ⎟⎠ ⎛ ΔH m ⎞ = ⎡⎣VCr 3+ ⎤⎦ ⋅ cons tan t ⋅ exp ⎜ − ⎝ RT ⎟⎠

(Eq 5)

Chapter 5: Oxidation / 61

Fig. 4

e BCr 3+ =

Metal with oxide scale. (a) A protective scale that prevents gas access. (b) Schematic of electrochemical oxidation through a protective oxide scale that serves as electrolyte and electron lead. The case is for mobile cations

3FDCr 3+ RT

(Eq 6)

where zi is the ion charge, F is Faraday’s constant (96,457 C.eq–1). Compared to wustite (FeO), the equilibrium constant of Eq 4 is quite low, resulting in a low degree of nonstoichiometry in Cr2–δO3 compared to Fe1–δO (where δ can be as large as 0.05), and thus the transport of Cr3+ through its scale is much slower than the transport of Fe2+ through FeO and thus the difference in parabolic rate constants in Fig. 3.

Kinetics and Oxidation Rates: Wagner’s Theory The parabolic oxidation rate was introduced without explanation in Eq 3. It was first described in terms of oxide defect structure and resulting transport properties by Wagner (Ref 6), and the theory is explained in most of the monographs on oxidation, such as Chapter 3 in Ref 3 and Chapter 4 in Ref 1. This treatment follows the derivation in Ref 1. Consider a general case, as shown in Fig. 4 under the assumptions that (a) the scale is compact and adherent, (b) electrode reactions are rapid enough to be in equilibrium at the interface and surface, (c) nonstoichiometry is small and uniform throughout the scale (i.e., defects are in thermal equilibrium throughout the scale), and (d) double-layer

effects are ignored (i.e., scales are relatively thick compared to range of space charge effects). As a case study, let us assume that the mobile ion defect is cations due to metal vacancies in the scale. The molar flux J (moles/m.s) of a particle i in an electrolyte subjected to an electrochemical potential gradient was shown to be: Ji = − ci Bi

∂ ( μ i + zi Fφ ∂x

)

(Eq 7)

where zi is the ion charge, F is Faraday’s constant (96,457 C/gram equivalent), and φ is the electric field (V). The electronic or ionic conductivity κ in an electrolyte can be computed through: κ = F 2 ∑ zi Bi ci 2

(Eq 8)

The contribution of a given ion specie type or electron defect type to this conductivity is denoted as the partial conductivity and computed as: 2

κ i = F 2 zi Bi ci

(Eq 9)

Inserting Eq 9 into Eq 7 yields: Ji = −

∂ ( μ i + zi Fφ

κi 2

F zi

2

∂x

)

(Eq 10)

62 / Stainless Steels for Design Engineers

Now, if the mobile particles are a single type of metal cations (e.g., Cr3+) and electrons, then two fluxes are present: Jc = −

∂ ( μ c + zc Fφ

κc 2

F zc

∂x

2

)

(Eq 11)

and Je = −

∂ ( μ e + ze Fφ

κe 2

F ze

∂x

2

)

(E q 12)

(Eq 13)

And at the oxide-scale/metal interface, the anode reaction is in equilibrium, that is: z + M = M + zce − and therefore, c

μ M = μ c + zc μ e

(Eq 14)

Combining Eq 11 to 14, the potential gradient is eliminated, and the chemical potential gradients can be replaced by the metal (M) potential gradient, and the following equation results: Jc = −

κ cκ e

zc2 F 2 ( κ c + κ e

)

∂μ M ∂x

μ"

(Eq 16)

M

Now, if the growth of the oxide scale is controlled by the flux of cations: ∂C dx ∝ JC = − DC C ∂x dt

(Eq 18)

The constant k is the parabolic rate constant. A mass balance can be written where the flux of cations for a period of time dt is equated to the amount of metal being accumulated as cations inside the scale of thickness dx: (Eq 19)

By combining Eq 18 and 19 and inserting Eq 16 for the flux, an expression for the parabolic rate constant is obtained: 1 k= 2 2 zc F CC

μ 'M

κ cκ e dμ + κe M c

∫κ

μ ''M

(Eq 20)

If the mobility and thus partial conductivity of electrons is significantly higher than that of the ions (a reasonable assumption), then Eq 20 can be simplified as: 1 k= 2 2 zc F CC

μ 'M

∫ κ dμ c

(Eq 21)

M

μM

(Eq 15)

To obtain an explicit function for the cation flux, Eq 15 needs to be integrated after variable separation, keeping in mind that conductivities and metal chemical potential may vary within the scale. Integrating from the gas-scale surface (x = 0, μM = μ′′M) to the scale-metal interface (x = X, μM = μ′′M) one obtains. M κκ 1 Jc = − 2 2 ∫ c e dμ M κ zc F x μ ' c + κ e

dx k = dt X (t )

Jc ⋅ dt = CC dx

Electrical neutrality requires that: Jc Z c + Je Z e = 0

assumed, that is, a linear drop across X, at all times:

(Eq 17)

The concentration drop across the scale is constant since interface and surface reactions are at equilibrium. If quasi steady state is

Since from diffusion theory we know that Dc = BcRT, and inserting this in Eq 9, one obtains κ c = F 2 zc

2

Dc c RT c

and inserting this into Eq 21 results in: 1 k= RT

μ 'M

∫ D dμ c

M

(Eq 22)

μ ''M

It is often more convenient to express Eq 22 in terms of oxygen potentials rather than the metal potentials. It was assumed at the onset of this analysis that the deviation from stoichiometry is small and constant throughout the scale. Therefore, the oxide potential is constant and: zc μ + μ M = μ MO = constant zC / 2 4 O2

(Eq 23)

Chapter 5: Oxidation / 63

And thus, dμ O = dRT ln PO ∝ − dμ Me. Therefore, 2

2

Eq 22 can be written: ln PO''

k∝

∫

PO''

2

2

Dc d ln PO = 2

ln PO'

∫

PO'

2

2

Dc dP PO O2

(Eq 24)

2

In the case of Cr2O3, if bulk diffusion is dominating, Eq 4 and 5 inserted into the diffusion coefficient in Eq 24 for C = Cr3+, results in: PO''

2

k∝

∫

PO'

2

PO''

2 P 3/ 4 dPO = ∫ P −1/ 4 dPO 2 2 PO P' 2

O2

=

( ) − (P )

3 ⎡ '' P 4 ⎢⎣ O2

3/ 4

' O2

3/ 4

⎤ ⎥⎦

(Eq 25)

Thus, the parabolic rate constant would be predicted to vary with the power of three-quarters of the external oxygen partial pressure. Grain boundary diffusion has however been identified to be important in the case of Cr3+ transport (Ref 8). The observed growth rate of Cr2O3 polycrystalline films is far too fast to be accounted for by bulk diffusion of chromium ions; instead, grain boundary diffusion would be expected to dominate (Ref 9).

The Volatile Nature of Cr2O3 At high enough temperatures and high enough oxygen partial pressures, the formation of a gaseous hexavalent chromium oxide CrO3* could lead to thinning of the Cr2O3 scale according to the following reaction: 3 Cr2 O3 (s) + O 2 ( g) = 2CrO3 ( g) 2

(Eq 26)

Figure 5 shows the vapor pressure of the superoxide as a function of temperature and partial pressure of O2. The effect of this reaction on the oxidation kinetics can be described as follows: The thickness change described through the parabolic rate constant in Eq 18 is corrected for by the * Hexavalent chromium is now considered a human carcinogen and is rigorously regulated by both the Occupational Safety and Health Administration (OSHA) and the U.S. Environmental Protection Agency (EPA).

thickness loss due to evaporation, which is described by a first-order reaction kinetics expression with rate constant ke. Thus, the thickness change becomes: dx k = −k dt X (t ) e

(Eq 27)

This results in a so-called paralinear (as opposed to parabolic) rate for the oxide thickening (Fig. 6), and at a critical oxide scale thickness X( the rate of thinning due to evaporation equals the rate of thickening due to oxidation. In Eq 27, this means that dx/dt = 0; consequently, X = k/ke. While at first this seems to suggest that it does not affect the oxidation process in that the rate of oxidation does not increase, the formation and evaporation of chromium oxides results in greater chromium consumption in the alloy compared to what would be the case if evaporation did not occur. As a result of evaporation losses, stainless steels that depend on a protective chromium oxide layer are limited in use to temperatures up to 900 to 1000 °C. The presence of water vapor promotes the formation of even more volatile oxyhydroxides (e.g., CrO2(OH)2 ) (Ref 11, 12).

Spalling and Cracking of the Scale At elevated temperatures or during temperature cycling, there are multiple ways in which stresses can develop that may crack and blister the scale, rendering it nonprotective. The different causes of stress generation are described in Chapter 5 in Ref 1. So-called growth stresses arise due to changes caused by the oxidation process itself. These include differences in lattice mismatch, alloy depletion in the metal, point-defect gradients in scales containing oxides such as FeO, with large deviation from stoichiometry, recrystallization, and volume differences between the oxide and metal. The last is perhaps the most commonly mentioned and is characterized by the Pilling-Bedworth ratio, abbreviated as PBR (Ref 13). PBR =

VOxide VmOxide = Metal VMetal Vm * ν

(Eq 28)

Here, the subscript m stands for molar volume, and υ is the number of metal atoms needed to form a stoichiometric unit of the oxide (in the

64 / Stainless Steels for Design Engineers

Fig. 5

Chromium-oxygen system species volatility as a function of temperature and oxygen pressure. Source: Ref 10

case of Cr2O3, υ is 2, and in the case of FeO, it is 1). When PBR is greater than 1, then the oxide is expected to be in compression and is likely to be protective, whereas if it is less than 1, the oxide is in tension and thus nonprotective. There are, however, many exceptions to this, partly because the stress state often depends more on the mechanisms and conditions of the oxidation process rather than the properties of metal and oxides. Thermal, stresses are caused by differences in thermal expansion between the oxide and metal, and the stresses generated in oxide scales can be estimated: σ Oxide =

)

− EOxide ( α Oxide − α Metal ΔT ⎛ ⎞ t E 1 − ν p ⎜ 1 + 2 Oxide Oxide ⎟ t Metal E Metal ⎠ ⎝

(

)

(Eq 29)

The equation is written for a case shown schematically in Fig. 7, where both sides on a metal undergo oxidation. Here, σ is the stress, υp is Poisson’s ratio (it has been assumed that there is no mismatch), α is the coefficient of thermal expansion, E is the modulus of elastic-

Fig. 6

Schematic of paralinear oxidation as a result of evaporation of chromium superoxide

ity, t is thickness, and ΔT is the temperature change. In general, α is larger for the metal than the oxide; thus, during cooling the stresses are expected to be compressive and during heating tensile. Thermal stresses can cause spalling of the protective oxide layer, and it is most severe under cyclic conditions. Effect of Silicon, Aluminum, and Molybdenum. Due to concerns about the cost of chromium and its (former) classification as a

Chapter 5: Oxidation / 65

tox

tm

tox

Oxide

Metal

Oxide

Fig. 7

Schematic of a cross section of oxidized sample indicating dimensions in Eq 29 for predicting thermal

stresses

strategic material, there were efforts to try to substitute less-expensive elements such as aluminum and silicon that also are known to form protective layers (Ref 14), even though they have significant metallurgical and mechanical drawbacks. Silicon additions of 1.5 wt% or more have the effect of forming a continuous amorphous subsurface layer in iron-silicon and Fe-Cr-Si alloys that is relatively impervious to transport of ions. The mechanism for the evolution of such a layer is as follows: (1) The more readily available iron or chromium first forms a surface layer, and this causes an enrichment of silicon at the oxide-metal interface. (2) As sufficient silicon is enriched, the SiO2 layer is formed. It has been reported that alloys with chromium content as low as 6 wt% and silicon content of 1.5 wt% perform in terms of oxidation as well as commercial stainless steels. Also, an addition of 4 wt% Si to a Fe14wt%Cr14wt%Ni alloy resulted in a 200-fold reduction in weight gain at 900 °C However, this SiO2 layer seems to promote oxide spalling, especially in cyclic service. Aluminum forms a very stable thin outer layer of Al2O3 that initially reduces the oxidation rate. Alumina is among the most stable and defect-free oxides, giving it an extremely low diffusion rate. In consequence, if sufficient aluminum is present to maintain the protective alumina scale, the aluminum-bearing alloys provide the greatest oxidation resistance attainable in engineering alloys. If, however, the aluminum content is not sufficient to force alumina scale formation at the scale-metal interface, an

internal Cr2O3 layer forms and thickens. Eventually, the surface aluminum-oxide layer flakes off. Also, due to the low oxygen potential needed to form Al2O3, internal oxidation may result below the metal-scale interface in alloys in which formation of a continuous alumina scale film does not occur. Molybdenum is suggested to strain the lattice due to its larger size and consequently increase the rate of bulk diffusion of elements (Ref 11), which can enhance the rate of initial Cr2O3 formation. Molybdenum is usually considered detrimental for oxidation resistance. Molybdenum normally forms MoO2 oxide, but this can oxidize further to form the low-melting and volatile MoO3. If the MoO3 evaporates, there is little problem, but if its volatilization is inhibited by low atmosphere circulation, liquid MoO3 can accumulate and dissolve the protective Cr2O3 scale, leading to catastrophic oxidation. Effect of Rare Earth Additions. Cerium, lanthanum, and yttrium additions are known to improve oxidation resistance of high temperature nickel- and iron-based alloys (Ref 6). Rare earth additions have been suggested to have a multitude of beneficial effects, such as reducing the growth kinetics of Cr2O3 scales, stabilizing Cr2O3 scales at lower chromium levels, increasing adhesion, and preventing spalling of the oxide scale during thermal cycling. The explanation for any of this does not seem clear, but some hypotheses have been suggested. The effect on the growth kinetics could be because the reactive element ions collect at grain boundaries and block fast path diffusion. The improved adherence could be because these elements getter tramp elements such as sulfur and suppress void formation at the interface. Furthermore, they might form so-called oxide pegs at the interface (Ref 16). The precise role of the rare earth additions to Cr2O3 oxide protection and the mechanism by which they are incorporated into the scale during the surface treatment processes remain unknown. An understanding of these fundamental issues would help to develop optimum alloy chemistries for selected high-temperature and pressure applications and to further develop the surface infusion process. The lack of fundamental understanding of how rare earths improve oxidation resistance has not stopped the development of several alloys that benefit from the effect.

66 / Stainless Steels for Design Engineers

Oxidation Under Less-Oxidizing Atmospheres The two types of alloy oxidation behaviors, (a) a noble matrix metal with less-noble alloying elements and (b) both matrix element and alloying elements are nonnoble, were mentioned. When designing against oxidizing environments, case b is perhaps the most relevant, and most of the discussion has been devoted to this. However, during annealing for microstructural control, steels are exposed to furnace gases at high temperatures that have relatively low oxygen or steam contents, for which case a will apply, that is the atmosphere does not cause iron (or nickel) to oxidize but chromium (and aluminum, silicon, molybdenum, etc.) does. For simplicity, assume a binary system A-B of “noble” iron and “reactive” chromium. In this case, depending on the concentration of the reactive element and atmosphere, the oxide of the reactive element can, in principle, form either on the surface (as has been discussed so far) or internally as discrete oxide particles in a metal matrix through oxygen diffusion into the metal. Both cases are shown schematically in Fig. 8. Let us discuss the conditions that promote one or the other of these by starting with a situation in which (a) no surface oxide exists and (b) the oxygen atmosphere is such that the solubility of oxygen within a distance X is enough to thermodynamically render Cr2O3 stable according to Eq 2 but none of the iron oxides. The derivation is done in terms of both a generic system A-B causing an oxide BOν and for iron-chromium causing Cr2O3. Assume for a start that the DO >> DCr, and thus while oxygen diffuses into the alloy, chromium does not counterdiffuse. The flux of oxygen inward into the metal is then the cause of increased mass. Assuming a

Fig. 8

Schematic of two cases in a less-oxidizing atmosphere. (a) Adsorption of oxygen leading to internal oxidation and (b) external oxidation as the B element migrates.

quasi-steady-state situation as shown in Fig. 9, the flux can be written: JO =

N X − N OS NS dm = − DO O = DO O dt Vm X Vm X

(Eq 30)

Within the depth X, the oxygen solubility is such that Eq 2 is negative enough that Cr2O3 forms. Beyond X, it is not. At the distance X, the oxygen concentration is negligibly low compared to the surface composition, which is in equilibrium with the gas phase. The molar volume Vm is used to obtain the flux in units of moles per square meter. Within the layer 0 < x < X, all the chromium is assumed to be oxidized; therefore, the accumulated mass due to oxygen addition: o 3 N B0 υX N Cr 2 X m= = Vm Vm

(Eq 31)

Differentiating Eq 31 with time, equating to Eq 3, and separating variables results in: XdX =

N OS DO υN BO

dt =

N OS DO 3 0 N 2 Cr

(Eq 32)

Integrating Eq 32 from x = 0 to x = X results in an expression for the internal oxidation depth X: ⎛ 2N S D X =⎜ O0O ⎝ υN B

Fig. 9

⎞ t⎟ ⎠

1/ 2

⎛ ⎜ 2N S D =⎜ O O 0 ⎜ 3 N Cr ⎝ 2

⎞ ⎟ t⎟ ⎟ ⎠

1/ 2

(Eq 33)

Quasi-steady-state approximation of the moving boundary problem of internal oxidation. Counterdiffusion of B is assumed to be negligible

Chapter 5: Oxidation / 67

This expression predicts a parabolic dependence of X with time, just as Eq 3 did for the external oxidation. For a more rigorous derivation (without assuming quasi steady state), refer to Appendix B in Ref 1. Now, let us see what causes this to transition into an external scale. It was assumed in the derivation of Eq 33 that counterdiffusion of chromium does not occur. When considering Eq 33, is clear that the rate of penetration of the internal oxidation front will decrease with (a) increasing NB0, (b) decreasing NOS, and (c) decreasing DO. If DCr was not negligible, there would be a gradual change in oxidation morphology if the ratio (NB0DB)/(NOSDO). Gradually, if the ratio were increased there would be a slowing of the inward penetration of the internal oxide front and an enrichment of BOυ in the internally oxidized zone. Wagner developed a model (Ref 16), based on that at some point, when the volume fraction of BOυ versus volume metal in the internally oxidized zone reaches a critical value g*, there is a transition from internal to external oxidation; specifically, this happens when: 1/ 2

⎡ πg* S DOVM ⎤ N >⎢ NO ⎥ DBVOx ⎦ ⎣ 2ν 0 B

Fig. 10

Temperature dependence of metal dusting of iron. Source: Ref 18

trogen absorption, but if the oxygen is depleted before all surfaces are oxidized, the remaining material can be rapidly nitrided by the residual, essentially pure, nitrogen atmosphere.

REFERENCES (Eq 34)

If more than one reactive elements were present (such as is the case in stainless steels in which aluminum, silicon, molybdenum, niobium, etc., may be present), this will decrease the inward flux of oxygen, and thus the transition to external may occur at a lower solute (CCr0) concentration than what is predicted by Eq 34. Metal Dusting. Under reducing conditions, in products of combustion atmospheres, oxidation and carburization may occur simultaneously and at a higher rate than exhibited in pure oxidation. Under even more reducing conditions, the condition called metal dusting may occur. Metal dusting is often characterized by the generation of large, smooth pits that look as if metal had been scooped from the surface. The underlying phenomenon is the formation of metal carbides, which manifests itself as the breakup of bulk metal to metal powder. This occurs at temperatures at which the carbide is most stable (Fig. 10). During oxidation in air, if large surface area is present under conditions of restricted air supply, oxygen can be depleted to the point that oxidation essentially ceases. Oxide films inhibit ni-

1. N. Birks, G.H. Meier, and F.S. Pettit, Introduction to the High-Temperature Oxidation of Metals, 2nd ed., 2006, Cambridge University Press, New York 2. P. Kofstad, High Temperature Corrosion, 1988, Elsevier Applied Science, London 3. K. Hauffe, Oxidation of Metals, Plenum Press, New York, 1965 4. F.D. Richardson and J.H.E. Jeffes, J. Iron Steel Inst., Vol 160, 1948, p 261 5. C. Wagner and K. Grünewald, Z. Phys. Chem., Vol 40B, 1938, p 455 6. J.H. Park, W.E. King, N.L. Peterson, and S.J. Rothman, The Effect of Reactive Element on Self-Diffusion in Cr2O3, Norman L. Peterson Memorial Symposium, Oxidation of Metals and Associated Mass Transport, edited by M.A. Dayananda, S.I. Rothman, and W.E. King, AIME, Warrendale, PA, 1998, p 103–107 7. C. Wagner and K. Grünewald, Z. Phys. Chem., Vol 40B, 1938, p 455 8. D. Caplan and G.I. Sproule, Effect of Oxide Grain Structure on the High Temperature Oxidation of Cr, Oxid. Met., Vol 9, 1975, p 459–472 9. B.B. Ebbinghaus, Combust. Flame, Vol 93, 1993, p 119–137

68 / Stainless Steels for Design Engineers

10. K. Hilpert et al., JECS, Vol 143/11, 1996, p 3642–3647 11. N.B. Pilling and R.E. Bedworth, J. Inst. Met., Vol 29, 1923, p 529 12. J.K. Tien and J.M. Davidson, Oxide Spallation Mechanisms, Stress Effects and the Oxidation of Metals, ed. J.V. Cathcart, AIME, New York, 1975, p 200 13. J. Rawers, Understanding the Oxidation Protection of Fe-Cr-Si Alloys, Norman L. Peterson Memorial Symposium, Oxidation

of Metals and Associated Mass Transport, edited by M.A. Dayananda, S.I. Rothman, and W.E. King, AIME, Warrendale, PA, 1998, p 323–340 14. E.J. Felten, J. Electrochem. Soc., Vol 108, 1961, p 490 15. C. Wagner, J. Electrochem. Soc., Vol 103, 1956, p 571 16. C.M. Chun, J.D. Mumford, and T.A. Ramanarayanan, J. Electrochem. Soc., Vol 147, 2000, p 3680

Stainless Steels for Design Engineers Michael F. McGuire, p 69-90 DOI: 10.1361/ssde2008p069

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 6

Austenitic Stainless Steels Summary AUSTENITIC STAINLESS STEELS are the most common and familiar types of stainless steel. They are most easily recognized as nonmagnetic. They are extremely formable and weldable, and they can be successfully used from cryogenic temperatures to the red-hot temperatures of furnaces and jet engines. They contain between about 16 and 25% chromium, and they can also contain nitrogen in solution, both of which contribute to their high corrosion resistance. Were it not for the cost of the nickel that helps stabilize their austenitic structure, these alloys would be used even more widely.

because their greater thermal expansion coefficient tends to cause the protective oxide coating to spall. 2. They can experience stress corrosion cracking (SCC) if used in an environment to which they have insufficient corrosion resistance. 3. The fatigue endurance limit is only about 30% of the tensile strength (vs. ~50 to 60% for ferritic stainless steels). This, combined with their high thermal expansion coefficients, makes them especially susceptible to thermal fatigue. However, the risks of these limitations can be avoidable by taking proper precautions.

Introduction

Alloy Families in Perspective

Austenitic stainless steels have many advantages from a metallurgical point of view. They can be made soft enough (i.e., with a yield strength about 200 MPa) to be easily formed by the same tools that work with carbon steel, but they can also be made incredibly strong by cold work, up to yield strengths of over 2000 MPa (290 ksi). Their austenitic (fcc, face-centered cubic) structure is very tough and ductile down to absolute zero. They also do not lose their strength at elevated temperatures as rapidly as ferritic (bcc, body-centered cubic) iron base alloys. The least corrosion-resistant versions can withstand the normal corrosive attack of the everyday environment that people experience, while the most corrosion-resistant grades can even withstand boiling seawater. If these alloys were to have any relative weaknesses, they would be:

The fundamental criterion in the selection of a stainless steel is generally that it can survive with virtually no corrosion in the environment in which it is to be used. Good engineering practice sometimes requires that materials be selected for sufficient, but finite, service life. This is especially true for high-temperature service, for which creep and oxidation lead to limited life for all materials. The choice among the stainless steels that can be used in that environment is then based on the alloy from which the component can be produced at the lowest cost, including maintenance, over the intended service life. The ferritic stainless steels are less expensive for the same corrosion resistance but sometimes are found lacking because of: • Lack of toughness, as is the case at subambient temperatures or in thicknesses greater than about 1.5 mm • Lack of great ductility, specifically if more than about 30% elongation is needed

1. Austenitic stainless steels are less resistant to cyclic oxidation than are ferritic grades

70 / Stainless Steels for Design Engineers

• Susceptibility to high-temperature embrittling phases when moderately alloyed The less-expensive martensitic grades are used instead of austenitic when high strength and hardness are better achieved by heat treating rather than by cold work, and mechanical properties are more important than corrosion resistance. This is also the case for the more expensive PH grades, which can achieve corrosion resistance only matching the least corrosion resistant of the austenitic alloys. Duplex grades match austenitic grades in corrosion resistance and have higher strength in the annealed condition but present the designer with challenges with regard to embrittling phases that can form with prolonged exposure to elevated temperatures and only moderate ductility like the ferritic alloys. So, the austenitic grades are the most commonly used grades of stainless mainly because, in many instances, they provide very predictable levels of corrosion resistance with excellent mechanical properties. Using them wisely can save the design engineer significant costs in his or her product. They are a user-friendly metal alloy with life-cycle cost of fully manufactured products lower than many other materials.

Fig. 1

The austenitic alloys can have compositions anywhere in the portion of the Delong diagram labeled austenite shown in Fig. 1 (Ref 1). This diagram was designed to show which phases are present in alloys in the as-solidified condition, such as found in welds. Thus it also applies to castings and continuously cast products. As a practical matter of castability, the composition of most commercial alloys falls along the zone of several percent ferrite as cast. The salient feature of austenitic alloys is that as chromium and molybdenum are increased to increase specific properties, usually corrosion resistance, nickel or other austenite stabilizers must be added if the austenitic structure is to be preserved. The traditional way of displaying the austenitic stainless steels is to present 302 as a base. Figure 2 shows one such diagram. Diagrams such as these treat alloys as an evolutionary family tree and subtly mislead. Many alloys were pushed toward obsolescence because of advances in processing. For instance, 321 was developed as an alloy in which the detrimental effects of carbon were negated by addition of titanium. The widespread adoption of the argon oxygen decarburization (AOD) in the 1970s made this alloy unnecessary, except for special circumstances, since carbon could be cheaply

Schaeffler-Delong stainless steels constitution diagram. Adapted from Ref 1, 2

Chapter 6: Austenitic Stainless Steels / 71

Fig. 2

The austenitic stainless family. Source: Ref 3

removed routinely. Likewise, 302 gave way to the lower-carbon 304, for which the even lowercarbon 304L is commonly substituted and dually certified to qualify as either grade. While low carbon prevents sensitization, stabilized grades may still be preferred for special applications such as type 321 in aerospace and type 347 in refinery service. Similar inertia keeps the higher-nickel 300 series as the de facto standard when the more cost-efficient high-manganese 200 series is the logical basic grade. The relevant types of austenitic alloys can nonetheless be rationalized with this diagram. As chromium is added, oxidation resistance and corrosion resistance increase. Because

nickel equivalents (manganese, nitrogen, carbon, etc.) must also be added in matching amounts, austenite stability is also increased. If molybdenum, a chromium equivalent, is added, corrosion resistance but not oxidation resistance is enhanced. And, if nitrogen is the austenite stabilizer added to balance increases chromium or molybdenum, then corrosion resistance is also increased. With small exceptions, that is the rationale of austenitic grade design. Silicon is used as an alloy to promote oxidation resistance and resistance to corrosion by oxidizing acids. Copper is used to promote resistance to sulfuric acid. Rare earths make a more stably oxidation-resisting scale. Niobium increases

72 / Stainless Steels for Design Engineers

Lean Alloys

creep resistance. Sulfur and selenium increase machinability. In this chapter, austenitic alloys are classified into three groups: • Lean alloys, such as 201 and 301, are generally used when high strength or high formability is the main objective since the lower, yet tailorable, austenite stability of these alloys gives a great range of work-hardening rates and great ductility. Richer alloys, such as 305, with minimal work hardening are the high-alloy, lowest work-hardening rate grade for this purpose. The general-purpose alloy 304 is within this group. • Chromium nickel alloys when the objective is high temperature oxidation resistance. This can be enhanced by silicon and rare earths. If the application requires high-temperature strength, carbon, nitrogen, niobium, and molybdenum can be added. 302B, 309, 310, 347, and various proprietary alloys are found in this group. • Chromium, molybdenum, nickel, and nitrogen alloys when corrosion resistance is the main objective. Alloys such as silicon and copper are added for resistance to specific environments. This group includes 316L, 317L, 904L, and many proprietary grades.

Lean austenitic alloys constitute the largest portion of all stainless steel produced. These are principally 201, 301, and 304. Alloys with less than 20% chromium and 14% nickel fall into this unofficial category. Since they are stainless, it is generally taken for granted that these alloys will not corrode, and these alloys have sufficient corrosion resistance to be used in any indoor or outdoor environment, excluding coastal. These grades are easily weldable and formable and can be given many attractive and useful surface finishes, so they are very much generalpurpose alloys. Table 1 lists some typical compositions of the most commonly used lean austenitic alloys. These typical compositions vary with end use, raw material cost factors, and the preference of a given manufacturer. The compositions of standard alloys are often finetuned to the intended end use. In this table, the word drawing indicates higher nickel for lower work hardening, while tubing indicates alloys with higher sulfur to facilitate gas tungsten arc welding (GTAW) penetration. Tensile indicates lower alloy levels to increase the work-hardening rate for material that is intended to be used in the cold-worked, high-strength condition. 316L is included in its most common tubing end use chemistry even though it is a corrosion-resisting alloy because it is so pervasively used as a service center sheet item. The main difference among the lean austenitic alloys lies in their work-hardening rate: the leaner the alloy, the lower the austenite stability. As unstable alloys are deformed, they transform from austenite to the much harder martensite. This increases the work-hardening rate and enhances ductility since it delays the onset of necking since greater localized

Wrought alloys generally have cast counterparts that differ primarily in silicon content. Versions that require enhanced machinability have a high content of controlled inclusions, sulfides, or oxysulfides, which improve machinability at the expense of corrosion resistance. Carbon is kept below 0.03% and designated an L grade when prolonged heating due to multipass welding of heavy section (greater than about 2 mm) or when welds requiring a postweld stress relief are anticipated.

Table 1 Typical compositions of the most commonly used lean austenitic alloys Alloy

Designation

C

N

Cr

Ni

Mo

Mn

Si

Other

Other

Other

201 201 drawing 201LN 301 tensile 301 drawing 303 304 304 drawing 304 extra drawing 304L tubing 305 321 316L

S20100 S220100 S20153 S30100 S30100 S30300 S30400 S30400 S30400 S30403 S30500 S32100 S31603

0.08 0.08 0.02 0.08 0.08 ... 0.05 0.05 0.06 0.02 0.05 0.05 0.02

0.07 0.07 0.13 0.4 0.04 ... 0.05 0.04 0.04 0.09 0.02 0.01 0.0

16.3 16.9 16.3 16.6 17.4 ... 18.3 18.4 18.3 18.3 18.8 17.7 16.4

4.5 5.4 4.5 6.8 7.4 ... 8.1 8.6 9.1 8.1 12.1 9.1 10.5

0.2 0.02 0.2 0.2 0.02 ... 0.3 0.3 0.3 0.3 0.2 0.03 2.1

7.1 7.1 7.1 1.0 1.7 ... 1.8 1.8 1.8 1.8 0.8 1.0 1.8

0.45 0.5 0.45 0.45 0.45 ... 0.45 0.45 0.45 0.45 0.60 0.45 0.50

0.001 S 0.001 S 0.001 S 0.001 S 0.007 S ... 0.001 S 0.001 S 0.001 S 0.013 S 0.001 S 0.001 S 0.010 S

0.03 P 0.30 P 0.03 P 0.03 P 0.03 P ... 0.03 P 0.03 P 0.030 P 0.030 P 0.02 P 0.03 P 0.03 P

0.2 Cu 0.6 Cu 0.5 Cu 0.3 Cu 0.6 Cu ... 0.3 Cu 0.3 Cu 0.4 Cu 0.4 Ci 0.2 Cu 0.4 Ti 0.4 Cu

Chapter 6: Austenitic Stainless Steels / 73

deformation is more than offset by greater localized strain hardening. These grades are best viewed as a continuum with a lower boundary at 16%Cr-6%Ni and an upper boundary at 19%Cr-12%Ni. This represents the range from minimum to maximum austenite stability. Since that is the main distinction within this grade family, let us examine its basis. Martensite and Austenite. Stability. The formation of martensite at room temperature may be thermodynamically possible, but the driving force for its formation may be insufficient for it to form spontaneously. However, since martensite forms from unstable austenite by a diffusionless shear mechanism, it can occur if that shear is provided mechanically by external forces. This happens during deformation, and the degree to which it occurs varies with composition according to (Ref 4):

such as occurs when they are sensitized or when solute segregation occurs, as from welding, then the equation applies on a microscopic scale. Sensitized zones (i.e., the regions near grain boundaries where chromium carbides have precipitated) will have a much higher tendency to transform to martensite. Figures 3(a) and (b) show the changes in phase structure as a function of composition over ranges that encompass these alloys. Martensite can be present in two different forms. The α′-form is the bcc magnetic form, while ε is a nonmagnetic, hcp (hexagonal closepacked) version. The formation of ε versus α′ is related to the stacking fault energy of the alloy, which is given by (Ref 6):

Md30 (°C) = 551 – 462(%C + %N) – 9.2(%Si) – 8.1(%Mn) – 13.7(%Cr) – 29(%Ni + Cu) – 18.5(%Mo) – 68(%Nb) – 1.42 (GS – 8) (Eq 1)

Y300SF (mJ m-2) = Y0SF + 1.59Ni – 1.34Mn + 0.06Mn2 – 1.75Cr + 0.01Cr2 + 15.21Mo – 5.59Si –60.69(C + 1.2N)1/2 + 26.27(C + 1.2N) (Cr + Mn + Mo)1/2 + 0.61[Ni•(Cr + Mn)]1/2 (Eq 2)

This is the temperature at which 50% of the austenite transforms to martensite with 30% true strain (Ref 5). It should be noted that even elements that are chromium equivalents in promoting ferrite are austenite stabilizers in that they impede martensite formation. This temperature is the common index of austenite stability. This regression analysis was generated for homogeneous alloys. If alloys are inhomogeneous,

Epsilon martensite formation is favored in alloys of lower stacking fault energy. The fcc structures deform by slip between (111) planes. Viewed from these planes, the structure is a series of ABCABC atom arrangements. Slip between planes can result in an ABCA/CAB structure. This so-called stacking fault generates an hcp structure. With lower stacking fault energies, these are more readily

Fig. 3

(a) Iron-chromium phase diagram at 8% nickel; (b) iron-nickel phase diagram at 18% chromium

74 / Stainless Steels for Design Engineers

Fig. 4

Variation of martensite formation with temperature and true strain for 304. Source: Ref 7

formed, and ε predominates. The stacking fault can also be viewed as two partial dislocations with the material between them faulted. These partial dislocations, when generated in abundance, cannot readily slip past one another and thus pile up, increasing work-hardening rates. As in carbon and alloy steels, the martensite transformation can take place simply by cooling, but in the lean austenitic alloys the temperatures are well below ambient. The more stable alloys do not transform even with cryogenic treatment. Figure 4 shows the variation of martensite formation with temperature and true strain for 304. Martensite formed in these alloys is quite stable and does not revert until heated well above the temperatures (Fig. 5) at which it was formed. The carbon levels of austenitic stainless steels are always relatively low, so strain-induced martensite is self-tempering and not brittle. Martensite has been found to form in unstable austenite due to the electrochemically induced supersaturation by hydrogen (Ref 9). Under conditions of cathodic charging, superficial layers were found to transform to ε under conditions of intense hydrostatic compression. During subsequent outgassing, α′ was found to form due to reversals in the stress state. Martensite thus formed is, of course, susceptible to hydrogen embrittlement. Mechanical Properties. The tensile properties in the annealed state not surprisingly relate well to composition. The 0.2% yield strength

and tensile strength, respectively, are reported (Ref 10) to follow the equations: YS( MPa) = 15.4[4.4 + 23(%C) + 32(%N) + 0.24(%Cr ) + 0.94(%Mo) + 1.3(%Si ) + 1.2(%V) + 0.29(%W ) + 2.6(%Nb) + 1.7(%Ti ) + 0.82(%Al ) + 0.16(%Ferrite ) + 0.46(d −1/1/ 2 )

(Eq 3)

TS ( MPa) = 15.4[29 + 35(%C) + 55(%N) + 2.4(%Si ) + 0.11((%Ni ) + 1.2(%Mo) + 5.0(%Nb) + 3.0(%Ti ) + 1.2(%Al ) + 0.14(%Ferrite ) + 0.82(d −1/ 2 )

(Eq 4)

In each case, d is the grain diameter in millimeters. Another researcher (Ref 11) gave the relationships as: YS ( MPa ) = 120 + 210 N + 0.02 + 2 Mn + 2Cr + 14 Mo + 10 Cu + (6.15 − 0.054δ)δ + (7 + 35(N + 0.2))d −1/ 2

(Eq 5)

TS = 470 + 600( N + 0.2) + 14 Mo + 1.5δ + 8d −1/ 2

(Eq 6)

Again, d is grain diameter in millimeters, and δ is percent ferrite. The claimed accuracy for the latter set of equations is 20 MPa and is said

Chapter 6: Austenitic Stainless Steels / 75

Fig. 6

Fig. 5

Reversion of martensite formed by cold work. Source: Ref 8

to apply to both austenitics and duplex stainless steels, but clearly the tensile strength relationship must break down for leaner alloys, such as 301, in which tensile strength increases with decreasing alloy content because of the effect of increasing alloying causing less transformation to martensite, which inarguably produces higher tensile strengths in austenitic stainless steels. Equation 3 must also be favored over Eq 5 in that it accounts for carbon explicitly. One other hardening mechanism is possible in austenitic stainless steels, and that is precipitation hardening. Most precipitation-hardening stainless steels are unstable austenite, which is transformed to martensite before the precipitation hardening takes place. One commercial alloy, A-286, is entirely austenitic and employs the precipitation within the austenite matrix of Ni3 (titanium, aluminum) for strengthening. This is dealt with in a separate section. Austenitic stainless steels do not have a clear yield point but can begin to deform at as little as 40% of the yield strength. As a rule of thumb, behavior at less than half the yield strength is considered fully elastic and stresses below twothirds of the yield strength produce negligible plastic deformation. This quasi-elastic behavior is a consequence of the many active slip systems in the fcc structure. Even highly coldworked material exhibits this phenomenon, although stress-relieving cold-worked material will cause dislocations to “lock in place” and

Variation of impact strength with temperature for (a) austenitic, (b) duplex, and (c) ferritic stainless steels

form more stable dislocation arrays that break loose at a higher and distinct yield point. The tensile properties of austenitic stainless steels with unstable austenite, that is, those with Md30 temperatures (Eq 1) near room temperature, are very strain rate dependent. This is simply due to the influence of adiabatic heating during testing increasing the stability of the austenite. Tests run under constant temperature conditions, either by slow strain rates or use of heat sinks, produce lower tensile strengths. Thus, reported tensile strengths should not be taken as an absolute value but a result that can be significantly changed by changes in testing procedure, even with accepted norms and standards. Highly cold-worked austenitic stainless steels are often used for their robust mechanical properties. Few metallic materials can match the very high strengths they can achieve. Very lean 301 can be cold worked to yield strengths on the order of 2000 MPa because of its unstable austenite transforming to martensite. When cold worked to lower degrees, it can provide very high strength while keeping impressive ductility. Austenitic stainless steels have exceptional toughness. The ambient temperature impact strength of austenitic stainless steels is quite high. This is not surprising in view of their high tensile strengths and high elongations. What is most remarkable is the absence of a transition temperature, which characterizes ferritic and martensitic materials. Figure 6 shows impact strength of the various stainless steel types versus temperature. This again is due to the multiplicity of slip systems in the fcc structure and the fact that they do not require thermal activation. This makes the austenitic stainless steels, especially the 200 series, the optimal cryogenic

76 / Stainless Steels for Design Engineers

material, surpassing the 9% nickel martensitic steels in cost, toughness, and, of course, corrosion resistance. Precipitation of Carbides and Nitrides. Carbon is normally considered as an undesirable impurity in austenitic stainless steel. While it stabilizes the austenite structure, it has a great thermodynamic affinity for chromium. Because of this affinity, chromium carbides, M23C6, form whenever carbon reaches levels of supersatura-

tion in austenite, and diffusion rates are sufficient for carbon and chromium to segregate into precipitates. The solubility of carbon in austenite is over 0.4% at solidification but decreases greatly with decreasing temperature. The solubility is given by (Ref 12): log (C ppm ) = 7771 −

6272 T (°K )

(Eq 7)

The equilibrium diagram for carbon in a basic 18%Cr10%Ni alloy is shown in Fig. 7. At room temperature, very little carbon is soluble in austenite; even the 0.03% of L grades is mostly in a supersaturated solution. The absence of carbides in austenitic stainless is due to the slow diffusion of carbon and the even slower diffusion of chromium in austenite. At a carbon level of 0.06%, which is found in most 304, supersaturation is reached below about 850 °C. Below this temperature, supersaturation increases exponentially, while diffusion decreases exponentially. This results in precipitation rates that vary with temperature and carbon level as shown in Fig. 8. At these temperatures, grain boundary diffusion is much more rapid than bulk diffusion, and grain boundaries provide excellent nucleation sites, so precipitation occurs along grain boundaries. And, because carbon diffuses several orders of magnitude more rapidly than chromium, carbon diffuses to and combines with chromium essentially in situ, depleting the grain boundaries of chromium in solution.

Fig. 7

Carbon solubility in 18–10 austenitic stainless. Source: Ref 13

Fig. 8

The precipitation rates for Cr23C6 as a function of carbon content

Chapter 6: Austenitic Stainless Steels / 77

Fig. 10

Fig. 9

Depletion of chromium from the austenite near grain boundaries due to chromium carbide precipitation. Source: Ref 14

Figure 9 shows that the local chromium depletion is such that the chromium level can become low enough that it has not even enough to be stainless and certainly much lower corrosion resistance than the surrounding area. This zone, because it is lower in chromium, also has very unstable austenite and is quite prone to martensite formation. Figure 10 shows how the locus of precipitation changes with time and temperature. Carbon relatively far from grain boundaries in the interior of grains remains in supersaturation until much longer times and much greater supersaturation since bulk diffusion is required for the nucleation and growth of these precipitates. The key observation is that any solid-state precipitation of a chromium-rich precipitate will necessarily cause local chromium depletion and a resulting loss of corrosion resistance.

Variation of carbide precipitation locus with time. Source: Ref 16

Much longer term heat treatment is required to eliminate these depleted zones by rehomogenization of slowly diffusing chromium than the short time required to form them. This is very evident for carbides, but also true for oxides. Underneath chromium-rich oxide scales is a layer depleted in chromium and lower in corrosion resistance. This is why not only scale from welding must be removed, but also the underlying chromium-depleted zone. Other precipitation processes that give rise to chromium depletion are α and χ and the solidstate precipitation of oxides, nitrides, and sulfides. Chromium precipitates that form in the liquid alloy do not cause depletion of chromium locally because no chromium gradients are set up around them during precipitation as diffusion in the liquid is very rapid. Thus, primary carbides, oxides, and sulfides are not per se harmful to corrosion resistance. But, if the same compounds form and grow in the solid state, chromium depletion occurs (Ref 15). Alloying elements can have a major influence on carbide precipitation by their influence on the solubility of carbon in austenite. Molybdenum and nickel accelerate the precipitation by diminishing the solubility of carbon. Chromium and nitrogen increase the solubility of carbon and thus retard and diminish precipitation. Nitrogen is especially useful in this regard (Fig. 11). Increasing austenite grain size accelerates precipitation, as does cold work, especially in the interior of grains, where diffusion is enhanced by increased defect density. Nitrogen is much more soluble than carbon and does not give rise to sensitization phenomena as does carbon even though Cr2N can be a

78 / Stainless Steels for Design Engineers

Fig. 11

Delay in carbide precipitation induced by nitrogen level. Source: Ref 17

stable phase when the solubility limit is exceeded. The solubility is over 0.15% in austenite, so its precipitation seldom has the possibility of occurring, but it does become an issue in ferritic stainless steels in this regard, for which solubility is much lower. Manganese and chromium increase the solubility of nitrogen in austenite. Stabilization. Before carbon was easily lowered to harmless levels, it was found that adding more powerful carbide formers than chromium could preclude the precipitation of chromium carbides. Titanium and niobium are the most useful elements in this regard. They form carbides with solubility that follows the following equation type: log [ M] [ X] = + A − H

RT

(Eq 8)

For titanium carbide and niobium carbide, the respective solubilities are: log [Ti ] [C] = 2.97 −

6780 T

log [ Nb] [C] = 4.55 −

9350 T

(Eq 9)

(Eq 10)

Oxides and sulfides are more energetically favorable than are carbides and nitrides of these metals. Thus, any additions made to form carbides must be sufficient to account for the prior formation of these compounds. Nitrogen also competes with carbon for available titanium or niobium. Thus, for successful gettering of all carbon, there must be sufficient titanium or niobium to combine stoichiometrically with all these species present. This requires in rough terms that titanium exceed four times the carbon plus nitrogen, or that niobium exceed eight times, after accounting for the oxygen and sulfur. It would be a mistake

Fig. 12

Variation of hardness with depth and therefore carbon content in colossal supersaturation

to ignore the titanium-consuming capacity of oxygen and sulfur unless they have been minimized by refining, which can be done quite readily. Even if sufficient titanium or niobium is present to combine with all carbon, kinetic considerations may result in that not occurring. High temperatures, such as encountered in welding, dissociate carbides. If quenched from this state, carbon can be free to form Cr23C6 if it is reheated to temperatures above 500 °C. Carbon has always been considered totally undesirable from a corrosion point of view because of its tendency to form chromium carbides. Recently, however, new processes have been developed to supersaturate carbon in austenite below the temperatures at which it has sufficient mobility to form carbides. This socalled colossally supersaturated austenite results in very high hardness (Fig. 12) and corrosion resistance over limited depths. From this, however, we can see that carbon, like nitrogen, is actually beneficial to corrosion resistance in solid solution, although this is not observed at normal concentrations. It is possible to see that if it could be kept in solution it would be appropriate to give it a factor of around 10 in the pitting resistance equivalent number (PREN) equation: PREN = %Cr + 3.3(%Mo) + 30(%N) + 10(%C)

(Eq 11)

This is consistent with the similar thermodynamic interaction coefficients that carbon and nitrogen share with regard to chromium.

Chapter 6: Austenitic Stainless Steels / 79

High-Temperature Alloys The austenitic stainless steels can have an exceptional combination of strength and corrosion resistance at temperatures above 500 °C. They are often called on to resist attack by oxygen, sulfur, carburizing, nitriding, halogens, and molten salts. Austenitic stainless steels are the most creep resistant of the stainless steels. Alloying with carbon, nitrogen, and niobium produces the greatest strength at elevated temperatures. Refer to the properties database for comparisons among the grades. This discussion concentrates on their resistance to hightemperature environments, which is their salient characteristic. Oxidation Resistance. Resistance to oxidation comes from the protective Cr2O3 scale that forms on the surface of the material. Above about 18% chromium, a continuous scale forms. The scale acts as a barrier to oxygen and greatly slows further oxidation of metal below the scale. Below the composition at which complete Cr2O3 coverage occurs, the film will also contain the less-protective spinel FeCr2O4. The Cr2O3 scale is more protective because it better restricts the diffusion of oxygen to the interface between the scale and the base metal, which is where the oxidation reaction occurs. As the oxide grows, the path to the interface lengthens, and the rate of oxidation slows. This generates the parabolic-type oxide growth rate that characterizes these alloys. The rate of oxide growth is expressed simply as: R = kt

(Eq 12)

in which the rate is in units of mass gained per unit of surface area and time. This rate is a strong function primarily of chromium level, as can be seen in Fig. 13. The rate increases exponentially with temperature since diffusion governs this phenomenon. The rate drops dramatically as chromium reaches the concentration necessary to generate the protective Cr2O3 layer. Above this sufficiency level, further increases in chromium are not as beneficial; they mainly provide a reservoir of chromium to re-form the Cr2O3. As long as chromium can diffuse to the interface at a sufficient rate to satisfy the incoming flux of oxygen, the parabolic rate is maintained. If there is insufficient chromium flux, then the oxygen penetrates beyond the interface and forms Cr2O3 in situ. The oxide will change to

Fig. 13

Variation of parabolic oxidation rate with chromium level and temperature. Source: Ref 18

the less-protective FeCr2O4, and the scale growth rate will increase beyond the parabolic relationship, leading to breakaway oxidation. The breakaway temperature increases with increasing chromium level. The austenitic alloys benefit over the ferritic alloys from the presence of nickel. For a given chromium level, oxidation rates decrease with increasing nickel content. Figures 14, 15, and 16 display this relationship. The optimal range for the iron base stainless steels, shown in Fig. 14, is reached by the commercial alloy 310, with 25Cr-20Ni composition. Alloying can alter the oxidation-resisting performance of the austenitic stainless steels. Some elements form more protective oxide layers than Cr2O3. Aluminum and silicon are most useful in this regard. Aluminum forms a layer of Al2O3 that is more restrictive to oxygen diffusion than is Cr2O3, as does silicon through the formation of SiO2. The alloys 302B, 153MA, and 253MA all use elevated silicon levels. Aluminum’s strong ferrite-promoting tendency restricts its utility in austenitic grades, however. While the gains from using under 3% silicon are impressive, rare earths can yield even greater benefits from mere trace additions. 153MA (UNS S30415) is a variation on 304 using silicon and cerium. Cerium appears to act at the metal-scale interface such that the oxides

80 / Stainless Steels for Design Engineers

Fig. 14

Influence of nickel on oxidation of iron-chromium alloys. Source: Ref 19

Fig. 15

Isooxidation curves. Source: Ref 20

formed are thinner, tougher, more adherent, and more protective, although there is no consensus on the mechanism. Figure 17 shows the improvement quantitatively (Ref 21). Because austenitic stainless steels have a greater thermal expansion coefficient than fer-

Fig. 16

Corrosion rates for various stainless steels and nickel base alloys. Source: Ref 20

ritic alloys, they stress their scale more during thermal cycling. This can fracture the scale, causing spalling and rapid subsequent oxidation of the underlying metal. This serious performance flaw also is remedied by rare earths, as shown in Fig. 18 (Ref 21).

Chapter 6: Austenitic Stainless Steels / 81

Fig. 17

Fig. 18

Comparison of rare earth-alloyed stainless alloys to conventional stainless alloys, 4833 = 309S, 4845 = 310S. Source: Ref 21

310S (4845) versus rare earth-modified 253MA in cyclic oxidation. Source: Ref 21

While the mechanism by which rare earths make the scale tougher and more adherent are vague, their effect in making austenitic alloys much better at resisting high-temperature oxidation, especially cyclic oxidation, are undeniable. Alloying elements can also be detrimental to oxidation resistance. Manganese, an even more potent oxide former than chromium, forms a manganese-chromium spinel that is less protective than the Cr2O3. Molybdenum and tungsten, which are refractory metals and are beneficial to aqueous corrosion resistance, form volatile, lowmelting oxides (MoO3 and WO3) that promote catastrophic oxidation (Ref 22, 23). Vanadium also forms an oxide, V2O5, which melts at 660 °C and can also cause catastrophic oxidation.

The formation of an oxide on stainless steel should be understood to imply de facto the depletion of chromium from the underlying metal surface. Whether the scale is formed in service, during heat treating, or during welding the surface, once the oxide is removed, there will be less chromium than the bulk alloy, often by a very significant amount, and therefore the corrosion resistant will be less. This is why welds must have not only their heat tint removed, but also the underlying metal which is depleted in chromium, to a depth on the order of 10 μ (Ref 24). Microstructure can also affect oxidation resistance. As a generalization, it can be said that changes that promote the diffusion of chromium assist in the formation of a protective scale and improve oxidation resistance. Thus, cold work and finer grain size are positive factors via their enhancement of chromium diffusion. At the very highest temperatures, Cr2O3 can be further oxidized to CrO3, which has significant vapor pressure above about 1100 °C. The compositions of some of the main high-temperature austenitic alloys mentioned here are given in Table 2. Other Environments. The most common addition species that aggravates high-temperature oxidation is water vapor. At 10%, water vapor will increase oxidation by a factor of ten. It acts by increasing the porosity of the oxide scale and by promoting formation of the volatile CrO2

82 / Stainless Steels for Design Engineers

(OH)2 species. As a rule of thumb, maximum service temperatures should be reduced by 50 to 100 °C in the presence of steam. Halogens can attack oxide scales and cause their degradation or volatilization. Carburization and nitriding are best prevented by an oxide layer that forms at very low oxygen partial pressures as chromium and silicon contents are increased. Austenitic alloys have no advantage over ferritic in this regard. Intermetallic Phases. Transition elements may combine to form intermetallic phases in which the formula can vary from B4A to A4B. Table 3 lists the most common secondary phases encountered in austenitic stainless steels, i.e., apart from austenite and ferrite. Sigma formation is retarded by nitrogen, so alloys such as 153MA are less prone to it. Lower chromium and higher nickel are beneficial. Silicon and aluminum are detrimental, as is molybdenum. The most relevant is σ. It can contain as little as four iron to one chromium or molybdenum in a tetragonal structure. Thus, it can exist in many conventional austenitic alloys. Other relevant phases are χ and Laves. The greatest risk from these phases is the loss of room temperature toughness, followed by some loss of corrosion resistance. Table 2 Alloy

In lean austenitic alloys used in high-temperature, 600 to 1000 °C service, formation times are relatively long, on the order of 100 h or more. In richer alloys, such as 310, times can be as short as several hours. If the temperature at which the alloy is to be used is one in this temperature range, then some σ is a foregone conclusion, and while σ will have little detrimental effect on short-term properties at these temperatures, long-term properties such as creep, stress rupture, and especially rupture ductility are degraded by σ. For alloys, σ is an even greater concern as these are prone to its formation and can inadvertently form some during processing. If such alloys are intended for use near room temperature, then their toughness will be seriously reduced by the brittle σ, which will form first at triple points and then throughout grain boundaries. Because of this morphology, just a few percent intermetallic phase can cause toughness to decrease by an order of magnitude. High-Temperature Mechanical Properties. Above about 500 °C, yield strength is less appropriate than creep strength in assessing the adequacy of an austenitic stainless steel for structural purposes. The resistance of a material to creep is generally measured by the creep rupture strength, which is the stress that causes

Notable high-temperature austenitic alloys Designation

C

Cr

Ni

Mo

Mn

Si

Other

Max temp, °C

8.1 8.1 9.1 9.5 12.2 11.1 10.5 19.2 35.0 35.0 34.5

... ... ... ... ... ... ... ... ... ... 2.4

1.8 1.8 1.8 0.6 1.7 1.8 0.6 1.6 1.5 1.7 1.1

2.5 0.50 0.50 1.3 0.50 2.0 1.5 0.60 ... 0.90 0.40

... ...

950 820 820 1000 1040 1040 1100 1090 1200+ 1200 1200

Table 3

Secondary phases in austenitic stainless steel

NbC NbN TiC TiN Z-phase M23C6 M6C σ-phase Laves phase χ-phase G-phase

0.07 0.08 0.03 0.15 0.07 0.07 0.17 0.03 0.15 ... 0.04

17.8 18.8 17.8 18.5 23.0 19.8 21.0 24.6 25.0 18.0 21.0

S30215 S30409 S32109 S30415 S30909 DIN 1.4828 S30815 S31008 S35315 S33000 S35125

Precipitate

0.15 0.08 0.06 0.05 0.08 0.08 0.08 0.05 0.05 0.06 0.04

N

302B 304H 321H 153MA 309S 309Si 253MA 310S 353MA 330 332Mo

Structure

fcc(a) fcc fcc fcc Tetragonal fcc Diamond cubic Tetragonal Hexagonal bcc(b) fcc

(a) fcc, face-centered cubic; (b) body-centered cubic

Parameter, (Å)

a = 4.47 a = 4.40 a = 4.33 a = 4.24 a = 3.037, c = 7.391 a = 10.57–10.68 a = 10.62–11.28 a = 8.80, c = 4.54 a = 4.73, c = 7.72 a = 8.807–8.878 a = 11.2

0.6 Ti 0.05 Ce ... ... 1.0 Al, 0.05 Ce ... 0.05 Ce ... 0.40 Nb

Composition

NbC NbN TiC TiN CrNbN Cr16Fe5 Mo2C (e.g.) (FeCr)21Mo3 C; Fe3Nb3C; M5SiC Fe,Ni,Cr,Mo Fe2Mo, Fe2Nb Fe36Cr12 Mo10 Ni16Nb6 Si7, Ni16Ti6 Si7

Chapter 6: Austenitic Stainless Steels / 83

rupture after 10,000 or 100,000 h. If deformation is a greater concern, however, the creep deformation strength, that is, the stress that results in a strain of 1% after 10,000 or 100,000 h, can be used as a basis for design. Cold work and precipitates tend to be ineffective strengtheners at temperatures that produce solution annealing and precipitate coarsening (overaging). Thus, solid solution hardening is

Fig. 19

Charpy V toughness after 200 hr aging

Fig. 20

Relative 100,000-h creep strength

preferred. Substitutional elements have limited effect, but interstitial solid solution elements, such as carbon and nitrogen, are quite useful. Nitrogen is the better addition in this regard, plus it has the collateral benefit of strongly retarding intermetallic phase precipitation. Figures 19 to 22 compare mechanical properties of the major high-temperature austenitic alloys (Ref 25).

84 / Stainless Steels for Design Engineers

Fig. 21

100,000-h creep rupture strength

Fig. 22

High-temperature, short-time yield strength

Corrosion-Resistant Austenitic Alloys Stainless steels are almost always chosen at least in part for their corrosion resistance. In normal atmospheric conditions, alloys with more than 10.5% chromium do not rust. Austenitic alloys require higher levels than this to stabilize the austenitic structure at room

temperature, thus giving the common perception that they are superior in corrosion resistance. The main advantage austenitic alloys have is their ability to utilize the powerful and inexpensive alloying element nitrogen. That is the key aspect of the more modern austenitic stainless steels that have come into use in the last two decades.

Chapter 6: Austenitic Stainless Steels / 85

The ion that is most aggressive against stainless steel is one of the most pervasive in our environment. The chloride ion is found in abundance over the entire earth. It is, of course, in seawater, but also in the rain, on roads, in food, and even in our bodies. Chlorides destabilize the passive film. If the conditions of chloride concentration, temperature, and acidity are sufficiently aggressive to break down the film, then active corrosion ensues. If this is highly localized because of a local weakness in the passive film, then pitting occurs. Such a pit may be unstable and repassivate, or it may grow without limit. Other halides have the same effect, but they are less ubiquitous. Because of the specific virulence of the chloride ion and because of its universal presence, corrosion-resistant austenitic stainless steels all look like they were designed to resist chloridepitting attack. Pitting in stainless steels is in most instances the threshold level of corrosion. Crevice corrosion is, however, more severe and usually design limiting vis-à-vis corrosion. Crevices can exist not only in deliberate joints but also under environmental deposits, paint films, weld splatter, etc. There are other environments in which the resistance follows different rules, such as oxidizing acids, bases, and organic acids, but these are best regarded as exceptions.

Fig. 23

The main factors in the resistance of austenitic alloys to pitting attack are generally given by: PREN = %Cr + 3.3(%Mo) + 30(%N)

(Eq 13)

Pitting resistance is measured by ASTM G 48 (practice C) in which the lowest temperature at which pitting occurs in a 6% FeCl3 is measured. This is the critical pitting temperature, CPT. The relationship between PREN and CPT is shown in Fig. 23 (Ref 26). The function of chromium in the passive film is intuitively clear. As the chromium content of an alloy increases, the ready reservoir of chromium to form the chromium-rich layer is facilitated. The roles of molybdenum and nitrogen are subtler and are still subject to controversy. It is the subject of much research, which has been summarized in reviews. The obvious paradox is how can elements that are not active in the passive film be so effective in maintaining its integrity. We do know that the essential chromium in the matrix of stainless steels is quite reactive and will form compounds with carbon, oxygen, sulfur, and other transition elements. When it does, it is no longer effective as a passive film former. The regions from which the chromium diffused to form the chromium-rich phase will be poor in chromium unless subjected to a lengthy homogenizing anneal. Most theories of pitting founder at the start because they assume

Critical pitting temperature versus pitting resistance equivalent number (PREN); SUS 329J4L = S31260, YUS 270 = S31254. Source: Ref 26

86 / Stainless Steels for Design Engineers

a homogeneous passive film, which is an impossible goal in reality. The chapter on corrosion deals with this topic in more depth. The passive layer is extremely thin compared to oxide layers. It is on the order of 1 to 10 nm thick. Its formation does not cause chromium depletion beneath it, as oxide layers do. As the alloy content of chromium and molybdenum increase, the film is thinner, and the current density required to form the film is correspondingly reduced. The corrosion-resistant austenitic stainless steel grades range from 316 to the various highmolybdenum, high-nitrogen alloys commercialized in the last ten years, the most notable of which are listed in Table 4 with their typical analyses. Early grades were based on alloying with chromium and molybdenum with sufficient nickel added to preserve the austenitic structure. Each of these elements facilitates the formation of the passive film and reduces the corrosion rate in the active state. Further experimentation showed that molybdenum was not beneficial under highly oxidizing conditions, but that silicon was helpful under such conditions. Copper was beneficial against sulfuric acid. Alloy development came in stages. First, 317 was the most significant corrosion-resistant alloy. Then, more chromium and molybdenum were added, and the class of alloys known collectively as the 6%Mo alloys was commercialized. Allegheny’s AL-6X™ is representative of this group. With PRENs of around 40, these alloys were resistant to seawater at ambient but not at elevated temperatures. This left a great deal wanting in corrosion resistance. These alloys were also very difficult to process, at least in part because they rapidly formed brittle grain boundary σ-phases in as little as several minutes at some temperatures. This limited chromium and molybdenum levels to a total of about 30%. Table 4 Alloy

316L 316Ti 317L 317LM 904L JS700 254SMO 4565 654SMO AL6-XN Al6-XN Plus

The discovery that nitrogen was beneficial against corrosion permitted a breakthrough in alloy development by the 1980s. Nitrogen was increased to around 0.20% from nominal levels of 0.05%. This was found to increase PREN by another five units, but more importantly, also suppressed σ formation to times that permitted thicker sections to be welded without embrittlement. Research into the thermodynamics of nitrogen in austenite showed that manganese increased the solubility of nitrogen appreciably. This permitted even higher levels of total alloying to be achieved. This was exploited in the alloys UNS S34565 and S32654, which contain 3 to 6% manganese and about 0.50% nitrogen. The PRENs of these alloys are over 50, which gives them a critical pitting temperature around 100 °C. Table 5 lists the performance of the various popular corrosion-resistant austenitic stainless steels. The advances are quite appreciable and made stainless steel a viable material for many applications for which previously nickel base or titanium alloys had been required. The obvious question in view of the success of the use of high manganese levels in conjunction with high nitrogen levels in the most highly alloyed austenitic stainless steels is when this approach will be used for the medium-alloyed austenitics to make alloys superior to 316, 317, and 904 without the high levels of nickel and molybdenum that render these alloys so expensive. It does not take much imagination to envision alloys equal to 904L in PREN with less nickel and molybdenum than 316L that would be almost totally resistant to intermetallic phase precipitation and have much greater resistance to SCC because of higher austenite stability. The same case could be made for a 317-equivalent alloy in corrosion resistance with less than 7% nickel and 0.5% molybdenum, essentially a 301 in alloy cost. In view of these trends in

Typical compositions of corrosion-resistant austenitic stainless steels Designation

C

N

Cr

Ni

Mo

Mn

Si

Other

Other

S31603 S31635 S31703 S31725 N80904 N08700 S31254 S34565 S32654 N08367 N08367

0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.01 0.02 0.02

0.03 0.03 0.06 0.06 0.06 0.06 0.20 0.45 0.50 0.22 0.24

16.4 16.4 18.4 18.4 19.5 19.5 20.0 24.0 24.0 20.5 21.8

10.5 10.5 12.5 13.7 24.0 25.0 18.0 18.0 22.0 24.0 25.3

2.1 2.1 3.1 4.1 4.1 4.4 6.1 4.5 7.2 6.2 6.7

1.8 1.8 1.7 1.7 1.7 1.7 0.8 6.0 3.0 0.4 0.3

0.5 0.5 0.5 0.5 0.5 0.5 0.4 ... ... 0.4 0.4

... 0.40 ... ... 1.3 Cu 0.4 Cu 0.8 Cu ... 0.5 Cu 0.2 Cu 0.2 Cu

... ... ... ... ... 0.3 Nb ... ... ... ... ...

Chapter 6: Austenitic Stainless Steels / 87

alloy development, the future use of 316 and 317 should be numbered. There is no justification for the use of scarce and expensive resources such as nickel and molybdenum when cheap, abundant replacements like manganese and nitrogen are available. The use of 316 as a standard alloy should in the future be eroded by more cost-effective alloys such as the lean duplex alloys like 2003. The same environments that cause pitting corrosion also cause crevice corrosion. A crevice is a volume in and out of which diffusion is restricted to a degree that corrosion products accumulate and cause the contained environment to become increasingly aggressive in pH and [Cl–]. Conditions that are below the threshold for pitting can cause crevice corrosion. The critical temperature for crevice corrosion is also measured in 6% FeCl3 (ASTM G-48 B or D). It is the lowest temperature at which crevice corrosion occurs. This temperature, the CCT (critical crevice temperature), is lower than the CPT. GTAW a wrought alloy also lowers the CPT to about the level of the CCT. The reason for this lowering of resistance to localized attack has been thought to be related to alloy depletion caused either by dendrite coring during solidification or chromium depletion around inclusions. The relation to crevices would thus seem to be that surfaces contain numerous flaws with respect to corrosion resistance, which, while not capable of sustaining pitting, can in a crevice dissolve and alter the environment sufficiently that the new harsher environment can generally destabilize the passive film and proceed autocatalytically. Stress corrosion cracking is the bane of austenitic stainless steels. SCC occurs when there is both a tensile stress of a sufficient magnitude and a sufficiently aggressive environment. The threshold stress varies with alloy and

thermomechanical history. As a rule of thumb, the environment to initiate SCC must be sufficiently severe to cause localized corrosive attack. The most dangerous situation is one in which the expectation of pitting is marginal. The mechanism of SCC has been a subject of intense academic controversy for many years. Because of this, much of the literature has focused on arguing a case rather than clarifying the phenomenon. What can be said about SCC in austenitic stainless steels with consensus? • Risk of SCC is low at room temperature and increases exponentially with temperature. • SCC is preceded by localized corrosive attack, which has an incubation time, and then proceeds in a discontinuous manner. • Fracture may be transgranular, intergranular, or both. It is almost entirely lacking in plastic deformation with little, if any, metal loss. • Alloying or treatments that delay localized attack or stabilize austenite can delay SCC up to the point of virtual immunity. • SCC is aggravated by increased chloride concentration and acidity, but also exists in caustic environments. • Stress must exceed a threshold for a given set of conditions for SCC to occur. • Anodic or cathodic polarization may prevent SCC under conditions at which it would otherwise be expected, or it may cause SCC in environments in which it would not otherwise occur. Austenitic stainless steels are not alone in their susceptibility to SCC. All stainless steels suffer from SCC under some set of conditions of environment and material thermomechanical history. The key is to choose an alloy that is resistant under the conditions of use. To a first approximation, this means using an alloy that will not pit under the conditions of use, then designing its

Table 5 Corrosion resistance ratings of various austenitic stainless steels, using 30 factor for nitrogen Alloy

316L 316Ti 317L 317LMN 904L JS700 254SMO 4565 654SMO AL6-XN AL-6XN Plus

Designation

PREN(a)

CPT oC

CCT oC

S31603 S31635 S31703 S31726 N80904 N08700 S31254 S34565 S32654 N08367

24 23 30 34 35 36 46 53 63 50 50 min

15 15 25 30 40 43 73 90 105 78 95

–3 –3 0 4 15 15 38 50 75 43 60

(a) PREN, pitting resistance equivalent number.

88 / Stainless Steels for Design Engineers

use to be below the threshold stress completes a sound design approach if residual stresses can be accurately known. Otherwise, assuming that the metal will have residual stresses equal to 100% of the yield strength is the prudent engineering approach. Figure 24 shows threshold stress for a number of alloys. Special Corrosive Environments. Knowledge of the ability of the various stainless steels to resist specific environments is essential to the design process. This information is extensive since it must correlate many environments and temperatures for many materials. Hence, refer to the publications of organizations such as the National Association of Corrosion Engineers (NACE) or to the Web sites of companies such as Allegheny Ludlum or Outukumpu, where such information is available freely. The more reputable producers will give assistance on specific questions. Engineering forums on the Internet, such as www.eng-tips.com, should also be considered a resource. The following discussion presents just the principles of the resistance of austenitic stainless steels to specific, more common environments. Sulfuric acid is common, aggressive, and must be contained. Figure 25 shows the isocorrosion curves for several alloys in pure sulfuric acid. Alloy 20, 904L, and alloy 825 were developed specifically for sulfuric acid service. Each contains several percent copper that, while not beneficial against pitting, concentrates in the passive film and diminishes general corrosion. Molybdenum and tungsten are also very beneficial for resistance to sulfuric acid. Phosphoric acid is similar to sulfuric acid in its effect on austenitics but somewhat less aggressive.

Fig. 24

Threshold stress for stress corrosion cracking (SCC) for various alloys. Source: Ref 27

Nitric acid is not particularly aggressive against stainless steels. Resistance to it is proportional to chromium content. So, attack, should it take place, is preferably at grain boundaries, where segregation of elements such as carbon, phosphorus, and silicon can lower chromium locally. These elements are kept low for nitric acid service. Pitting is not a risk. Standard usage is: • Below 50% concentrations and below 100 °C, 304L and 17% Cr ferritics are used. • Around the 65% aziotrope, 310 is most resistant, especially a low carbon version, but 304 NAG with low carbon, phosphorus, and silicon is more often used. • For 98% solutions or for lower concentrations that contain other stronger oxidizers, alloys with 4% Si, 18% Cr, and 15% Ni or 5% Si, 17% Cr, and 17% Ni (UNS S30600 and S30601) have been developed. Hydrochloric acid, not surprisingly, is quite aggressive against stainless steel. It is very effective in destabilizing the passive film. Thus, resistance to hydrochloric acid is simply an extreme case of resistance to pitting in chlorides with resistance given by Eq 12. Only the most highly alloyed austenitic alloys, such as AL6XN®, should be considered and then under conditions that are tolerable, such as those shown in Fig. 26. Strong bases such as NaOH and KOH are not especially aggressive against stainless. The 17% chromium alloys can be used up to 50 °C, while 304L can be used to 90 °C. As is the case with nitric acid, chromium and nickel are beneficial, while molybdenum is counterproductive. The 25% chromium alloys such as 310 or an equal

Fig. 25

Isocorrosion in pure sulfuric acid. Source: Ref 28

Chapter 6: Austenitic Stainless Steels / 89

Fig. 26

Resistance to hydrochloric acid. Source: Ref 28

chromium duplex can be used to 150 °C, above which temperature nickel base alloys are required. High-chromium ferritic stainless steels are also very good choices. Organic acids are generally less aggressive against stainless than are mineral acids since they are less dissociated in solution. They become hazardous when they contain chloride ions, at high temperatures, or when they dissociate strongly, such as with formic acid. Because of the large number of organic compounds that may be considered, refer to the various corrosion tables. Surface Finish. The corrosion resistance of austenitic stainless steels is quite dependent on surface condition, as are other stainless steels. Treatments that enhance surface concentrations of beneficial elements or remove detrimental constituents can greatly alter performance. Oxide formation depletes surface chromium, so strong pickling or electropolishing of the descaled surface is especially important. Studies have shown chromium depletion of a maximum of 6% extending 10 μ before reaching bulk chromium levels. This is equivalent to the depletion seen in sensitization. The increase in attack rate from this depletion is huge. A 1000fold increase in weight loss in the ASTM G 48 B test has been seen by a superficial loss of 6% chromium. Likewise, surface abrasion, especially coarse abrasion, has a major detrimental effect. The 120 grit #3 finish often seen on stainless reduces pitting resistance by as much as the equivalent of 5 PREN, that is, equal to a reduction in chromium content of 5%. Rolled finishes are much preferred. The mechanism for this has not been clearly established; exposure of MnS inclusions, the microcrevices abrasion produces, and residual stress have been cited as possible

contributing causes. Powder injection-molded stainless components often have porosity that is generally spherical. When exposed to the surface, such pores act as crevices and lower the pitting potential also. All of these factors are operative and can act in unison. Very fine abrasive polishing causes little residual stress and has very minimal crevice creation. Thus, mirror-type polished finishes do not degrade corrosion resistance, but they do not enhance corrosion resistance as does electropolished mirror finishes, which remove chromium-depleted sites, which can initiate pitting. REFERENCES

1. A.L. Schaeffler, Constitution Diagram for Stainless Steel Weld Metal, Met. Prog., Vol 56, Nov 1949, p 680–688 2. W.T. Delong, A Modified Phases Diagram for Stainless Steel Weld Metals, Met. Prog., Vol 77, Feb 1960, p 98 3. Design Guidelines for Selection and Use of Stainless Steel, SSINA,1998, p 3 4. K.-J. Blom, “Press Formability of Stainless Steels,” paper presented at Stainless steels ‘77 5. F.B. Pickering, “Physical Metallurgical Developments in Stainless Steel,” paper presented at Stainless Steel ‘84, Goteborg 6. Q.-X. Dai et al., Chin. Phys., Vol 11, 2002, p 596–600, doi:10.1088/1009-1963/11/6/315 7. Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 564 8. Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 565 9. P. Marshall, Austenitic Stainless Steels, Microstructure and Mechanical Properties, Elsevier, 1984 10. Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 579 11. H. Nordberg, Mechanical Properties of Austenitic and Duplex Stainless Steels, Innovation in Stainless Steels ‘93 (Firenze), 1993, p 2.217 12. Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 566 13. S.J. Rosenberg and C.R. Irish, J Res. Nat. Bar. Stand., Vol 48, 1952, p 40 14. Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 410 15. M. McGuire, “A Diffusion Model for the Influence of Oxygen and Sulfur on the NonEquilibrium Distribution of Chromium in

90 / Stainless Steels for Design Engineers

16. 17. 18. 19. 20. 21. 22.

Austenitic Stainless Steel Welds and Slabs,” paper presented at Proceedings MS&T ‘04,2004 R. Stickler and A. Vinckier, Trans. ASM, Vol 54, 1961, p 362 Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 568 Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 448 Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993., p 453 Aciers Inoxidables, Les Editions de Physique, Les Ulis, Paris, 1993, p 454 www.outukumpu.com W.C. Leslie, Mechanism of Rapid Oxidation at High Temperature, Trans. ASM, Vol 41, 1958, p 1213–1219

23. N. J. Grant, Accelerated Oxidation of Metals at High Temperature, Trans. ASM, Vol 44, 1961, p 128–137 24. J.F. Grubb, paper 04291 presented at Corrosion 2004, NACE, 2004, p 1–15 25. ACOM Files, High Temperature Stainless Steels, www.outukumpu.com 26. J. Okamoto et al., A Super-Austenitic Stainless Steel for Tubing and Piping Applications, Nippon Steel Technical Report 90, July 2004 27. ACOM Files, High Temperature Stainless Steels, www.outukumpu.com 28. Allegheny Technologies, “AL6-XN® Alloy”

Stainless Steels for Design Engineers Michael F. McGuire, p 91-107 DOI: 10.1361/ssde2008p091

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 7

Duplex Stainless Steels Summary THE NEWEST FAMILY of stainless steels is the duplex alloys. The mixture of ferrite and austenite in their structure gives them higher strength than either phase by itself. Duplex alloys have at least 20% chromium, so they are considered as highly corrosion-resistant alloys but not high-temperature alloys because of embrittling phases. Their low nickel content makes them more economical than austenitic alloys of the same level of corrosion resistance, especially when their greater strength can be utilized to reduce the amount of material required. They should largely replace alloys such as 316L and317L in the future.

Introduction Duplex stainless steels are the newest and fastest-growing alloy group in the stainless steel family. They are called duplex because at room temperature they consist of two phases, ferrite and austenite. Discovered in the 1920s, they languished in a suboptimized and underutilized state until recently. They possess excellent strength, toughness, and corrosion resistance. They also display exceptional resistance to stress corrosion cracking (SCC) and corrosion fatigue. The leaner grades, such as 2304, correspond to 316L in corrosion resistance but have double the yield strength, while the higher alloy grades like 2507 compete with the 6% molybdenum superaustenitics in corrosion resistance while still possessing much greater strength. Their limitations lie in their lack of cryogenic toughness and their inability to withstand temperatures much above 300 °C without forming embrittling phases. But between –100 and 300 °C

they are exceptional materials. Whether these duplex alloys will grow to the full extent of their potential depends on several factors: • Will high nickel and molybdenum prices be sufficient motivation to drive designers to explore alternatives to traditional austenitic grades? • Will producers overcome their inhibition to aggressively market these grades through their cost-saving potential? • Will producers perfect the techniques to produce these grades reliably so that their availability is unquestioned? • Will design codes change to correctly reflect the duplex materials’ higher ratio of yield strength to tensile strength? Why are there such issues with a family of alloys that has been successfully used for 20 years? The concept of duplex stainless steels is simple: islands of austenite in a continuous matrix of highly alloyed ferrite, as seen in Fig. 1. This

Fig. 1

Wrought 2205 duplex microstructure

92 / Stainless Steels for Design Engineers

combination in principle offers high strength because of the possibility of refining the dual-phase grain structure and thereby raising yield strength according to the Hall-Petch relationship as well as by solid solution hardening, especially with nitrogen. In addition, the absence of a continuous austenite phase provides relief from SCC by having any propagation of cracks in austenite arrested by the ferrite phase. The optimization of the alloy system had to wait for two events, both related to nitrogen. First, the control of nitrogen in the refining by the argon oxygen decarburization (AOD) process allowed the control nitrogen content up to the solubility limit. Second, the understanding of the thermodynamics of the alloy system became understood and reduced to a computer model. At this point, the alloys developed over the first 50 years of development became obsolete, and new grades with higher nitrogen vastly improved the performance and user friendliness. Why this was so important can be seen by studying the structure of these alloys.

Structure and Alloy Design The ideal structure of a duplex grade would be a stable 50-to-50 ratio of austenite to ferrite at all temperatures at which it is to be used without other phases. The austenite would be islands

Fig. 2

in the ferrite matrix, and each phase would have equal corrosion resistance despite having different compositions. It took a long time for that to be accomplished. Figure 2 shows a simple Fe-Cr-Ni constitutional diagram (Ref 1). The salient points are that the typical successful alloys nearly bisect the two-phase field for austenite and ferrite. It is also obvious that the composition of the austenite and the ferrite must be quite different. Ferrite contains a great deal more chromium than austenite; hence, its pitting corrosion resistance contribution from chromium is much greater than the resistance of the austenite because in duplex grades: PREN = %Cr + 3.3 × %Mo + 16 × %N

(Eq 1)

If one were to add molybdenum to increase pitting resistance, it would preferentially partition to the ferrite, further exacerbating the differential between the two phases. This is where nitrogen saves the day. Additions of nitrogen concentrate nearly entirely in the austenite. This lowers the activity of chromium and thereby effectively attracts more chromium to the austenite phase than would otherwise be present. This stabilizes the austenite, keeping the ratio of ferrite to austenite more nearly constant with temperature. The pitting resistance of the austenite increases signifi-

The Fe-Cr-Ni phase diagrams. The shaded area results from nitrogen additions

Chapter 7: Duplex Stainless Steels / 93

cantly to approximately that of the ferrite. In addition, the nitrogen solid solution strengthens the austenite and retards the formation of intermetallic phases, which is not bad for an element that costs nothing. ThermoCalc, developed by the Swedish Royal Academy, has been an especially valuable tool in helping us understand and design better duplex stainless steels. Without being able to computer model the thermodynamics of the system, it would be impossible to project the partitioning of the various potential alloying elements. Figure 3 shows isopleth diagrams for a basic 2205 composition in which nickel is varied. The 2205 is the workhorse grade of duplex. It has a pitting resistance equivalent number

(PREN) of about 35 and fills a niche in corrosion resistance where austenitics and ferritics are lacking, greater than 317L stainless, PREN = 30, and below the 6% molybdenum grades, such as AL-6XN alloy, with PRENs of around 45. Ferritics have a gap between 442 (18Cr-2Mo) and the super ferritics (28Cr-4Mo). By varying the chromium, nickel, and molybdenum, leaner alloys can be devised that save cost based on reduced molybdenum and nickel. Conversely, more corrosion-resistant alloys with higher PRENs can also be mapped, such as Fig. 4, with the same diagrams varying nickel for a higher molybdenum level. This composition includes the important 2507 alloy. Partitioning of elements (Fig. 5) between austenite and ferrite is an important issue. The partitioning tendency is a strong function of temperature. Figure 6 shows that as temperature

Fig. 3

Fig. 5

Partitioning tendencies of various elements between ferrite and austenite. Source: Ref 2

Fig. 6

Variation of partitioning ratio with temperature. Source: Ref 2

Fig. 4

The iron-nickel diagram for 22% Cr, 3% Mo, 0.15% N

The iron-nickel diagram for 25% Cr, 4% Mo, and 0.25% N: N is a nitride, χ is chi, σ is sigma, α is ferrite, and γ is austenite

94 / Stainless Steels for Design Engineers

increases, the partitioning diminishes until at just above 1300 °C it approaches unity for all normal substitution-alloying elements (Ref 2). For nitrogen, however, the tendency is to increasingly segregate to austenite as temperature increases (Ref 2). A danger in these alloys is that austenite formed from ferrite on heating, such as during welding or annealing, will contain only the low amount of nitrogen that was in the ferrite from which it was formed, until diffusion can restore equilibrium. If the heating time does not permit this, this so-called secondary austenite will have low nitrogen and therefore low pitting corrosion resistance, as shown in Fig. 7(e). Nitrogen alters the phase stability, making austenite stable to higher temperatures. This helps keep welds from becoming excessively ferritic and disturbing the desirable 50-to-50 ratio of austenite to ferrite. Secondary austenite with low nitrogen is remedied by diffusion if the phase forms at higher temperatures at which diffusion of nitrogen can rehomogenize the nitrogen level. A crucial aspect of alloy design in the duplex alloys involves the avoidance of unwanted phases. The duplex stainless steels have all the potential problems with embrittling phases of the ferritic and austenitic stainless steels combined since they contain both as phases. Ferrite forms two main embrittling phases, α′ and σ. The α′ is generally believed to be a result of the miscibility gap that exists in the ironchromium system, by which ferrite undergoes spinodal decomposition into the iron-rich α, normal ferrite, and the chromium-rich α′, which is a brittle ordered alloy. Higher levels of chromium or the presence of copper or molybdenum exacerbate this reaction, which has a formation that follows an Arrhenius-type curve with a maximum at around 400 °C. Figure 8 shows the α′ formation kinetics for five duplex alloys (Ref 2). While duplex grades have good oxidation resistance and high-temperature strength, the α′ problem restricts their use to below about 315 °C. Ferrite and austenite both form intermetallic phases, of which the most prominent and dangerous is σ, a tetragonal phase richer in chromium and molybdenum than the ferrite from which it forms. It is brittle and forms at grain boundaries, so its precipitation has the immediate effect of lowering toughness. Cold work accelerates the precipitation process by up to an order of magnitude by virtue of its dual effect on nu-

cleation and diffusion. The areas around the newly formed σ are naturally somewhat diminished in chromium and molybdenum, so the alloy’s resistance to localized corrosion is compromised also. Figure 9 shows the TTT (time-temperaturetransformation) diagram for various high-alloy stainless steel, including austenitic, ferritic, and duplex. Alloys of all structures, ferritic, austenitic, and duplex, with high chromium and molybdenum encounter the σ problem fairly equally and in proportion to their alloy content (Ref 2). This is the reason that the use of nitrogen instead of molybdenum is so beneficial to the leaner alloys, not just in cost, but for the major reduction in rate of formation of sigma. Figure 10 shows the large reduction in σ formation enjoyed by the lean alloy AL 2003™ material compared to the higher molybdenum 2205 alloy (Allegheny Ludlum). There are other intermetallic phases in addition to σ. They include χ, R, π, and τ. These are of more research than practical interest because σ, with its bad consequences, forms sooner and in greater quantity under the same conditions compared to the others. Carbides and nitrides can also form in duplex alloys. The nitride Cr2N can form when saturated ferrite is quenched from a high temperature, as can occur in the welding process. It is possible that this would result in nearby chromium depletion and a decrease in corrosion resistance. Carbide formation does not as easily cause chromium depletion in duplex alloys because the precipitation at the ferrite-austenite grain boundary does not deplete the austenite as greatly in chromium locally because of the neighboring ferrite having a much higher diffusivity for chromium. The point is generally moot since all modern duplex grades contain less than 0.030% carbon. Table 1 lists the duplex grades currently available commercially. Figure 7 shows a series of duplex photomicrographs.

Mechanical Properties In many ways, the duplex stainless alloys represent a best of both worlds in combining traits from the austenitic and ferritic alloys. They offer high as-annealed strength with good toughness and ductility. Table 1 lists the major grades of duplex stainless steels;

Chapter 7: Duplex Stainless Steels / 95

Fig. 7

(a) As-cast duplex structure, austenite in a ferrite matrix. (b) 2205 annealed; austenite phase contains twins. (c) 2507 as-welded; weld is highly ferritic because of rapid cooling rate. (d) Same weld as (c) after homogenization anneal. (e) 7-Mo Plus with ( (dark areas) that has induced the formation of secondary austenite (arrows)

Table 2 lists typical and minimum properties for the major duplex alloys and those of some comparable ferritic and austenitic alloys for comparison.

The most striking and unexpected characteristic of the duplex grades is their high yield strength, more than double that of comparable austenitic grades.

96 / Stainless Steels for Design Engineers

Fig. 8

Kinetics of ( formation

The strength of the duplex grades is driven by the strength of the continuous ferrite phase. It owes its strengthening primarily to: • Solid solution hardening by nickel, molybdenum, chromium, copper, and manganese • Interstitial solid solution hardening by carbon and nitrogen • Strengthening by grain refinement These components have been related to the mechanical properties by the following equations (Ref 3): Fig. 9

Sigma formation kinetics at various alloy levels

Rp0.2 = 120 + 210 N + 0.02 +2( Mn + Cr ) + 14 Mo + 10 Cu + (6.15 − 0.54δ)δ + (7 + 35( N + 0.02))d −1/ 2 Rp1.0 = Rp0.2 + 40 ± 9

(Eq 2) (Eq 3)

Rm = 470 + 600( N + 0.02) +14 Mo + 1.5δ + 8d −1/ 2

Fig. 10

Delay in ( precipitation in lean duplex 2003

(Eq 4)

where δ is the ferrite content in percent, d is the lamellar spacing, and results are in megapascals. The influence of nitrogen is interesting in that at lower levels (e.g., below 0.1% nitrogen) austenite is the weaker phase, but additional ni-

Chapter 7: Duplex Stainless Steels / 97

Table 1 UNS

S32900 S31200 S31260 S31500 S31830 S32001 S32003 S32101 S32205 S32304 S32520 S32550 S32750 S32760 S32906 S32950 S39274 S39277

Table 2

Duplex compositions Name

329 44LN DP3 3RE60 2205(old) 19 D 2003 2101 2205 2304 Uranus 52N+ 255 2507 Zeron 100 2906 7–Mo Plus DP3W AF 918

N

Cr

Ni

Mo

Mn

Si

Cu

W

P

S

0.08 0.03 0.03 0.30 0.03 0.03 0.03 0.04 0.03 0.03 0.03

C

... 0.14–0.20 0.10–0.30 0.05–0.10 0.08–0.20 0.05–0.17 0.14–0.20 0.20–0.25 0.14–0.20 0.05–0.20 0.20–0.35

23.0–28.0 24.0–26.0 24.0–26.0 18.0–19.0 21.0–23.0 19.5–21.5 19.5–21.0 21.0–22.0 22.0–23.0 21.5–23.5 24.0–26.0

2.5–5.0 5.5–6.0 5.5–7.5 4.25–5.25 2.5–3.5 1.0–3.0 3.0–4.0 1.35–1.70 4.5–6.5 3.0–5.0 5.5–8.0

1.0–2.0 1.2–2.0 2.5–3.5 2.5–3.0 2.5–3.5

3.0–5.0

1.0 2.0 1.0 1.2–2.0 2.0 4.0–6.0 2.0 4.0–6.0 1.0 2.5 1.5

0.75 1.0 0.75 1.4–2.0 1.0 1.0 1.0 1.0 2.0 1.0 0.8

... ... 0.2–0.8 ... ... ... ... 0.1–0.8 ... 0.05–0.6 0.5–3.0

... ... 0.1–0.5 ... ... ... ... ... ... ... ...

0.040 0.040 0.030 0.030 0.030 0.040 0.040 0.040 0.030 0.040 0.035

0.030 0.030 0.030 0.030 0.020 0.030 0.030 0.030 0.020 0.040 0.020

0.04 0.03 0.03 0.03 0.03 0.03 0.025

0.10–0.25 0.20–0.30 0.20–0.30 0.30–0.40 0.15–0.35 0.24–0.32 0.23–0.33

24.0–27.0 24.0–26.0 24.0–26.0 28.0–30.0 26.0–29.0 24.0–26.0 24.0–26.0

6.0–8.0 6.0–8.0 6.0–8.0 5.8–7.5 3.5–5.2 6.0–8.0– 6.5–8.0

2.9–3.9 3.0–5.0 3.0–5.0 1.5–2.6 1.0–2.5 2.5–3.5 3.0–4.0

1.5 1.2 1.0 0.8–1.5 2.0 1.0 0.8

1.0 0.8 1.0 0.5 0.6 0.8 0.8

1.5–3.0 . . . 0.5 ... 0.5–1.0 0.5–1.0 0.8 ... ... ... 0.2–0.8 1.5–2.5 1.2–2.0 0.8–1.2

0.040 0.035 0.030 0.030 0.035 0.030 0.030

0.030 0.020 0.010 0.030 0.010 0.020 0.020

1.5–2.0 0.1–0.8 3.0–3.5

Duplex mechanical properties

Grade

Name

Rp0.2

Rm

A5

HB

RC

Charpy-V –40 °C, J

S31200

44LN

450

690

25

293

31

...

S31260

DP3

485

690

20

290

31

...

S31830

2205(old)

450

62

25

293

31

...

S32003

2003

450

620

25

290

30

40

S32001

19D

450

640

25

290

25

...

S32101

2101

450

650

25

290

32

40

S32205

2205(new)

460

640

25

290

32

40

S32304

2304

400

600

25

290

31

40

S32520

Uranus 52N+

550

770

25

...

28

...

S32550

Ferralium

550

760

15

302

32

...

S32750

2507

550

795

15

310

32

40

S32760

Zeron 100

550

750

25

270

...

...

S32960

7-Mo Plus

485

690

15

293

32

...

trogen strengthens the austenite so that above 0.2% nitrogen, the austenite becomes the stronger phase. The two phases are elongated parallel to the major strain axis of working such as from hot or cold rolling. As working increases, the microstructure and properties become increasingly anisotropic, with the austenite taking on a (110) [223] texture and the ferrite (100) [011] to (211) [011] (Ref 2). Because the ferrite phase controls mechanical properties, the dependence of these properties on temperature is significant since flow in body-centered cubic (bcc) structures is thermally activated. Figure 11 shows the variation of yield and tensile strengths of various grades along with that of austenite and ferrite of similar composition. Because use of these alloys

above 300 °C is not recommended, no highertemperature properties are shown. Since they have a ductile-to-brittle transition, they also are not well suited to cryogenic use. Impact Strength. Toughness is a significant consideration when using duplex alloys to replace the extremely tough austenitic alloys. Duplex alloy low-temperature toughness is intermediate to that of ferritic and austenitic alloys. This having been said, it should be noted that the duplex alloys can have excellent toughness levels, such as 100 J at –100 °C in the solution-annealed condition. As would be expected, toughness improves with decreasing grain size and deteriorates with cold work. The most deleterious effect on toughness comes from the precipitation of intermetallic phases, such as α′ and σ, which cause a sharp decrease

98 / Stainless Steels for Design Engineers

Fig. 11

Variation of ferrite, austenite, and duplex with temperature. Source: Ref 4

in toughness level and a concurrent increase in transition temperature. The combined effect of cold work and α′ can be seen in Fig. 12. Lean alloys such as 2001, 2003, and 2101 have a much slower rate of formation of α′ and are much less at risk for loss of toughness from exposure in the 300 to 600 °C range, as was shown in Fig. 11. Fatigue. Fatigue tests on duplex stainless steels indicate that they possess a fatigue limit of about 50% of the yield strength when tested in air (Ref 4). The ratio of the fatigue strength in a hostile environment to that in air is a useful measure of the complementary strong points of the duplex grades (i.e., strength and corrosion resistance). Figure 13 shows that ratio for various alloys plotted versus their PREN. As an alloy’s resistance to corrosive attack increases, its fatigue limit in a given environment approaches that in air, indicating simply that corrosion plays an increasingly small role in fatigue crack propagation as corrosion resistance increases. While this is intuitively reasonable, it is not diminished because the duplex reward the user with a higher level of yield strength and fatigue strength in air, so the net useful strength under cyclic loading is much greater than that of equivalent-PREN austenitic alloys.

Increase in transition temperature with α′ formation with aging for (a) annealed 2705 and (b) coldworked 2205. Source: Ref 4

Fig. 12

Chapter 7: Duplex Stainless Steels / 99

Fig. 13

Influence of pitting resistance equivalent number (PREN) to fatigue strength in NaCl solution versus in air. Source: Ref 2

Forming and Machining The higher strength and lower ductility of the duplex grades compared to austenitics gives them correspondingly less ability to be cold formed. Duplex alloys have sufficient ductility to be cold drawn; they behave like ferritics or austenitics of similar alloy level. This, however, is an alloy level at which excellent formability is seldom expected. Nevertheless, duplex alloys can be cold formed like austenitic alloys. Operations such as bending, drawing, and pressing can readily be performed. Bend radii should be at least twice sheet thickness. Tubing can be expanded into tube sheets, but care must be taken to produce tight roller-expanded joints. Tubing bend radii should be at least twice tubing outside diameter (OD).

Heavily formed sections should be fully annealed, not just stress relieved, whenever there is a potential for SCC in the service environment.

Corrosion Resistance Because duplex alloys are made up of two phases, ferrite and austenite, each must carry its own weight in resisting corrosion. Early alloys that were lacking in nitrogen generally had a ferrite phase that, because of the greater partitioning of the chromium and molybdenum to the ferrite, had higher corrosion resistance than the austenite. As nitrogen is added, it enriches the austenite phase preferentially until the corrosion resistance of the austenite phase reaches that of

100 / Stainless Steels for Design Engineers

the ferrite. This approach is common to all more recently developed alloys starting with the revision of 2205 from UNS S31803 to S32205, which has primarily higher nitrogen. The net result is a type of alloy that has most of the highly desirable corrosion resistance characteristics of superferritic grades without their limiting lack of mechanical properties, mainly toughness.

The duplex alloys offer important advantages in performance over the austenitic grades in a number of significant aggressive media, including sulfuric acid, hydrochloric acid, sodium hydroxide, phosphoric acid, and organic acids. This performance extends to situations in which

the aggressiveness of these media is enhanced by contamination. Sulfuric Acid. Figure 14 shows the behavior of S32304 compared to 304 and 316 in sulfuric acid. Figure 15 shows additional, more highly alloyed duplex grades. The use of copper as an alloying element in S32550 (1.5%) and S32760 (0.5%) gives them much better performance than the otherwise similar S32750. In real-life situations, such as seen in flue gas desulfurization, sulfuric acid can be contaminated with chlorides. While this contamination is deadly to 316 and 317, it has only a minor effect on the copper-alloyed duplexes (Fig. 16). Hydrochloric Acid. Historically stainless steels have had their poorest performance when confronted by hydrochloric acid. Here again, the

Fig. 14

The 0.1 mm isocorrosion curves. Source: Ref 5

Fig. 15

Fig. 16

Isocorrosion (0.1 mm/yr) performances of several austenitic and duplex alloys. Source: Ref 6

General Corrosion

The 0.1 mm isocorrosion curves. Source: Ref 5

Chapter 7: Duplex Stainless Steels / 101

copper/tungsten-alloyed duplexes show exceptionally good performance, as seen in Fig. 17. This extends the usefulness of stainless steels to an environment that had previously been off limits. Indeed, the duplex stainless steels in general can be said to be relatively indifferent to the pH of chloride solutions and are affected rather by the chloride concentration and temperature. Nitric Acid. It is fairly well known and accepted that resistance to nitric acid, which was one of the first uses of stainless steel, depends almost entirely on the chromium content. Molybdenum, in all other instances a very beneficial alloying element, has a strongly negative influence on resistance to this highly oxidizing acid. Consequently, only the leanest-molybdenum duplex alloys, such as S32304, should be

considered for use with nitric acid, and even then no advantage can be claimed. Sodium Hydroxide. Much of the older published data on the behavior of stainless steel has seemed to promote the notion that higher nickel levels were beneficial in strong bases. There seems now to be little to support that notion. Figures 18 and 19 clearly indicate, respectively, that the duplex alloys with their relatively low nickel levels significantly outperform the higher nickel 304L and 316L, with performance improving with increasing chromium content. The advantage is magnified when the environment is contaminated with chlorides, as is the case of the white liquors of kraft digesters. Phosphoric Acid. While pure phosphoric acid is not a very corrosive medium for stainless

Fig. 17

Isocorrosion (0.1 mm/yr) performance of duplex in HCl compared to 316L. Source: Ref 6

Fig. 18

Corrosion rates in boiling NaOH. Source: Ref 7

Fig. 19

Corrosion rates in white liquors plus chlorides. Source: Ref 8

102 / Stainless Steels for Design Engineers

Fig. 20 Minimum temperatures for wet phosphoric acid (WPA) with an isocorrosion rate of 0.127 mm/yr. Source: Ref 9

In combinations of acetic and formic acid, the superiority of duplex alloys is quite evident, as seen in Fig. 22. S32750 shows virtual immunity, while in mixtures contaminated with halides its performance ranks very closely to expensive nickel-based superalloys such as N06625 and N06455. Even the lower alloyed S32205 can offer an order of magnitude improvement over S31703 in hot contaminated acetic acid. Pitting Corrosion

Fig. 21

Isocorrosion (0.1 mm/yr) performances of various alloys. Source: Ref 9

steel, contaminants again can render it so. Halides are particularly common and aggressive contaminants. Figure 20 shows the substantial improvement in performance of the duplex alloys over 316L when contaminants are present. Performance again improves with increasing chromium, molybdenum, and nitrogen levels. Organic Acids. Duplex alloys perform particularly well in organic acids and have an excellent record in industrial plants. In acetic acid, 304L handles lower temperatures and concentrations. Alloys such as S32205 perform well. In formic acid, the most aggressive organic acid, S32750 is resistant at all concentrations almost to the boiling point, outperforming even titanium (see Fig. 21).

The different analysis of the two main phases in duplex alloys means that each has its own pitting resistance equivalent number, PREN. The ferrite phase has the relationship common to ferritic grades: PREN + %Cr + 3.3%Mo

(Eq 5)

while the austenite obeys the more familiar: PREN + %Cr + 3.3%Mo + 30%N

(Eq 6)

The duplex grades partition these critical elements in such a way that the overall PREN of most alloys comes out to be approximately Eq 1. If one has the actual analysis of each phase, then the proper relationship to use is Eq 2. These relationships are incomplete in that they only address the major alloying elements. Tungsten has half the value of molybdenum and is frequently included: PREN = %Cr + 3.3(%Mo + 0.5 × %W) + 16%N (Eq 7)

Chapter 7: Duplex Stainless Steels / 103

Fig. 22 Corrosion rates for various alloys in 50% acetic plus formic acid, boiling. Source: Ref 10 If nonwrought material is involved, as in ascast and welded alloys, these relationships greatly overstate PREN. This is because nonequilibrium-diminished concentrations of chromium are often found around precipitates, especially (manganese, chromium) S inclusions (Ref 11, 12) and because of lower alloy content locally due to solidification segregation, principally of molybdenum. This is most significant in welded tubing, which can have higher sulfur levels to increase weld penetration. Tube welds can be reequilibrated by high-temperature annealing, but field girth welds will show diminished corrosion resistance if unannealed. So, untreated welds can have PREN’s 5 to 15 lower than the parent alloy, which equates to the localized lowering of chromium levels. The critical pitting temperature (CPT) of welds often decreases to near the critical crevice corrosion temperature (CCT) of the parent metal. The precipitation of chromium- or molybdenum-rich second (third, in this case) phases, such as σ or α′ inevitably results in diminishment of these key alloying elements in the region surrounding the precipitate, which will make it more prone to localized corrosion. This can also occur when secondary austenite forms during the heating of alloys to high

temperatures. This austenite, which forms from ferrite, has very little nitrogen, which clearly lowers its pitting corrosion resistance. The duplex alloys stand up very well in comparison to corresponding superaustenitic alloys. Figure 23 shows how CPT varies with PREN. This ranking is not always linear, as Fig. 24 shows, with pitting potential dropping fairly rapidly with temperature and at different rates for different alloys. Figures 25 and 26 show the influence of pH and chloride concentration, respectively. In 3% NaCl (Fig. 26), the rankings show a minor variation with pH and a rational relationship to alloy content. The influence of chloride concentration is strong over a wide range of concentrations. These tests are best for judging relative performance of alloys and must be used cautiously when extrapolating lab results to service performance. The degree to which short-term tests, whether potentiostatic or strictly immersion, reflect long-term performance has not been well established. Crevice Corrosion Crevices exist both by design and inadvertently. Crevices are occluded volumes of liquid

104 / Stainless Steels for Design Engineers

Fig. 23

Fig. 24

Critical pitting temperature in seawater measured potentiostatically versus pitting resistance equivalent number (PREN). Source: Ref 13

Variation of pitting potential with temperature. SCE, saturated calomel electrode. Source: Ref 14

in which oxygen and corrosion products reach levels quite different from the exterior environment and become highly corrosive. Thus, the tighter the crevice is, the greater the restriction of diffusion between the crevice and the bulk and therefore the greater the chance of crevice corrosion occurring. An alloy’s susceptibility to crevice corrosion is proportional to its resistance to pitting corrosion under the same conditions. The CCT is lower than the CPT by about 10 to 30 °C.

The difference increases with total alloy content, as can be seen in Fig. 27. Interestingly, the difference is approximately the same as is the difference in CPT between the wrought alloy and the welded alloy. Stress Corrosion Cracking Stress corrosion cracking (SCC) has long been the Achilles’ heel of stainless steels. Only soft ferritic stainless steels are immune to it. It

Chapter 7: Duplex Stainless Steels / 105

Fig. 25

Variation of critical pitting temperature (CPT) with pH. Source: Ref 14

Fig. 26

Critical pitting temperature (CPT) as a function of NaCl concentration. SCE, saturated calomel electrode. Source: Ref 5

Fig. 27

Critical crevice temperature (CCT) and critical pitting temperature (CPT). Source: Ref 15

occurs at temperatures and in environments where stainless would be a perfect material if only it did not stress corrosion crack. The arrival of duplex stainless steels has to a very large degree ameliorated, if not solved, that problem. SCC is unfortunately poorly understood. Like pitting, whose initiation mechanism has not been identified, SCC has both its initiation and propagation mechanisms still open to debate. But the duplex alloys have good strength mainly through fine grain size and solid solution hardening, which seems to avoid the hydrogentrapping dislocation types that seem to be associated with hydrogen failures. So, while we cannot state the mechanism for SCC, we can map out the conditions under which duplex alloys are susceptible to SCC. The major environmental factors that affect SCC are chloride concentration and temperature. Figure 28 shows the remarkable advantage the duplex alloys have over the comparable austenitic alloys with regard to the temperatures at which they may be used without SCC. The duplex alloys in this regard are governed in their behavior by their ferrite matrix, through which cracks must propagate (Ref 16). Ferritic stainless steels are known for their resistance to SCC in the annealed condition. The advantage of the duplex lies in their composite-type microstructure with the crackarresting austenite phase and the toughening fine grain structure. The duplex alloys show a higher threshold stress for SCC as a percentage of their yield strength (Fig. 29) than austenitic alloys. This is in spite of their higher yield strength, again giving these alloys more usable strength. In ferrite, SCC susceptibility is a maximum below 100 °C, while in austenite susceptibility appears to begin around 50 °C and increase monotonically with temperature. The temperature at which SCC occurs at the fastest rate increases with nickel content. This is also characteristic of ferritic and martensitic materials and mirrors their hydrogen embrittlement behavior. H2S also accelerates failure in chloride environments (Fig. 30), and cold work accelerates failure and lowers threshold stress values. While duplex alloys behave in many regards like ferritic alloys in their SCC or hydrogen embrittlement response, they do not have the same relationship between susceptibility and bulk hardness. Other ferritic and martensitic alloys display pronounced susceptibility to these failures modes when their hardness

106 / Stainless Steels for Design Engineers

Fig. 28

Stress corrosion cracking (SCC) in neutral aerated NaCl. Testing duration 1000 hr. Source: Ref 5

Fig. 30

Fig. 29

Constant load stress corrosion cracking (SCC) tests in aerated MgCl2 at 150 °C. Source: Ref 5

Suggested chloride and pH limits for cold-worked duplex alloys. Source: Ref 17

exceeds Rc 22. The duplex alloys have annealed hardness over Rc 30 without being in danger. This probably simply indicates that hardness as a measure of susceptibility is valid only insofar as it reflects a certain yield strength threshold as it does in tempered martensite and is not valid for ferrite/austenite composite structures. Thus, it is very important to understand duplex SCC behavior as a separate study and not interpret it in terms of austenitic or martensitic SCC.

Chapter 7: Duplex Stainless Steels / 107

REFERENCES

1. P. Lacombe, B. Baroux, and G. Beranger, Stainless Steels, Les Editions de Physique, 2003 2. R.N. Gunn, Duplex Stainless Steels, Abington Publishing, 1997, p 28 3. H. Nordberg H, Innovation of Stainless Steel, Conf. Proc., AIM, Florence, 1994, p 2.217–2.229 4. Charles, Duplex Stainless Steels ’91, Vol 1, Beaune, Les Editions de Physique, 1991, p 3–48 5. S. Bernhardsson, Duplex Stainless Steels ’91, Vol 1, Beaune, Les Editions de Physique, 1991, p 137–150 6. J. Nichols J, 12th International Corrosion Congress, Houston, NACE, p 1237 7. E.-M. Horn, Werkstoffe und Korrosion, Vol 42, 1991, p 511–519 8. J.P. Audouard, Stainless Steel Europe, April 1992, p 45

9. Avesta Sheffield, Corrosion Handbook for Stainless Steels, 1994 10. B. Walden et al., Stainless Steel ’93, Florence, AIM, 1993, p 3.47 11. M.F. McGuire, MS&T Conf. Proc., 2004, p 831–846 12. M. Ryan, D. Williams, R. Chater, B. Hutton, and D. McPhail, Why Stainless Steel Corrodes, Nature, Vol 412, p 770 13. C.V. Roscoe et al., Duplex Stainless Steels ’86, The Hague, Nederlands Instituut voor Lasteckniek, 1986, p 126–135 14. J.M. Drugli et al., Paper 270 presented at Corrosion ’90, Las Vegas, NACE, 1990 15. S. Bernhardsson, Paper 164 presented at Corrosion ’90, Las Vegas, NACE, 1990 16. T. Kudo, H. Tsuge, and A. Seki, Stainless Steel ’87, The Institute of Metals, 1988, p 168–175 17. R. Francis, Duplex Stainless Steels ’94, Vol 3, Glasgow, TWI, 1994, paper KIV

Stainless Steels for Design Engineers Michael F. McGuire, p 109-122 DOI: 10.1361/ssde2008p109

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 8

Ferritic Stainless Steels Summary THE FERRITIC STAINLESS STEELS are the lowest-cost highly corrosion- and oxidationresisting alloys in existence. They are useful mainly as light-gauge sheet since their toughness drops off rapidly for heavier sections. Even as they have grown in use more than any other type of stainless, they could still economically displace the popular but expensive 304 for many routine applications.

Introduction Ferritic stainless steels are simplest, lowestcost stainless steels. In their minimal form, they contain simply enough chromium to overcome their inherent level of carbon impurity and hit the 11% chromium in solution required for “stainlessness.” Early in the 20th century, 430 came into being, and the attainable levels of carbon removal required 16% chromium for this to occur. So much extra chromium was required because during annealing, to develop the fully ferritic structure, carbon combines with chromium, rendering it useless as a corrosion fighter. In October 1967, the first commercial use of argon oxygen decarburization (AOD) changed the world for ferritic stainless steel. This process, in which argon and oxygen are blown through the molten metal to selectively remove carbon without removing chromium (described in detail elsewhere in this book), reduced the carbon plus nitrogen levels sufficiently that their effect could be nearly negated by small additions of titanium or niobium, which combine strongly with carbon and nitrogen and effectively remove them from solution. This process is called stabilization, and the

technology was documented long before AOD was invented (Ref 1). It was not until carbon and nitrogen levels were brought down to AOD levels that it became truly practical for ferritic alloys. The level of carbon plus nitrogen was lowered from around 0.10% to around 0.04%, and less-expensive high-carbon ferrochromium could be used instead of expensive low-carbon versions. Thus, there exist two types of ferritics: the early high-carbon types such as 430, 434, 436, and 446 and the more modern stabilized alloys led by 409 and 439. The older, unstabilized grades are not always fully ferritic. Their carbon levels cause them to form some high-temperature austenite, which transforms to martensite if quenched. This makes their welds brittle. To be used, they are normally in the annealed condition, which requires a lengthy subcritical anneal to avoid martensite and to evenly distribute chromium after all carbides have stably formed. The newer stabilized alloys behave as if they are interstitial free. They are ferritic at all temperatures (excluding for the moment the possibility of extraneous phases such as (α' and σ) and can be easily welded without fear of unwanted phases. Stabilization does not preclude excessive grain growth in the fusion or heat-affected zone (HAZ) of welds, which can render them brittle. The mechanical properties of ferritic stainless steels appear similar to austenitics strengthwise, but they lack the ductility of austenitics, and they are limited at low temperatures by brittleness and at high temperatures by softness. The lower thermal expansion coefficient of ferritics makes their scale more compatible with the base alloy and provides them with a lesser tendency to spall. This makes them excellent for high-temperature applications with thermal cycles, provided their strength is adequate.

110 / Stainless Steels for Design Engineers

The corrosion resistance of ferritics is hampered by their inability to utilize nitrogen. The absence of nickel, which characterizes these alloys, is not a problem since nickel adds little to corrosion resistance. The titanium stabilization of the modern alloys has quite a beneficial effect since titanium is a powerful deoxidizer and desulfurizer, both of which can cause local chromium depletion and pitting. Ferritics, moreover, are essentially free from stress-corrosion cracking (SCC) since they are below the threshold hardness for hydrogen embrittlement in body-centered cubic (bcc) ferrous alloys. There are a few exceptions. The main attraction of ferritic stainless steels over austenitics is their cost. The old comparison of 430 versus 304 is a bit unfair since 304 is richer in chromium. A fair comparison might be between 439 and 304. The corrosion resistance of these two alloys is barely distinguishable under normal ambient conditions. They are both very formable and weldable. The vast majority of the objects made commercially from 304 could be switched to 439 with no adverse consequence. But, if nickel is selling for $7 per pound, then the total cost of 304 versus 439 is doubled by its presence. No design engineer can afford to ignore this level of incentive to learn to use ferritic stainless steels.

Ferritic Stainless Alloys The ferritic stainless alloys generally group in low (10.5 to 12.0%), medium (16 to 19%), and high (greater than 25%) chromium. They can be stabilized or not. These distinctions are somewhat imposed after the fact. Rather than giving them an order that they truly do not possess, the most significant alloys are all listed in Table 1 with their compositions. The low-chromium ferritic stainless steels began with the development of MF-1, the predecessor of 409, in the 1960s. Its excellent corrosion resistance, compared to carbon steel; relatively low cost; good welding; and formability permitted it to replace aluminized carbon steel and cast iron in automotive exhaust systems, opening up what eventually became the largest single market for stainless steel. It was made possible by the very low carbon plus nitrogen levels the AOD process provided and the use of stabilization. Thus, 409 was an improvement on 405 in which aluminum performed a quasi stabilization, and low carbon suppressed martensite.

A similar predecessor was 410S, a low-carbon version of 410 to which some understabilizing amount of titanium is added but that still requires annealing for full ferritic properties. The key issue of the 11% chromium ferritics is how to deal with carbon and nitrogen. The 405 and 410S take the approach of minimizing it and live with annealing. The 409 uses full titanium stabilization. The hidden problem with using only titanium is that unless nitrogen levels are made very low, the amount of titanium required to combine with it can reach levels at which the first TiN precipitates in the molten metal. This slaggy precipitate agglomerates, causing casting problems and surface defects. This gave 409 a reputation as a grade unsuitable for applications that required good appearance because the titanium streaks were difficult to avoid and greatly highlighted by polishing. This has largely been overcome by better refining techniques to reduce carbon plus nitrogen to levels below 0.02% and the use of dual stabilization by titanium and niobium; 468 (UNS S40930) is such an alloy. The historical archetype of ferritic stainless steels was 430, which has existed since the 1920s and is still widely used. Its drawbacks are lack of weldability, relatively poor corrosion resistance because so much of its chromium is tied up as carbides, and modest formability. The new archetype for this medium-chromium level is 439. With 17% chromium and single (439) or dual stabilization (468), this alloy overcomes the problems of 430 and can readily replace 304 in most applications with significant cost savings. In North America, 439 is mainly used as a higher-temperature automotive exhaust alloy, but in Europe 430Ti is used extensively in more visually challenging applications, such as appliances. There, it is generally used instead of 439 whenever the part can be designed to be formed from it. Now, 434 and 436 are little used as their historical application in automotive trim finds little place in today’s automotive styling. A modern offshoot of these alloys, which are basically molybdenum enhanced 430, is 444. This alloy has roughly the corrosion resistance of 316L but is fully resistant to SCC in the welded or annealed condition. This makes it especially useful for applications such as hot water heaters, heat exchangers, and food- and beverage-processing equipment. Both the nominally 11 and 18% chromium alloys are sometimes modified to enhance their high-temperature strength or oxidation

0.01

...

0.07 0.03

0.01

0.03

S40910

S40920

AK alloy

S40930 S40940

S40975

AK alloy typical AK alloy typical ATI alloy typical ATI alloy typical Outukumpu typical S42900 S43000 S43020

S43023

S43036

S43035 S43932

ATI, AK alloys S46800

AK alloy typical AK alloy typical Outukumpu typical

409

409

409 ultraform

466 409Cb

409Ni

11 Cr-Cb

430Se

430Ti

439 439LT

439 HP 439 ultraform 468

18 Cr-Cb

4742

18SR

429 430 430F

4724

Alfa II

Alfa I

12 SR

0.08

S40900

409

0.02

0.02

...

...

0.08

...

0.015

0.02

0.1

0.04 0.03

...

0.04

0.12

... ... ...

0.12 0.12 0.12

...

...

0.025

0.08

18

17.3

18

18.0–20.0

17.5

17.0–19.0 17.0–19

16.0–19.5

16.0–18.0

14.0–16.0 16.0–18.0 16.0–18.0

13.5

13

13

12

0.015

...

11.35

10.5–11.7

10.5–11.75 10.5–11.7

10.5–11.7

10.5–11.75

10.5–11.7

10.5–11.75

11.5–14.5 12.0–13.0

Cr

0.015

0.03

0.025

0.02

0.01

0.03

0.02 ...

0.03

0.03

0.02 0.06

0.03

0.03

...

... ...

0.08 0.05

S40500 AK alloy

405 400

N

C

Designation

Ferritic stainless compositions

Alloy

Table 1

...

0.25 0.7

0.3

1 0.3

...

0.35

1 1

1

1.25

1 1 1.25

0.7

0.035

0.035

...

0.25

1

0.5

0.2

0.5 0.5

1

...

0.75 0.75 ...

...

...

...

...

0.2

0.5–1

1 1

0.75

0.5 0.05 0.5

1

0.5

1 1

0.5

1 1

Mn

0.5

0.6 ...

Ni

...

1.3

...

0.45

(continued)

...

...

...

...

...

0.45 1

... ...

...

...

...

...

0.04

...

0.25

... ...

0.25

Ti + Nb: 0.20 + 4x(C + N) to 1.10

0.35

0.20 + 4x(C+N) to 1.10 0.20 + 4x(C+N) to 0.75 Ti+Nb

...

...

0.55

...

...

... ...

...

...

... 0.20 + 4x(C+N) to 1.10

... ... ...

...

...

... ... ...

...

0.4

...

0.4

0.35 0.6

...

...

... ...

...

...

0.17

...

... ...

Nb

0.3

...

0.03

0.001

0.03 0.03

0.04 0.04 0.02

0.03

0.03 0.3 0.15 min 0.06

...

...

...

...

...

...

0.8 + 8x(C+N) Ti + Nb 10xC to 0.75 Nb

0.01 0.04 0.03

8x(C + N)

8x(C + N) to 0.15–0.50

6x(C + N) to 0.5

6x(C + N) to 0.75

... ...

Ti

0.01

0.02

0.02

0.45

0.03 0.03

S

0.04

0.06

0.04 0.04 0.06

...

... ... ...

...

...

...

...

...

0.04

0.04 0.04

0.04

0.04

0.04

0.45

0.04 0.03

P

...

...

...

...

...

... ...

...

...

...

...

... ...

Mo

1 1

1

1

1 1 1

1

0.03

0.03

...

1.3

1

1 1

1

1

1

1

1 1

Si

1.7 Al 1 Al

...

...

...

... 0.15 Al

0.15 Al

0.15 Se

... ... ...

1 Al

4 Al

3 Al

1.2 Al

...

...

... ...

...

...

...

...

0.10–0.30 Al 0.25 Al

Other

Chapter 8: Ferritic Stainless Steels / 111

S44200 ATI alloy typical S44400

442 436S

S44627

S44635 S44660 S44735 S44600 Cast alloy

E-Brite, 26-1

Monit Sea-cure 29-4C 446 CC-50

453

4762

433

ATI alloy typical Outukumpu typical ATI alloy typical

S43400 S43600 S44100

444, YUS 190-EM

Designation

434 436 441, 4509, 430J1L

0.025 0.025 0.025 0.2 0.5

0.01

0.03

0.08

0.01

0.025

0.2 0.01

0.12 0.12 0.03

C

0.035 0.035 ... 0.25 ...

0.015

...

...

...

0.035

... 0.015

... ... ...

N

Ferritic stainless compositions (continued)

Alloy

Table 1 Cr

24.5–26.0 25.0–27.0 28.0–30 23.0–27.0 26.0–30

25.0–27.5

22

24

20

17.5–19.5

18.0–23.0 17.3

16.0–18.0 16.0–18.8 17.5–18.5

Ni

3.5–4.5 1.5–3.5 0.5 0.6 4

0.5

0.3

...

0.25

1

0.6 0.3

... ... ...

Mn

1 1 1 1.5 1

0.4

0.3

0.7

0.3

1

1 0.2

1 1 1

Si

0.75 1 0.75 1 1.5

0.4

0.3

1.4

0.4

1

1 0.4

1 1 1

Mo

3.5–4.5 2.5–3.5 3.5–4.5 ... ...

0.75–1.25

...

...

...

0.75–1.25

... 1.2

0.75–1.25 0.75–1.25 ...

P

0.04 0.04 0.04 0.04 ...

0.02

0.02

...

0.02

0.04

0.04 0.02

0.04 0.04 0.04

S

...

...

Ti + Nb: 0.20 + 4x(C + N) to 0.80 Ti + Nb: 0.20 + 4x(C + N) to 0.80 Ti + Nb: 0.20 + 4x(C + N) to 0.80 ... ...

0.03 0.03 0.03 0.03 ...

0.02

...

...

Ti + Nb: 0.20 + 4x(C + N) to 0.80

8x(C + N) min

0.1-0.6

Ti

... ...

0.02

0.03

...

0.001

0.03

0.04 0.001

0.03 0.03 0.015

...

... ...

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

0.5–0.20

...

...

Other

1.5 Al 0.60 Al 0.1 REM 0.2 Cu 0.5 Cu + Ni ... ... ... ... ...

10x(C + N) . . .

...

... ...

... ... Nb + Ta 5xC 0.7 . . . 9xC 0.3–1.0 . . .

Nb

112 / Stainless Steels for Design Engineers

Chapter 8: Ferritic Stainless Steels / 113

resistance. Again, the driving force has been the requirements of the hot end of exhaust systems (e.g., exhaust manifolds). Alloying with niobium and molybdenum adds to the high-temperature strength, while additions of chromium, silicon, and aluminum increase oxidation resistance. There exists an array of proprietary alloys as shown in Table 1; these are usually developed for specific automotive needs and employ all or some of these alloying variations. The use of silicon and aluminum decreases formability and can accelerate (formation, so their use involves trade-offs. Alloys with more than 20% chromium are used specifically for high-oxidation or corrosion resistance. Despite the relative lack of hightemperature strength, these alloys are particularly useful because of their high-oxidation resistance, which they derive from the tight adherence of their oxide scale. The close match between the thermal expansion coefficient of the scale and the alloy prevents spallation of the oxide, which would lead to breakaway oxidation. This was the purpose of the earliest highchromium ferritic stainless, 446. The performance of 446 has been exceeded by lower alloyed grades, such as the aluminum-alloyed ferritics. A prime example of the state of the art is 453, which has not only 22% chromium and 0.6% aluminum but also rare earths in trace amounts (i.e., 0.1%). As in austenitic alloys, rare earths act as very powerful oxide and sulfide formers that concentrate at the metal-oxide interface and stabilize it, again preventing spallation. This type of alloys finds use in high-temperature applications such as planar oxide fuel cells. The high-chromium alloys, when used for corrosion resistance, are usually called superferritics. In the 1960s, E-Brite® was developed. To obtain high toughness, it was vacuum refined to very low carbon plus nitrogen levels. It was followed by the more capable 29-4®. Later, this alloy was stabilized and became the still-popular AL 29-4C®. (E-Brite now has a new life as a fuel cell material based on its oxidation resistance and very low thermal expansion coefficient.) These alloys saw success as replacements for 316L when SCC was a problem. This alloy and its close neighbor SeaCure® are used primarily in tubing where corrosion resistance is most important. It was developed for welded condenser tubing where seawater or brackish water is involved. It is also used in heat ex-

changers and extensively in condensing portions of high-efficiency residential furnaces. The lower-alloyed Seacure had a slight toughness advantage that permitted it to be used at wall thicknesses of 1/16 in. when AL29-4C® was too brittle. As with other ferritics, these alloys are generally only suitably tough when used in thin section size (i.e., less than several millimeters). It is difficult to say ferritic stainless steels are underutilized since they account for about half the world’s production of stainless, but there are many applications in which more expensive austenitic stainless steels are used needlessly. Ferritic stainless steels are a viable alternative to nickel-bearing austenitics when thickness is 2 mm or less and drawing and bending instead of stretch forming is permitted. There are many applications where the longer corrosion life of low-chromium ferritics should economically replace carbon steel, as they have in automotive exhaust systems. There are no technical barriers to obtaining these savings; design engineers need to learn how to use these alloys.

Metallurgy of Ferritic Stainless Steels Chromium stabilizes the ferritic structure at high temperatures. Thus, above about 11% chromium, austenite does not exist at any temperature in pure iron chromium alloys, as seen in Fig. 1. However, iron-chromium alloys devoid of carbon are not practical, so early metallurgists saw the diagram shown in Fig. 2 with the level

Fig. 1

Iron-chromium phase diagram from Thermocalc

114 / Stainless Steels for Design Engineers

of carbon at 0.20%, which represented the purity level attainable in arc furnace refining. Carbon is essentially insoluble in ferrite at ambient temperatures, and carbides of chromium and iron will form to the extent carbon is available. Since carbon diffuses interstitially much more rapidly than chromium can substitutionally, chromium is combined in situ, especially along grain boundaries, which are fast-diffusion paths. This locally depletes chromium, and the alloy is sensitized. This can be eliminated by a

Fig. 2

Iron-chromium phase diagram at 0.20% carbon

Fig. 3

Iron-chromium diagram at low carbon levels Source: Ref 2

sufficiently long homogenization anneal at a low enough temperature that carbon and nitrogen have very little solubility. This is standard in the processing of unstabilized ferritic stainless steels, such as 430. Rapid cooling of unstabilized alloys causes carbon and nitrogen to precipitate within grains. This severely embrittles the material and does not avoid sensitization. This is called high-temperature embrittlement because it comes from putting carbon and nitrogen into solution at a high temperature and then causing it to precipitate in a harmful manner. These alloys were only ferritic at room temperature if they were given a subcritical anneal to transform austenite to ferrite. Otherwise, at room temperature they would be ferrite plus martensite. There are alloys that are intended to use a mixed ferrite/martensite structure, but they are treated later as a variation from the normal ferritic alloys. The introduction of AOD refining permitted much lower levels of carbon, as seen in Fig. 3, opening the door for fully ferritic stainless steels. Carbon and nitrogen added together produce about the same effect as carbon alone. So, unstabilized fully ferritic alloys are not feasible below 20% chromium without extreme refining techniques, such as electron beam refining, which are not commercially viable for low-cost alloys. Thus, nearly all modern ferritic alloys

Chapter 8: Ferritic Stainless Steels / 115

are “stabilized.” This means that a strong carbide former such as titanium or niobium is added in sufficient quantity to combine with all the carbon plus nitrogen, removing them from solution. These reactions are simply: Ti + C = TiC

(Eq 1)

Ti + N + TiN

(Eq 2)

Nb + C = NbC

(Eq 3)

Nb + N = NbN

(Eq 4)

Titanium is the stronger getter for carbon and nitrogen. The thermodynamic driving force for carbide and nitride formation is given by (Ti )(C) =

−7700 + 2.75 T

(Eq 5)

(Ti )( N) =

−15790 + 5.40 T

(Eq 6)

It must be noted that titanium has an even higher affinity for oxygen and sulfur than for carbon, so that the removal of carbon from solution is preceded by the removal of oxygen, nitrogen, and sulfur in that order. This will be seen to have a major influence on corrosion resistance as the MnS inclusions generally associated with the initiation of pitting are not found in titanium-stabilized grades of normally low sulfur. In practice, the removal of oxygen begins in the molten state with the formation of titanium sulfide and nitride and next in the molten or solid state, depending on concentrations. It is desirable to keep sulfur and nitrogen low enough that precipitation is in the solid state so that precipitates do not agglomerate and cause large primary inclusions that become unsightly surface defects. TiCS forms in the solid state if sulfur is present; if not, TiC forms. Essentially all carbon is removed from solution below 1250 °C if carbon and nitrogen are kept as low as possible and a stoichiometric amount of titanium is available (i.e., greater than about four times the carbon plus nitrogen). The stabilization formula in various specifications is more than four times the carbon plus nitrogen because experimentally it has been found that sometimes understabilization occurs. This is due to the influence of oxygen and sulfur having prior compound formation with the titanium and less importantly that kinetic

factors prevented TiC formation. The latter effect was real in early austenitic alloys, such as 321, leading to knife-line corrosion attack after welding, but does not exist in low interstitial ferritic alloys, which have much greater diffusion rates than austenitic alloys. But, since carbon mobility is quite high, it is not practical to quench alloys quickly enough to prevent carbide precipitation as is possible in austenitics (detailed in the Chapter 6, “Austenitic Stainless Steels”). Figure 4 shows the time-temperaturetransformation (TTT) curve for an unstabilized 430-type alloy with carbon plus nitrogen of 0.08% (Ref 3). Stabilization causes nonchromium carbides to form at high temperatures, precluding chromium carbide precipitation. The net effect is that modern stabilized ferritic alloys behave as interstitial free and can be mapped using the pure iron-chromium diagram shown in Fig. 1. The rate of diffusion of carbon in ferrite is around 100 times greater than that of carbon in austenite. The solubility of carbon in ferrite is vastly lower than it is in austenite. Because of these factors, the heat treatments to avoid sensitization are essentially reversed. Carbon in austenite can be retained in supersaturation for extended periods of time. This is why austenitic L grades do not sensitize even though they are slightly supersaturated. Sensitization occurs at higher levels of carbon by prolonged heating at 600 to 850 °C. In ferritics, carbon cannot be kept in supersaturation even by the most rapid quenching, and sensitization is alleviated by prolonged heating in the 600 to 850 °C range to allow chromium to equalize where carbide precipitation has previously made it inhomogeneous.

Fig. 4

430 time-temperature-transformation (TTT) curve. K, carbideSource: Ref 3

116 / Stainless Steels for Design Engineers

Ferritic alloys, like austenitic alloys, can form intermetallic phases. The most prominent is σ, which can be seen to form in higher-chromium stainless steels (i.e., those with chromium plus molybdenum of 20% or more). Formation of σ occurs when such alloys are held between 500 and 800 °C; it is a hard, brittle tetragonal phase with equal parts iron and chromium. Thus, its formation causes chromium depletion of the adjoining ferrite. Formation requires substitutional diffusion of chromium so is slower to form than carbides, minutes rather than seconds. Since cold work enhances substitutional diffusion, it accelerates σ formation. The σ forms preferentially along grain boundaries for diffusion reasons, and this causes it to have a major embrittling effect. The σ may be redissolved by solution annealing, but regaining full homogeneity is not immediate. Another embrittling phenomenon is the formation of α'. This was named 885 °F or 475 °C embrittlement before its cause was understood. Before the nature of α' was known, it was confused with temper embrittlement, which occurs in martensitic alloys at the same temperature. Temper embrittlement is the segregation of phosphorus to prior austenitic grain boundaries and does not occur in fully ferritic alloys. The α' is the ordered equiatomic chromium iron phase that forms by spinodal decomposition; it has the same composition as σ but exists at a lower

Fig. 5

Influence of α' formation on hardnessSource: Ref 4

temperature with the same structure as ferrite but with the chromium and iron atoms in an ordered bcc matrix in which iron and chromium occupy sites equivalent to two interlocking simple cubic matrices. Because the lattice so closely matches that of ferrite, the precipitate is coherent and causes hardening. The α' embrittlement causes an extreme loss of toughness as well as hardening. It also causes a loss in corrosion resistance via the chromium depletion of that part of the matrix that surrenders chromium to the α'-phase. Figures 5 and 6 show the hardening effect of α' and the resulting loss of toughness, respectively (Ref 6).

Mechanical Behavior Ferritic stainless steels are quite similar in their mechanical behavior to carbon steel. The main influence of chromium is to produce some solid solution hardening. Let us review the strengthening mechanisms of bcc iron. Pure iron is an extremely soft material with a yield strength well under 10,000 psi. This softness is not seen in practice because steel is never pure. Carbon has an extremely powerful effect on hardening, as does nitrogen. The influence of substitutional alloying elements is also quite significant. According to

Chapter 8: Ferritic Stainless Steels / 117

Fig. 6

Influence of α' formation on toughness Source: Ref 5

Paxton (Ref 7), the misfit of solute atoms causes lattice strains proportional to the amount dissolved and provides strengthening through the lattice friction term. This mechanism also increases the impact transition temperature unfavorably. Elements that produce a refining of grain size are the exception to this general rule in carbon steel, but the lack of an austenite-to-ferrite transformation in stabilized ferritic stainless steels negates this benefit for them. Figure 7 shows that fairly common ingredients and impurities have strong hardening effects (Ref 6). Manganese and silicon are normally deoxidizers, but in titanium-stabilized alloys, titanium takes over the deoxidizing role so their presence can be limited. Phosphorus is virtually impossible to refine from stainless steel, so its presence at around 0.02% is normally a given unless low-phosphorus raw materials are used as a starting point. The worst toughness-inhibiting effects come from interstitial elements to grain boundaries: oxygen, carbon, and nitrogen. The effect of carbon plus nitrogen on transition temperature is profound, as seen in Fig. 8 (Ref 8). Stabilizing removes the interstitial carbon and nitrogen, along with oxygen and sulfur, from solution. This does not produce a major softening, however, because the precipitate itself has a hardening effect.

Fig. 7

Influence of substitutional elements on hardness of iron alloys

The softest ferritic stainless alloys are the 409 variations made for highly formed exhaust system components. They contain as little manganese, silicon, nickel, and other substitutional elements as possible and have a minimum of carbon plus nitrogen, so that the resulting precipitate fraction after titanium addition is as low as possible. To maximize softness and formability, titanium and niobium in excess of that required for stabilization must also be minimized as they will cause solid solution hardening.

118 / Stainless Steels for Design Engineers

Fig. 9

Fig. 8

Corrosion of titanium-stabilized 29% Cr plus 4% Mo alloys in ASTM A 763 Y test. Source: Ref 11

Influence of interstitial carbon and nitrogen on toughness transition temperature Source: Ref 8

Stabilization Stabilization is essential to ferritic stainless steels to avoid the precipitation of grain boundary carbides. Combined carbon plus nitrogen levels below 100 ppm are necessary to avoid both sensitization and embrittlement, but without proper heat treatment even alloys of this purity can incur debilitating loss of toughness due to carbide and nitride precipitates (Ref 9). These levels are not economically attainable for commercial alloys, so stabilization is the correct engineering answer. Stabilization is generally considered as the simple gettering of carbon and nitrogen by a suitable carbide and nitride former. It was not known until about 1980 just what the mechanisms of embrittlement were in the ferritic stainless steels, however. The distinguishing of α' from those related to interstitials and their stabilizers (Ref 10) permitted stabilizing elements to be optimized. Titanium combines with carbon and nitrogen stoichiometrically by: Ti = 4 × C + 3.4 × N

(Eq 7)

Niobium requires a greater weight percentage: Nb = 7.7 × C + 6.6 × N

(Eq 8)

Fig. 10

Corrosion of niobium-stabilized 29% Cr plus 4% Mo alloys in ASTM A 763 Y test. Source: Ref 11

As titanium and niobium are added to alloys, their corrosion resistance is improved (Figs. 9 and 10) (Ref 11). Maximum improvement in corrosion resistance levels off once full stabilization is reached. Excess amounts of the stabilizing elements have negligible effect, but titanium-stabilized alloys have a lower rate of corrosion than niobium-stabilized alloys. This is probably due to titanium’s ability to eliminate sulfur and oxygen from solution. Toughness improves for niobium-stabilized alloys up through full stabilization and then begins to decline. This is a result of excess stabilizing alloy acting as a solid solution hardener and therefore a toughness reducer. This toughness reduction is more pronounced with titanium, which is a stronger solid solution hardener (Figs. 11 and 12) (Ref 11). The upshot of this understanding was the introduction of dual stabilization, through which both weld and base metal toughness and corrosion resistance are optimized. The same study recommended that dual stabilization follow the following formula:

Chapter 8: Ferritic Stainless Steels / 119

(Ti + Nb) ≥ 6 × (C + N)

Fig. 11

Charpy V-notch impact ductile to brittle transition temperature (DBTT) of titanium-stabilized 29%Cr plus 4%Mo alloys test. Source: Ref 11

Fig. 12

Charpy V-notch impact ductile to brittle transition temperature (DBTT) of niobium-stabilized 29%Cr plus 4%Mo alloys test. Source: Ref 11

Fig. 13

(Eq 9)

The toughness of these alloys has a broad optimum that takes advantage of the corrosion-resisting benefits of titanium (Fig. 13) (Allegheny Ludlum). Other strong carbide formers such as zirconium and vanadium are ineffective stabilizers because their mobility at the temperatures at which they are thermodynamically capable of forming sufficiently large percentages of carbides and nitrides is too low to rid the matrix of these elements. They also have too great a tendency to form intermetallic compounds. Toughness in ferritic stainless steels is a major consideration. If ferritic alloys enjoyed the same toughness as austenitic alloys, there would be few instances when the use of the much more expensive nickel-bearing grades would be justified. Because stabilized alloys are ferritic at all temperatures, there is no automatic grain-refining transformation as exists in carbon steel. If grains grow large from annealing at high temperatures or welding, then the transition temperature increases. Section size also has an effect. Stabilized ferritic stainless steels are seldom used in thicknesses of over several millimeters because of decreasing toughness. Figure 14 shows how transition

Toughness of dual-stabilized low-alloy ferritic stainless. AL 466 is recognized as S40930

120 / Stainless Steels for Design Engineers

tion. But, the anisotropy does result in remarkable drawing characteristics, with ferritic stainless steels with elongations in tensile tests in the mid-30% range being nearly equal to austenitic stainless steels with over 50% elongation. The measure of anisotropy is the Lankford ratio. It is expressed as: R=

Fig. 14

Change in transition temperature with thickness for 29Cr-4Mo-2Ni alloy. Source: Ref 12

temperature can increase with thickness (Ref 12). This effect is due simply to stress states transitioning from biaxial to the more embrittling triaxial with increasing thickness.

Texture and Anisotropy The deformation of ferritic bcc materials is characterized by limited slip systems, high stacking fault energy, and lattice anisotropy. So, when ferritic stainless are deformed, dislocations tend not to dissociate as they do in austenitic stainless steels. The lack of dissociation of dislocations encourages cross slip. This minimizes dislocation tangles and work hardening. When ferritic stainless steels are deformed, certain crystallographic slip systems predominate, so that large deformations mechanically bring different grains via rotation into closer crystallographic alignment. This preferred deformation along easier slip planes results macroscopically in overall mechanical properties varying with direction with respect to the prior deformation. Thus, ferritic stainless steels, like low-carbon steels, have pronounced mechanical anisotropy. This is manifest in their deep drawing characteristics. Heavily cold-rolled and annealed ferritic stainless steels draw quite well. They resist thinning. When elongated, they contract in the width direction while keeping virtually the same thickness. This same phenomenon means that they cannot be stretch formed since plain strain quickly results in fracture because of the resistance to deformation in the thickness direc-

r0 + 2r45 + r90 4

(Eq 10)

When this expression equals 1, then a material is isotropic. As the value increases from 1, the drawability increases, as measured by the limiting drawing ratio (LDR), the ratio of the diameter of a disk to that of the deepest cylinder into which it can be drawn. The ferritic stainless steels in sheet form have LDRs of around 2.2 compared to 2.0 for 304. The good formability of ferritic stainless steels has some drawbacks. They are subject to ridging, which is the formation of visible ridges parallel to the direction of elongation. This is an artifact of texture in the material. A combination of careful chemistry design and thermomechanical processing is required to keep it under control. The approach centers on variables that increase stored energy from deformation to promote recrystallization over recovery during annealing. The ferritic stainless steels even carry forward some of the preferred grain orientation that come from initial solidification when growth of dendrites is along preferred crystallographic directions. Hot working merely reorients these similarly oriented grains en masse. Without phase changes or enough stored energy to provoke full recrystallization, randomness of grain orientation is never achieved. Titanium-stabilized steels show more texture and recovery versus recrystallization than do niobium-stabilized alloys. This is because titanium carbides and nitrides form at higher temperature and are therefore coarser. They thus present less obstruction to dislocation motion than finer niobium precipitates. Furthermore, niobium precipitates tend to dissociate to a greater degree than those of titanium. This puts niobium in solution during hot working where it can interact with dislocations. Thus, alloys at least partially stabilized with niobium can achieve greater recrystallization, which can translate to finer grain size and less anisotropy.

Chapter 8: Ferritic Stainless Steels / 121

Boron additions to ferritic stainless steels result in the formation of grain boundary carbides, M23(C, B)6. If added to titanium-stabilized steels, the carbides form on preexisting TiN particles and result in coarser overall precipitate arrays since finer, lower-temperature precipitating TiC or TiCS precipitates are at least partially precluded. The net result is coarser grain size and no major improvement in mechanical properties over the use of titanium alone (Ref 13). Additions of boron to niobium-stabilized steels does cause finer precipitates and grain size than would niobium alone (Ref 14).

High-Temperature Properties High-temperature mechanical properties of ferritic stainless steels are often important to their successful use because their oxidation resistance is excellent and better than austenitics, but their high-temperature strength is lower than that of austenitics. This has led to considerable development of high-temperature properties, primarily for the automotive market. Research has determined that high-temperature strength and creep resistance are best served by stabilizing grain size and having niobium in solid solution. Adding titanium to niobium-stabilized steels stabilizes the type of carbide, especially preventing the formation of the coarse M6C, whose growth decreases strength. The relatively high insolubility of TiC causes this. Niobium is concurrently made available for high-temperature solid solution strengthening.

Corrosion and Oxidation Resistance Corrosion resistance is chemistry dependent rather than structure dependent, so ferritic stainless steels behave just as do other stainless steels of the same crucial alloy content. The main alloying elements that provide resistance to localized corrosion, general corrosion, and crevice corrosion are chromium, molybdenum, and nitrogen. Since nitrogen is essentially insoluble in ferrite, it cannot contribute to the corrosion resistance of ferritic stainless steels as it can in austenite. Other alloying elements, such as copper and nickel, can add to corrosion resistance in special cases, but they are of secondary importance compared to

chromium and molybdenum. Likewise, other elements can have a negative effect. Any element that can combine with chromium or molybdenum can detract from corrosion resistance by their removal of these essential elements from solution. The most notorious of these is carbon, whose tendency to form chromium carbides causes areas around such carbides to be partially depleted of chromium. However, nitrogen, oxygen, and sulfur can also form chromium compounds and cause localized loss of corrosion resistance. Manganese sulfides, for instance, are almost always seen to be the locus of pitting corrosion (Ref 15). More careful examination has shown that such sulfides grow in the solid state as chromium/manganese sulfides and deplete their very close surroundings of chromium, inviting corrosion to begin at the inclusion-matrix interface, where chromium levels in solution are reduced (Ref 16). Other factors that lead to loss of localized corrosion resistance are the formation of chromium-rich phases such as α' and σ. Either of these with about 50% chromium will cause adjoining ferrite to have lower chromium levels. Because ferrite has a non-close-packed structure, diffusion rates, both substitutional and interstitial, are about two orders of magnitude higher than in austenite. That means that any deleterious chromium-depleting reaction can happen more rapidly. Alloys cannot be quenched rapidly enough to forestall sensitization, the precipitation of chromium carbides that depletes grain boundary regions of chromium. Instead, carbon must be neutralized by stabilization, or the chromium depletion must be removed by homogenization in longbox anneals. Note that the latter technique is also possible in austenitics but would require annealing for excessively long times, 102 h or so. The ferritic stainless steels are valued for their resistance to SCC. Even in environments that cause pitting, the normal initiation step for SCC, annealed ferritic stainless steels do not undergo SCC as long as alloying elements such as nickel, copper, and cobalt are kept below 0.5% in aggregate. Cold work sufficient to raise their hardness above Rc 20 to 22 can make them susceptible to both SCC and its cousin, hydrogen embrittlement. The more highly alloyed superferritic alloys are even susceptible to hydrogen embrittlement in the annealed condition (Ref 17). As with martensitic stainless steels,

122 / Stainless Steels for Design Engineers

this susceptibility is a maximum near room temperature and declines with increasing temperature, as opposed to austenitics, which see their maximum susceptibility above room temperature. This limits these alloys’ ability to employ cathodic protection safely to –0.80 Vsce, at which point corrosion in seawater is, if not eliminated, reduced to very low levels (Ref 18).

REFERENCES

1. F.M. Beckett and R. Franks, Trans AIME, Vol 113, 1934, p 126–143 2. Stainless Steel, Les Editions de Physiques, 1992, p 483 3. Stainless Steels, Les Editions de Physique, 2003 4. H.D. Newell, High Chromium Irons, Met. Prog., April 1947, p 617–626 5. P.J. Grobner, The 885 °C (475 °C) Embrittlement of Ferritic Stainless Steels, Metall. Trans., Vol 4, 1973, p 251–260 6. Handbook of Stainless Steels, Peckner and Bernstein, McGraw Hill, 1977, p 5–9, 5–12 7. H.W. Paxton, Alloying, ASM, 1998, p 213 8. H. Abo et al., Stainless Steel ‘77 9. J. Grubb and R. Wright, The Role of C and N in the Brittle Fracture of Fe-26Cr, Met. Trans. A, Vol 10A, Sept 1979, p 1247–1255

10. J. Grubb, R. Wright, and P. Farrar, “Micromechanisms of Brittle Fracture in TitaniumStabilized Stainless Steels,” Special Publication 706, ASTM, 1980 11. J. Grubb, Stabilization of High-Chromium Ferritic Stainless Steels, Proc. Int. Conf. Stainless Steels, ISIJ, Chiba, 1991 12. M.A. Streicher, Stainless Steel ‘77, p 27 13. E. El-Kashif, K. Asakura, T. Koseki, and K. Shibata, ISIJ Int., Vol 44, 2004, p 1568–1575 14. N. Fujita, K. Ohmura, E. Sato, and A. Yamamoto, Nippon Technical Report 71, Oct 1996 15. T. Suter, E. Webb, H. Bohni, and R. Alkire, Pit Initiation in I M NaCl With and Without Mechanical Stress, J. Electrochem. Soc., Vol 148 (No. 5), 2001, B174 16. M. Ryan, D. Williams, R. Chater, B. Hutton, and D. McPhail, Why Stainless Steel Corrodes, Nature, Vol 412, 2002, p 770 17. J. Grubb, “Hydrogen Embrittlement of Superferritic Stainless Steels,” paper presented at 1984 ASM Int’l Conference on New Developments in Stainless Steel Technology, Detroit, September 1984 18. J. Grubb and J. Maurer, “Use of Cathodic Protection With Superferritic Stainless Steels in Seawater,” paper presented at Corrosion 84, New Orleans, April 1984

Stainless Steels for Design Engineers Michael F. McGuire, p 123-135 DOI: 10.1361/ssde2008p123

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 9

Martensitic Stainless Steels Summary THE SMALLEST CATEGORY of stainless steels in usage volume is the martensitic stainless steels. This is mainly because these alloys are limited in corrosion resistance because of the necessity of keeping alloy levels low to produce the martensite structure. Even so, they fill an important niche as a strong, hard, and tough alloy of fairly good corrosion resistance and as a strong, stable, high-temperature alloy.

Introduction Nearly 100 years ago cutlery was first sold in Great Britain with a composition of 13% chromium and 0.25% carbon. This was the first commercial use of stainless steel and cutlery with the same basic analysis is still sold today. The useful alloys of martensitic stainless steel contain from roughly 11 to 18% chromium and up to 1.0% carbon. Relatively small amounts of nickel, molybdenum, tungsten, vanadium, and niobium are also added at times for specific purposes explained in this chapter. Those martensitic stainless steels in which elements such as copper and titanium are added to produce additional hardening through precipitation are discussed in Chapter 4, “Corrosion Types.” The designers and engineers already familiar with martensitic carbon and alloy steels will find nothing confusing about martensitic stainless steels. There is no aspect of martensitic steels that does not apply directly to stainless martensitic steels. The additional concerns one must have with stainless martensite relate mainly to those that are due to the strong ferritizing influence of chromium. Chromium strongly promotes the formation of ferrite,

which restricts the temperature and composition ranges over which it is possible to obtain a fully austenitic structure from which to form martensite. The presence of ferrite in a martensitic structure is detrimental to strength, hardness, and toughness. Ferrite can appear in the as-cast structure and be formed during austenitizing or tempering. All the usual concerns inherent in any martensitic alloys are still present; temper embrittlement, retained austenite, etc. Martensitic stainless steels are the most marginally corrosion resistant of all the stainless alloys. The requirement that they be fully austenitizable limits the amount of corrosion-resisting chromium and molybdenum they can contain. Much of the carbon in them detracts from the effective chromium content by forming chromium carbides. In addition they are always susceptible to stress corrosion cracking (SCC) when their hardness exceeds about Rc 22. These limitations combine to make their excellent properties usable in only mild environments compared to other stainless steels. Their high strength and hardness for their relatively low cost ensure their place as a very useful engineering material. Table 1 lists the most significant of the martensitic stainless steel alloys. The reader should be aware that some alloys which are quite similar are discussed primarily in other chapters dealing with specifically PH stainless steels or primarily ferritic stainless steels. The distinction between martensitic stainless steels and some other stainless alloy families is sometimes vague. Nearly all the precipitation-hardening stainless steels are used in the martensitic state, but their special hardening mechanism of precipitation within a martensitic matrix causes them to be categorized separately somewhat arbitrarily. By this conventional logic, some of the martensitic alloys containing molybdenum or

S41400 S41425

S41500 S41600 S41623

S41800 Wrought S42000 Wrought DIN Wrought 1.4116 Nominal S42020 Wrought S42023 Wrought S42200 Wrought S42400 Wrought Wrought S42500 ... Wrought ... Wrought

414 414 mod

415 416 416Se

418 420 4116

NT-CRS

HP13Cr-2

HP13Cr-1

JFE Nominal JFE Nominal Nippon Nominal

S41003

412

420F 420FSe 422 424 425 425mod Trinamet

S41040

410Cb

Wrought

Wrought

Wrought

Wrought Wrought Wrought

Wrought Wrought

Wrought

Wrought

Wrought

S41008

410

Form

Wrought Wrought Wrought

S40300 S41000 S41003

UNS

0.03

0.025

0.025

0.15 min 0.15 min 0.20-0.25 0.06 max 0.08-0.20 0.50-0.55 0.30 max

0.15-0.20 0.15 min 0.5

0.05 max 0.15 max 0.15 max

0.15 max 0.05

0.030 max

0.18 max

0.08

0.15 max 0.15 max 0.03

C

Mn

...

1.45

0.45

0.45

1.25 1.25 1 0.50-1.00 1 1 1

0.5 1

0.50-1.00 1.25 1.25

1 0.5-1.0

1.5

1

1.5

1 1 1

Compositions (wt%) of martensitic stainless steels

403 410 410S

Alloy

Table 1 S

...

...

...

...

...

...

Si

...

...

...

1 1 0.75 0.30-0.60 1 1 1

0.5 1

0.6 1 1

1 0.6

1

1

1

0.5 1

(continued)

0.15 min 0.06 0.03 0.03 0.01 0.03 0.03

0.03 0.03

0.03 0.15-0.30 0.06

0.03 0.005

0.03

0.03

0.03

0.03 0.03 0.03

Cr

12.7

13

12.0-14.0 12.0-14.0 11.0-13.5 12.0-14.0 14.0-16.0 13.0-14.0 12.0 -14 13

11.5-13.5 11.5-13.5 10.512.5 11.513.5 11.513.5 10.512.5 11.5-13.5 12.015 11.5-14.0 12.0-14.0 12.014 12.0-14.0 12.0-14.0 14.5 ...

...

1.4

2

0.6 0.6 0.75-1.25 0.30-0.70 0.30-0.70 0.80-1.20 1.00 -3 1

... ... 0.65

... 1.52 0.50-1.00 ... ...

0.6

Mo

... ... ...

...

...

4.5

5

4

... ... 0.50-1.00 3.50-4.50 1.00-2.00 0.5 ...

1.80-2.20 ... ...

1.25-2.50 4.07 3.50-5.50 ... ...

1.5

1.5

Ni

... ... 0.03 ...

1.5 Cu

...

... 0.15 min Se 0.75-1.25 W ... ... ... 2.00 -3.00 Cu ...

0.05-0.30 Nb 0.030 max N ... 0.06-0.12 N ... ... 0.15 min Se 2.50-3.50 W ... ...

N

... ...

Other

0.040 N

...

...

... ... 0.15-0.30V ... ... ... ...

... ... 0.15 V

... ... ...

... 0.30 Cu

...

...

...

... ... ...

Other

124 / Stainless Steels for Design Engineers

KL-HP 12Cr 431 440A 440B 440C 440F 440FSe BG-42 ATS-34 14-4 CrMo 154 CM CPM S30V CPM S60V CPM S90V CA-15 CA 15M CA-40 CA-40F CB-6N CB-6MN CA-28MVW

KL-12Cr

NT-CRSS

Alloy

PM

PM

Nominal

Nominal

J91150 J91151 J91153 J91154 J91650 J91540 J91422

Wrought PM

Nominal Nominal

Cast Cast Cast Cast Cast Cast Cast

0.15 max 0.15 max 0.20-0.40 0.20-0.40 0.06 max 0.06 max 0.20-0.28

2.2

2.15

1.05 1.45

0.01

0.20 max 0.60-0.75 0.75-0.95 0.95-1.20 0.95-1.20 0.95-1.20 1.15 1.05 1.05

Wrought

Wrought Wrought Wrought Wrought Wrought Wrought Wrought Wrought Wrought

C

0.01

0.02

Wrought

Form

Wrought

Nippon Nominal JFE Nominal JFE Nominal S43100 S44002 S44003 S44004 S44020 S44023 Nominal Nominal Nominal

UNS

...

...

Mn

... 1 1 1 1 0.5 1 0.50-1.00

0.4

0.45 ...

1 1 1 1 1.25 1.25 ... 0.4 0.5

2

Table 1 (continued) Compositions (wt%) of martensitic stainless steels

...

...

S

0.04 0.04 0.04 0.020-0.040 0.02 0.03 0.03

...

...

... ...

0.03 0.03 0.03 0.03 0.10-0.35 0.03 ... ... ...

...

Si

...

...

...

1.5 0.65 1.5 1.5 1 1 1

0.3

1 1 1 1 1 1 0.3 0.35 0.3

...

...

...

Cr

11.5-14.0 11.5-14.0 11.5-14.0 11.5-14.0 10.50-12.50 11.50-14.00 11.00-12.50

13

17

14 14

15.0-17.0 16.0-18.0 16.0-18.0 16.0-18.0 16.0-18.0 16.0-18.0 14.5 14 14

12

11

12.3 ...

Mo

0.40-1.00 0.90-1.25

0.5 0.15-1.00 0.5 0.5

1

0.4

4 2

... 0.75 0.75 0.75 0.40-0.60 0.6 4 4 4

2

2

Ni

1 1 1 1 1 3.50-4.50 0.50-1.00

...

...

... ...

1.25-2.50 ... ... ... 0.75 0.75 ... ... ...

5.5

2.4

5.8

Other

... ... ... ... ... ... 0.90-1.25 W

9.0 V

5.5 V

... 4.0 V

... ... ... ... ... 0.15 min Se 1.2 V ... ...

...

0.5 Cu

1.5 Cu

Other

... ... ... ... ... ... 0.20-0.30 V

...

...

... ...

... ... ...

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

0.010 N

0.010 N

0.015 N

Chapter 9: Martensitic Stainless Steels / 125

126 / Stainless Steels for Design Engineers

tungsten should also be considered precipitationhardening alloys, but they customarily are not and will not be in this work. The ferritic alloys often have compositions that allow them to be partially martensitic under some conditions. 430 (UNS S43000) and 3CR12 (UNS S41003) can contain some martensite if their heat treatment is such that austenite is allowed to form and is followed by rapid cooling. Even 409 (UNS S409XX) can form some austenite if chromium is at the high end of its possible range and nickel and manganese residual levels are high. The martensitic alloys themselves can be made to be partially ferritic by forcing their carbon contents to low levels as is customarily done with 410S (UNS S41003). Not understanding these alloys can lead to unexpected consequences in mechanical properties or corrosion performance.

on temperature. The amount is given by the Koistinen and Marburger equation (Ref 1): 1− Vα ′ = exp{β (Ms − T )}

(Eq 1)

The martensite is coherent with the parent austenite and resembles the passage of slip dislocations through the crystal. The sum of many such dislocations is shear, and this can be macroscopically visible as in Fig. 1. The formation of martensite is essentially mechanical (i.e., via deformation, not diffusion). The shear and volume expansion, about 4%, which accompanies the transformation, involves a great deal of strain energy that must be taken into account. This is shown diagrammatically in Fig. 2 (Ref 2).

Martensite Formation Martensite as a phenomenon deserves a brief review. Martensite forms as result of the diffusionless transformation of austenite. The austenite may be supersaturated with carbon or nitrogen, but that is not necessary for the transformation. The driving force for the transformation is simply the much lower free energy of the ferrite phase over the austenite phase, which can be attributed largely to large mutual repulsion between iron atoms that possess unpaired outer electrons with the same quantum number and magnetic polarity. This free-energy differential increases with decreasing temperature. At a certain temperature, the martensite start temperature Ms, the transformation occurs spontaneously via the coordinated movement of atoms in a shearing-type mode at very high speeds approaching the speed of sound in the material. The composition of the martensite is identical to that of the parent austenite. There is regularity to the relationship between the parent austenite and the martensite. Greninger and Troiano determined that the close-packed planes of the austenite {111} varied from the {011} of the martensite by only 0.2°. Further, the direction of the of the austenite was only 2.7° from the of the martensite. These relationships define the habit plane that constitutes the austenite martensite boundary. Martensite forms essentially independent of time and the fraction transformed depends only

Fig. 1

Martensite platelets emerging from the surface. Source: Ref 2

Fig. 2

The martensite reaction ab contrasted to the nucleation and growth-type transformation of austenite to ferrite, ac

Chapter 9: Martensitic Stainless Steels / 127

This energy differential between ferrite and martensite is stored in the high-strain energy matrix. Applied strains affect the transformation. Indeed, metastable austenite can readily be transformed to martensite by deformation. However, the untransformed austenite is hindered from transforming by the compression it receives from the already-formed martensite. Thus, some residual austenite is commonly found between lathes of martensite. At the Ms temperature, the body-centered cubic (bcc) phase becomes preferable energetically, but this temperature is too low for diffusion transformation, and a slight shear in the austenite lattice causes a rearrangement of the atoms from a face-centered cubic (fcc) to a distorted bcc structure. The amount of distortion is proportional to the amount of carbon in the interstices of the structure. These interstices are considerably smaller in the bcc structure even though it is expanded from the fcc. The octahedral sites change from 2.86 by 3.56 A to 2.86 by 2.86 A, as shown in Fig. 3 (Ref 3). The distortion is accommodated by accommodation from site to site at low carbon levels, but above about 0.018% carbon this can no longer be accommodated and a tetragonal distortion occurs (Ref 3). The carbon is in a state of supersaturation in the as-formed martensite. When the martensite is tempered, the carbon diffuses from these inter-

Fig. 3

Change in size of the octahedral interstitial site with the change from face-centered cubic (fcc) to bodycentered cubic (bcc). Source: Ref 2

stitial sites and forms various carbides, leaving the parent martensite less strained, softer, and tougher. Figure 4 shows that the large strain energy in martensite varies with the carbon content, and Fig. 5 (Ref 4) shows how hardness varies with carbon content. Nitrogen behaves similarly to carbon in both austenite and martensite, but its solubility is lower, and it is less significant as an alloying element accordingly. Hydrogen and boron, as interstitials, also raise hardness.

Phase Structure Figure 6(a) to (h) shows a series of photomicrographs of various martensitic alloys (Ref 5). A stainless martensitic alloy should have the following characteristics: • It must have at least 10.5% chromium to qualify as stainless and even more for better corrosion resistance. • It should be fully austenitic at some temperature. • The temperature at which austenite forms on heating should be sufficiently high to permit tempering above the temper embrittlement range. These criteria are somewhat challenging. Figure 7(a) shows that at low-carbon (0.05%) levels austenite is stable up to about 12% chromium, above which some δ-ferrite tends to be stable at all temperatures below the melting point. Increasing carbon slightly expands the chromium level at which full austenitization can occur (Fig. 7b) (Ref 3).

Fig. 4

Strain energy of martensite dependence on carbon content. Source: Ref 2

128 / Stainless Steels for Design Engineers

Fig. 5

Variation in martensite hardness with carbon content

The interplay between chromium and carbon is further explained in Fig. 8(a) and (b), in which it becomes clear that for higherchromium alloys the range over which full austenitization can occur is further restricted.

The variety of martensitic stainless steels would be very limited if only chromium and carbon were available as alloying elements, but fortunately nickel again can make an important contribution. Nickel greatly expands the

Chapter 9: Martensitic Stainless Steels / 129

Fig. 6

(a) Annealed 410 showing carbides within an equiaxed ferrite matrix. (b) 410 quenched and tempered. (c) 416 quenched and tempered: white ferrite and gray sulfides in a martensite matrix. (d) 420 quenched and tempered showing fine carbides in a martensite matrix. (e) 420 quenched and tempered showing surface decarburization. (f) 440A annealed displaying primary and smaller secondary carbides in a ferrite matrix. (g) 440B quenched and tempered displaying both primary and secondary carbides. (h) 440C quenched and tempered displaying significant primary carbides plus finer secondary carbides in a martensite matrix

130 / Stainless Steels for Design Engineers

Fig. 7

Iron-chromium phase diagrams at two low-carbon levels

1800

1800

Liquid

1400 α+γ

Liquid

1600

Temperature, °C

Temperature, °C

1600

γ

1200

1000

L+α

1400

σ

γ

α+γ

1200

γ + carbide 1000

γ + carbide 800

800 α + carbide

α + carbide 600

600 0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Mass, %C

Fig. 8

0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Mass, %C

(a) Iron-chromium phase diagram at 12% chromium; (b) iron-chromium diagram at 17% chromium

chromium levels and temperatures at which austenite is stable as is shown in Fig. 9 (Ref 3). Table 2 quantifies the influences of the various possible alloying elements on the key properties of martensitic stainless steels. It can be seen that the elements that promote austenite, with the exception of cobalt, all depress the Ms temperature. This puts a limit on the

amount of total alloy that can be used and in the end puts an upper limit on the ability of martensitic stainless steels to achieve high corrosion resistance. This is because as the main corrosion fighters, chromium and molybdenum, which are ferritizers, are increased, so must austenitizers such as nickel. The coordinated increase in these elements lowers the Ms to such a degree that the

Chapter 9: Martensitic Stainless Steels / 131

alloys become stably austenitic before much higher corrosion resistance is obtained. The ability to temper without austenite reversion is an important trait. Obviously, if transforming martensite to austenite during tempering caused subsequent untempered martensite or other undesirable phases, this would limit one’s ability to temper at a high enough temperature to achieve desired toughness. This limits the use of nickel while encouraging the use of elements like molybdenum. Copper has become an important alloying element in martensitic stainless steels because it greatly improves corrosion resistance in certain environments without diminishing an alloy’s ability to be tempered.

Thermal Processing The main concerns with processing martensitic stainless steels are austenitizing, quenching, tempering/stress relieving, and annealing.

Fig. 9 Table 2

The expansion of the range of austenite stability with nickel content

Austenitizing is complicated in martensitic stainless steels because many grades contain carbon at levels intended to produce carbides for wear resistance purposes. Since carbon solubility varies strongly with temperature at austenitizing temperatures (Fig. 7a and b), control of temperature is vital to have the correct balance of carbon in solution versus carbon as carbide since carbon in solution has such a strong influence on ferrite content, Ms, and mechanical properties. Austenitizing temperature also determines austenite grain size. This affects Ms, but more importantly it influences subsequent toughness. Phosphorus precipitates at prior austenite grain boundaries during tempering with a maximum effect at 475 °C. This is the infamous temper embrittlement. Figure 10 (Ref 3) shows the significant toughness change that occurs as increasing austenitizing temperature increases austenite grain size and permits greater phosphorus concentrations at grain boundaries. Refining phosphorus from any chromiumcontaining steel is quite challenging thermodynamically, so achieving low phosphorus levels depends mainly on restrictions on raw materials for melting. Because this is difficult or costly, grain size control is the main tool for controlling temper embrittlement. The higher-carbon grades, those above 0.20% carbon, should be heated gradually through stage heating to avoid cracking due to thermal stresses. Soaking at 800 °C until uniform temperature is achieved minimizes this risk. Another concern during austenitizing is superficial carbon loss, an example of which is shown in Fig. 6(e). Heating in air to 1050 °C can cause surface carbon to decrease by approximately 0.10% per hour, resulting in much lower surface hardness. This loss increases with base carbon level and austenitizing temperature. Carbon or nitrogen pickup could also occur if the atmosphere was rich in these elements. The carbon potential of the furnace atmosphere must be controlled to avoid potentially serious problems. If hydrogen atmospheres are used the

Influence of alloying elements on ferrite, Ms, and austenite start

Element

Lowering of % ferrite per % element Lowering of MS per % element Change of AC per % element

N

C

Ni

Co

Cu

Mn

Si

Mo

Cr

V

Al

–220

–210

–20

–7

–7

–6

6

5

14

18

54

–475

–475

–17

0 to 10

–17

–30

–11

–21

–17

–46

...

0

0

0 to 280 0 to 250 –30 to –115

–25 to –66 25 to 73 25 to 60 0 to 35 50 to 290 30 to 750

132 / Stainless Steels for Design Engineers

+120

P: 0.047% [ ]

P: 0.035%

+80 [ ] Transition temperature, °C

P: 0.021% +40 P: 0.003-0.004% [ ] 0

[ ]

[ ] [ ]

−40

−80 5

10

20

50

100

200

Austenite grain size, μm

Fig. 10

Influence of austenite grain size and phosphorus level on toughness

danger of embrittlement after quenching must be recognized. Stress relief without delay would be mandatory. The high chromium content of these alloys renders them very deep hardening. Air hardening is generally sufficient. Oil quenching which is faster may be slowed by heating the oil. Avoiding quench cracking and excessive warpage is almost always a greater concern than depth of hardening so air quenching is standard. Because the quenching and the transformation it causes are inevitably accompanied by residual stresses in a brittle material, stress relieving should be immediate to avoid cracking. Higher-carbon grades should not even be allowed below room temperature before stress relief. Pickling should never be done on asquenched material because this could easily result in hydrogen uptake and delayed cracking by hydrogen embrittlement. Heating as-quenched material to between 150 and 400 °C produces stress relieving. Besides the normal flow on a microscopic scale, which we understand as stress relieving, there is a slight growth in the number of fine cementite particles and a corresponding decrease in the amount of carbon in solid solution. This results in a slight decrease in hardness. At 400 °C, a further precipitation of M2X and M7C3 as well

as the transformation of M3C into M7C3 can result in a secondary hardening, a true precipitation-hardening effect. In the presence of strong carbide-forming alloying elements such as molybdenum, vanadium, and tungsten, the M2X carbide can become the more stable species and be responsible for the secondary hardening. At 500 °C, coarser M23C6 and M7C3 begin to grow at grain boundaries. This is accompanied by a pronounced softening. The hardening reduction with stress relief and tempering for a 12% Cr alloy is shown in Fig. 11 (Ref 6). Separately at the 475 °C range, the previously mentioned phosphorus segregation to prior austenitic grain boundaries occurs. This effect begins to disappear above 550 °C. Thus true tempering is conducted above this temperature. The microstructural changes at these temperatures are the above-mentioned loss of carbon from solid solution, carbide precipitation and coarsening, and, of course, stress relief. The result is a pronounced softening and toughening. If the material contains retained austenite, it may decompose to ferrite and carbide with a negative effect on toughness. The molybdenum, vanadium, and tungstenalloyed grades will resist softening during tempering because of the strength of the secondary hardening they undergo due to precipitation

Chapter 9: Martensitic Stainless Steels / 133

500 Initial hardness 300°C

450 300°C

350°C 350°C 400°C

400

400°C 450°C

450°C 500°C 550°C

Hardness

350

600°C 650°C 700°C

300

750°C

500°C 550°C

250

600°C 650°C 200

700°C 750°C

150 11

12

13

14

15

16

17

18

19

20

21

22

23

T (20 + LOG t) × 10−3

Fig. 11

Influence of tempering on hardness

hardening of carbides and nitrides. Nickel seems to amplify this action by its influence on diminishing the solubility of carbon in the matrix. Thus, the tempering of the higher-alloy martensitic stainless steels can truly be considered a precipitation-hardening reaction. The higher-carbon, higher-chromium grades are typically only stress relieved because the removal of chromium from solution by carbide formation at higher temperatures causes an unacceptable loss of corrosion resistance.

Applications High-Temperature Use. The basic 12 % Cr martensitic alloy has been the basis of alloying improvements that were done to produce better high-temperature performance, especially for turbines. The addition of vanadium and niobium, both of which form much more stable carbides than chromium, results in alloys that have vastly improved creep resistance in the 550 °C range, as shown in Fig. 12 (Ref 3). Tool and Cutlery Alloys. A high-profile use of martensitic stainless steels is in cutlery. Hunting knives, sport knives, and chefs’ cutting tools are highly valued items and contain some of the

most sophisticated martensitic stainless alloys. While 420 is the common alloy and is quite serviceable, much more wear- and corrosion-resistant alloys exist. At one time, 440C was the maximum step up from 420; however, further alloying with molybdenum for corrosion resistance and vanadium for hardness of the carbide phase has led to improvements. The wear resistance of a blade is largely determined by the hardness and amount of carbides while the toughness is governed by the matrix properties. These alloys are used at very high hardness levels, so cleanliness is very important to toughness, which measures the ability to withstand chipping in use. Electroslag remelting (ESR) or vacuum induction melting-vacuum arc remelting (VIM-VAR) provides the cleanliness required, while powder metallurgy is optimal for obtaining very fine carbide size and uniformity. The nominal analyses of some prominent grades are shown in Table 3. The martensitic alloys have a tendency toward centerline segregation during solidification as well as toward the formation of primary carbides. This has produced limitations in the amount of highly wear-resistant constituents such as vanadium carbide (hardness Rc 75), which can be introduced into the matrix in conventional production. Powder metal techniques

134 / Stainless Steels for Design Engineers

4 50 2 45

Applied stress, kg/mm2

40 35 3 30 1 25

20 1 (0.2C-10.5Cr) 2 (0.2C-10.5Cr-0.1Nb) 3 (0.2C-10.5Cr-0.1V) 4 (0.2C-10.5Cr-0.1V-0.1Nb)

16 3

10

30

100

300

1000

3000

10,000

Rupture life, h

Fig. 12 Table 3

Influence of vanadium and niobium on high-temperature properties

Tool and cutlery martensitic stainless steels alloy compositions

Alloy

UNS

420 4116

S42000 DIN 1.4116 Nominal S44002 S44004 Nominal Nominal Nominal

Wrought Wrought

0.15 min 0.5

1

Wrought Wrought Wrought Wrought Wrought

0.60-0.75 0.95-1.20 1.15 1.05 1.05

1 1

Nominal Nominal

Wrought PM

Nominal Nominal

440A 440C BG-42 ATS-34 14-4 CrMo 154 CM CPM S30V CPM S60V CPM S90V

Form

C

Mn

Cr

Mo

Ni

Other

Other

...

12.0-14.0 14.5

... 0.65

... ...

... ...

... 0.15 V

... 0.4 0.5

0.03 0.03 ... ... ...

1 1 0.3 0.35 0.3

16.0-18.0 16.0-18.0 14.5 14 14

0.75 0.75 4 4 4

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

... ... 1.2 V ... ...

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

1.05 1.45

0.45 ...

... ...

0.3 ...

14 14

4 2

... ...

... 4.0 V

... ...

PM

2.15

0.4

...

...

17

0.4

...

5.5 V

...

PM

2.2

...

...

13

1

...

9.0 V

...

...

...

S

Si

0.03 ...

1

are not subject to the same limitations as continuous casters and have alloys the production of alloys with high volume content of VC. One such alloy is Crucible CPM 90V with 14% Cr, 9% V, 1% Mo, and 2.3% C. This alloy has equal or better toughness and corrosion resistance as 440C but has ten times the wear resistance at the same macrohardness. Oil Country Tubular Good and Line Pipe. The need for corrosion resistance in oil production tubulars has grown as the quality of petroleum deposits has become less optimal. Use of stainless can eliminate for corrosion inhibitors in H2S and CO2 environments. This has led to the use of low-carbon martensitic stainless steels. Low carbon and nitrogen levels give

good toughness without tempering and minimize the loss of chromium to carbides, maintaining it in solution for corrosion resistance. The addition of nickel and molybdenum yields full austenite and martensite transformation and improves corrosion resistance. Table 4 lists several such alloys by JFE: the first two can be made to meet L80 specifications and produced as seamless. JFE reports production of over 100,000 tons per year of this product (Ref 7). The third alloy is a near match for the precipitation-hardening stainless Custom 450 (UNS S45000) (Ref 8). Like other precipitation-hardenable steels, it shows excellent resistance to SCC at high strength levels. Figures 13 and 14 show the improvements in corrosion resistance

Chapter 9: Martensitic Stainless Steels / 135

Table 4 Alloy

JFE Steel/Nippon Steel oil country tubular goods and line pipe alloys UNS

Form

C

Mn

S

Si

Cr

Mo

Ni

Other

HP13Cr-! JFE Nominal HP13Cr-2 JFE Nominal NT-CRS Nippon Nominal NT-CRSS Nippon Nominal KL-12Cr JFE Nominal KL-HP JFE 12Cr Nominal

Wrought

0.025

0.45

...

...

13

1

4

...

...

Wrought

0.025

0.45

...

...

13

2

5

...

...

Wrought

0.03

1.45

...

...

12.7

1.4

4.5

1.5 Cu

0.040 N

Wrought

0.02

2

...

...

12.3

2

5.8

1.5 Cu

0.015 N

Wrought

0.01

...

...

...

11

2.4

0.5 Cu

0.010 N

Wrought

0.01

...

...

...

12

5.5

...

0.010 N

Fig. 13

Corrosion rates of stainless versus carbon steel

... 2

Other

ronments to about Rc 22 to avoid SCC by the hydrogen embrittlement mechanism. The stainless can resist this failure mode at higher strengths. As specifying bodies such as the American Petroleum Institute (API) approve the use of stainless tubulars at higher strength levels than carbon steel tubulars are safely capable of handling, then the strength improvement, coupled with the orders of magnitude improvement in corrosion resistance, will cause a great increase in their use. Lower carbon levels permit the use of field welds without tempering so that similar alloys can be used for line pipe. These are corrosion resistant and yet meet X70 and X80 class specifications. These alloys are the last two in Table 2. These uses of martensitic stainless steels for oil production represent possibly the greatest growth area for any kind of stainless steel in the first decade of the 21st century. REFERENCES

Fig. 14

Corrosion rates for stainless oil country tubular goods (OCTG) alloys under severe operating conditions

over carbon steel L80 oil country tubular goods (OCTG) under test conditions representative of difficult real-use environments (Ref 7). The improvements in martensitic steels for these applications are hardly more than a thorough revisiting of the developments of the 1950s and 1960s. This does not diminish their importance. Carbon steels are limited in sour envi-

1. D.R. Koistinen, R.E. Marburger, “A General Equation Prescribing the Extent of the Austenite/Martensite Transformation in Pure Iron,” Acta Met, Vol 7, 1959, p 59 2. http://www.msm.cam.ac.uk/phase-trans/ 2002/martensite.html 3. Bletton, Aciers Inoxidables, Les Editions de Physique les Ulis, Paris, 1993, p 481 4. ASM Handbook Desk Edition, 1985, p 28–9 5. http://products.asminternational.org/mgo/ 6. K.J. Irvine et al., JISI, Vol 195, ISIJ International 1960, p 386–405 7. S. Deshimaru et al., “Steels for Production, Transportation and Storage of Energy, JFE Technical Report (No. 2), March 2004 p 55–67 8. M. Kimura et al., “High CR Stainless OCTG with High Strength and Superior Corrosion Resistance,” JFE Technical Report (No. 7), Jan 2006, p 7–13

Stainless Steels for Design Engineers Michael F. McGuire, p 137-146 DOI: 10.1361/ssde2008p137

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 10

Precipitation-Hardening Stainless Steels Summary

Introduction

THE PRECIPITATION-HARDENABLE (PH) grades are a highly specialized family of stainless steels whose existence derives from the need for very high-strength materials with good corrosion resistance. The workhorse alloys are the martensitic PH grades, which are used in many forms. Primarily used as forgings, bar, and other hot-worked forms, they can also be obtained in cold-rolled sheet and strip, although not with the flatness expected from non-PH stainless. The semiaustenitic alloys are more amenable to production as sheet, strip, and wire and are designed for applications that require extensive forming before hardening. The fully austenitic PH alloys fill a small niche where high mechanical properties are required at temperatures above or below which the other PH grades are found lacking, when a nonmagnetic material is required, or when the higher thermal expansive coefficient of an austenitic material is desired. In no case is corrosion resistance better than that of normal 304 found in PH stainless steels. If enhanced strength and very high corrosion resistance are required, then the designer should look to duplex stainless steels for the optimal material. If cost is a greater concern than corrosion resistance or toughness, then martensitic stainless steels should be considered for applications where strength and hardness over that of annealed ferritic and austenitic stainless is required. The increased use of titanium alloys and advanced composite materials may occur at the expense of the stainless PH alloys and at the same time may create some new niche applications for them.

The PH stainless steels exploit the low austenite stability possible in the chromium/nickel stainless steels by making the alloys so lean in composition that they can be made to transform nearly entirely to martensite by thermal or mechanical treatment. This martensite can then be further hardened by the coherent precipitation of intermetallic compounds, elemental copper, nitrides, or even phosphides. This precipitation hardening can also be made to occur in a fully austenitic matrix, and this also provides a commercial PH alloy. But, the martensitic PH grades are by far the more common. The border between the more highly alloyed martensitic stainless steels, which undergo secondary hardening during tempering, and the PH alloys is indeed vague. Some authors have astutely treated them as a single group. Here, we treat them separately because they are traditionally considered as separate alloys. The advantage of the PH alloys over the strictly martensitic stainless steels is that they attain great strength with higher toughness and corrosion resistance than can be obtained through the hardening of martensite through carbon. In addition, they can be fabricated in a relatively soft state and then hardened with very little dimensional change. The PH grades were developed at the beginning of World War II, with Stainless W (UNS S17600) by U.S. Steel generally acknowledged as the first. The later-developed grades are distinguished from the first by their more uniform, and therefore tougher, microstructure through the elimination of residual δ-ferrite and retained

138 / Stainless Steels for Design Engineers

austenite and by more astute alloy design and chemistry control. The mechanism of precipitation hardening is parallel to that used to strengthen aluminum alloys in which the precipitation of a coherent second phase from a supersaturated solid solution is produced by an aging heat treatment. The coherent precipitate strains the lattice and impedes the motion of dislocations, producing strengthening. Overaging causes the precipitates to lose coherency, and softening follows. The precipitate that causes the hardening is normally nickel (aluminum/titanium) (Ref 1). Figure 1 shows the compounds that can form from the precipitation of supersaturated aluminum and titanium in an iron alloy matrix. It is also possible to produce a hardening reaction by the precipitation of elemental copper (Ref 2). In nitrogen-bearing alloys, a hardening may be produced by the precipitation of Cr2N (Ref 3). The precipitation begins with the diffusion of the precipitating species to sites on the existing matrix. These enriched zones are called Guinier-Preston (GP) zones. Close dimensional matchup between the precipitating species and the parent matrix is required. The differential should be on the order of a percent. This allows not only coherency but also strain. The coherent precipitate is a effective barrier to dislocation

movement. As time and temperature of precipitation increase, the zones can grow to sizes that cannot accommodate the small size differential; coherency is lost, and with it the hardening effect diminishes. The precipitation has the dual function of stress relieving the martensite while further hardening the matrix through the precipitation of the coherent precipitate. The mechanical properties of the final microstructure depend on the initial strength of the matrix before aging, the amount of precipitate, and the coherency of the precipitate. The ideal microstructure for the initial matrix is 100% martensite. To the extent there is δ-ferrite or retained austenite, properties, especially yield strength and toughness in the transverse direction, are compromised. The aging temperatures can also be high enough that reversion of martensite to austenite occurs, which also lowers subsequent tensile properties. While the presence of persistent, large bands of either δ-ferrite or γ-austenite is undesirable, but both also have benefits. The presence of some fine bands of δ-ferrite promotes easier and more reproducible precipitation of chrome carbides at the δ/γ interface during the “austenite conditioning” or “trigger anneal” heat treatment step for semiaustenitic alloys (17-7, AM350, etc.). Although bands of stable austenite are undesirable, it is the presence of residual interlath

γ + Ni(AlTi) + Ni3AlTi

Ni(AlTi) Ni(AlTi) + Ni2AlTi Ni2AlTi

4

Ni2(AlTi) + Ni3AlTi

Aluminum, wt%

Ni2AlTi

γ + Ni2AlTi

γ + Ni(AlTi)

Ni3Ti σ

3

Limit of austenite ferrite region in solution treated conditions

2 γ + Ni2(AlTi) + Ni3AlTi γ + Ni3(AlTi) 1 γ + Ni3Ti + Ni3(AlTi)

γ

Cellular precipitation

γ + Ni3Ti 0

1

2

3 Titanium, wt%

Fig. 1

Possible aluminum/titanium precipitates

4

Chapter 10: Precipitation-Hardening Stainless Steels / 139

austenite that provides the work-hardening ability in many of these PH alloys. It is this work hardening that gives the PH alloys, especially the semiaustenitic ones, their unusual combination of high strength plus ductility and toughness in the fully hardened state. The complexity of PH steels comes from the processing involved in producing the martensitic structure in which the precipitation will occur. The most straightforward alloys are the martensitic, also called the martensitic PH alloys. These steels are supplied in the fully martensitic condition with hardness in the low Rc 30s. This is confusingly called the annealed condition, or condition A, even though the matrix is untempered martensite. After the material is fabricated, it is subjected to an aging treatment designated by the aging temperature in Fahrenheit (e.g., H-950). These aging temperatures range from 950 °F (510 °C) to 1150 °F (620 °C). A second major group of PH grades is the semiaustenitic. These grades in the normally furnished condition A are fully austenitic. This is accomplished by adding elements that lower the martensite start temperature, such as more chromium, molybdenum, and nickel. The austenite is more formable than martensite, and it has the possibility of superior corrosion resistance because of higher chromium content. This is balanced by the need to use either cold work, cryogenic treatment, or a destabilizing anneal to cause the matrix to become martensitic before its precipitation aging treatment. Last, if the austenite is made very stable by further alloying additions, a precipitation reaction can still be made to occur by the same type of aging treatment without martensite ever forming. The precipitation takes place in austenite and therefore results in lower room temperature Table 1 Alloy

strength than that of which the martensitic or semiaustenitic alloys are capable. The austenitic PH strength is better above 750 °F.

Martensitic Precipitation-Hardenable Stainless Steels The martensitic PH alloys are, as stated, fully martensitic at room temperature. Their martensite is a relatively soft, low-carbon (less than 0.05%) martensite as opposed to the higher carbon found in the martensitic stainless steels. The early alloys of this type, 17–7 PH and 17-4 PH, contained up to 10% δ-ferrite stringers, which caused poor through-thickness toughness. This would be expected from the Schaeffler-Delong diagram, but this is asking too much of the Schaeffler-Delong diagram, which was developed for welds, to predict the phase composition of alloys that have been homogenized by hot working. The inaccuracy of the diagram for more complex systems was overcome, and alloys were designed that had minimal δ-ferrite and still transformed entirely to martensite, if not at room temperature, at least at a reasonably attainable subzero temperature. This was done first by trial and error and more recently by using thermodynamic computer models, such as ThermoCalc, to predict equilibrium phase composition. This development was very significant for making the alloy family useful as a high-strength/hightoughness material for demanding applications requiring high mechanical properties and corrosion resistance. The most advanced PH alloys are martensitic PH grades by Cartech, Custom 465 and 475. Table 1 shows the more significant of these alloys compared on a strength basis; Table 2 shows

Mechanical properties of martensitic precipitation-hardenable alloys UNS

Condition

Yield, MPa

Tensile, MPa

Elongation, %

HRC

Stainless W 17-4 PH 15-5 PH

S17600 S17400 S15500

13-8 PH

S13800

1240 1210 1210 930 1450

1340 1310 1300 1025 1550

14 14 15 17.50 12

42 41 41 34 47

Custom 450

S45000

Custom 455

S45500

Custom 465

S46500

Custom 475

...

Ferrium S53

...

H-950 ( 510) H-925 (495) H-925 (495) H1100 (595) H-950 (510) H1050 (565) H-900 (480) H1100 (595) H-950 (510) H1050 (565) H-950 (510) H1000 (535) H-975 (525) H 1100 (595) ...

1270 460 1515 1205 1650 1500 1855 1315 1565

1350 970 1585 1310 1765 1600 2005 1572 1985

14 23 10 14 11 13 5 13 14–16

42 ... 48 40 49 48 54 48 54

Toughness, CVN ft-lb

... 40 20 70 25 70 60 180 8 25 13 28 ... 18

140 / Stainless Steels for Design Engineers

Table 2

Composition of martensitic precipitation-hardenable alloys

Alloy

Stainless W 17-4 PH 15-5 PH 13-8 PH Custom 450 Custom 455 Custom 465 Custom 465 (275) Custom 475 Ferrium S53

Designation

S17600 S17400 S15500 S13800 S45000 S45500 S46500 ... ... ...

C

Mn

Si

Cr

Ni

0.1 0 0 0 0 0 0 0 0 0.2

0.5 0.6 0.6 0.1 0.3 0.3 0.2 0.2 0.4 0.1

0.5 0.6 0.6 0.1 0.3 0.3 0.2 0.2 0.4 0.1

17 16 15 13 15 12 12 12 11 10

6.3 4.3 4.3 8.5 6 8.5 11 11 8 5.5

Mo

... ... ... 2 0.8 ... 1 1 5 2

Al

Cu

Ti

Other

0.2 ... ... 1.1 ... ... ... ... 1.2 0

... 3.2 3.2 ... 1.5 2.5 ... ... ... ... ... ...

1 ... ... ... ... 1 2 2 ... ...

P 0.3 ... ... ... 0.3 Nb 0.3 Nb ... 0.2 Nb 8.0 Co 1W 0.3 V 14 Co

Fig. 2

Typical microstructures of precipitation-hardenable (PH) stainless steels: (a) 15-5PH as-quenched martensite; (b) 13-8 PH solution treated and aged displaying fine martensite; (c) 17-7 PH displaying ferrite stringers in a martensite matrix; (d) 17-7 PH showing residual ferrite stringers and inclusions

their nominal compositions. Figure 2 shows a series of photomicrographs of PH alloys. The main advancement metallurgically in these alloys from the top, and earliest, in Stain-

less W to the latest in Custom 475 besides the elimination of δ-ferrite is in the volume fraction of the precipitating phase and the elimination of retained austenite. Alloy designers found that to

Chapter 10: Precipitation-Hardening Stainless Steels / 141

reduce δ-ferrite they also tend to stabilize austenite, which also happens to reduce the temperature at which martensite forms, MS. Last, the higher levels of molybdenum reduce the tendency to form secondary austenite during aging. To minimize δ-ferrite requires reducing chromium or molybdenum, which also reduces corrosion resistance. As a result, as alloy strength increases, corrosion resistance is compromised. The alloys with the greatest strength potential, Custom 465 and 475, barely qualify as stainless with around 11% chromium. But, from a utility point of view, these alloys are designed to have maximum mechanical properties with adequate corrosion resistance, so this is viewed as an acceptable compromise. The newest alloy is Ferrium S53, one of the recent alloys designed by computer-assisted thermodynamic calculations. It was designed to replace 300M, 4340, and AerMet 100 on an equal mechanical properties basis but also provide the corrosion resistance necessary to be used in aircraft components without cadmium plating. Its composition superficially seems deficient in chromium to provide “stainlessness,” but the cobalt level raises the thermodynamic activity of chromium sufficiently that the equivalent of 12% chromium in a non-cobalt-containing alloy is achieved. The precipitation hardening mechanism is the precipitation of Mo2C. It has been established that this hardening mechanism optimizes resistance to stress corrosion cracking (SCC) for a given strength level. The alloying characteristics of these grades are: • Low carbon, nitrogen, silicon, and manganese because these elements lower MS without contributing to age hardening • Low chromium to suppress δ-ferrite • Sufficient nickel to suppress δ-ferrite and provide for precipitates without excessively depressing MS

Table 3

• Molybdenum to offset loss of corrosion resistance by minimization of chromium, to increase the temperature at which austenite forms, and to form another hardening precipitate in the presence of cobalt • Cobalt to stabilize austenite while raising MS • Aluminum or titanium to form intermetallic precipitates with nickel or copper to precipitate as elemental copper It is possible to quantify these various influences on phases. This is summarized in Table 3 in terms of the influence of the element on different factors measured in degrees Centigrade for a 12% chromium alloy. Rapid quenching of these alloys is not required. They are air hardenable. But, the cooling of these alloys must be completed expeditiously through the final stages of martensite formation with minimal delay. During delays after the start of martensite transformation has occurred, the remaining austenite tends to stabilize, and full transformation to martensite does not occur. When this happens, the higher levels of austenite reduce subsequent mechanical properties after aging. As with any alloy used at such high strength levels, microstructural cleanliness is essential, but air melting and argon oxygen decarburization (AOD) refining are quite adequate. Corrosion Resistance. The martensitic PH stainless steels obey the same rules as other stainless steels with regard to corrosion resistance. The martensite carries no nitrogen in solution, so the resistance to pitting is given by: (Eq 1)

PREN = %Cr + 3.3%Mo

In the martensitic PH alloys, no chromium is rendered ineffective by the formation of Cr23C6 since carbon is either held low or stabilized by titanium or niobium. Thus, the corrosion

Influence of alloying elements on key transformations

Element

Lowering of % ferrite per % element

N

C

Ni

Co

Cu

Mn

Si

Mo

Cr

V

Al

–220

–210

–20

–7

–7

–6

6

5

14

18

54

–475

–17

0 to 10

–17

–30

–11

–21

–17

–46

0

0

25 to 60

0 to 35

Lowering of MS –475 per % element Change of AC 0 to 280 per % element

0 to 250 –30 to –115

–25 to –66 25 to 73

50 to 290 30 to 750

142 / Stainless Steels for Design Engineers

resistance will equal that of stabilized ferritic alloys of the same pitting resistance equivalent number (PREN) for which voluminous data are available. A greater concern is the risk of SCC in these alloys. While the mechanism of SCC in austenitic alloys is still debatable, it has long been clear that, for martensitic alloys, SCC is simply a manifestation of hydrogen embrittlement in which the hydrogen is provided by local corrosion. The existence of pitting is a sufficient, if not necessary, condition for SCC to occur if the temperature is within the range of susceptibility and the material is inherently susceptible. The material susceptibility is largely a function of resistance to crack propagation in any given alloy that is measured by fracture toughness. The martensitic PH grades have excellent toughness and low rates of crack propagation, but none should be considered immune to SCC since their hardness is never below the Rc 22 level, which is considered to be the threshold hardness for susceptibility to SCC in body-centered cubic (bcc) ferrous alloys. The suitability of high-strength alloys for use in potential SCC-provoking environments containing H2S is regulated in many locales by National Association of Corrosion Engineers (NACE) International Standard MR01-75. In it, the use of S17400 is permitted if it is double tempered at 620 °C and its hardness is 33 HRC or less, while S45000 can be used if it has been aged at 620 °C for 4 h and its hardness is 31 HRC or less. These permissible hardness levels are significantly higher than allowed in non-PH martensitic alloys, 22 HRC, which reflects the fact that the martensitic matrix has the toughness of a lower-hardness martensite. In marine environments, the PH alloys are susceptible to SCC if used at a high strength level. S17400 aged at 480 °C with a yield strength of 1240 MPa is susceptible to SCC, while higher aging temperatures (above 540 °C) producing lower strengths renders the materials immune at stresses near the yield strength, approximately 1170 MPa. The threshold strength for ordinary martensitic stainless steels would be 1030 MPa (Ref 4).

Semiaustenitic PrecipitationHardenable Stainless Steels If a martensitic stainless steel were alloyed more strongly with austenite-stabilizing ele-

ments, the austenite could be made stable at room temperature. This would make the alloy softer and more fabricable and, most importantly, permit them to be manufactured as coldrolled sheet and strip. If the austenite could then be transformed to martensite by cryogenic treatment, cold work, or special heat treatment, then it could be age hardened just like the martensitic PH grades. This has been accomplished for a group of alloys called the semiaustenitic PH grades. The “semi” signifies that the austenite in these alloys is metastable rather than stable at ambient temperatures. Also, it should be noted that these semiaustenitic alloys usually contain some δ-ferrite in their predominantly austenitic microstructure after annealing. These alloys are complex metallurgically because of the technique used to make the austenite stable at room temperature after a full solution anneal. The austenite is rendered stable by fairly high levels of carbon, a powerful austenite stabilizer, in solution. The amount of carbon that can be held in solution is a function of annealing temperature. The 1050 °C anneal of the condition A mill anneal puts all the carbon in solution, giving the austenite the stability of a normal 301-type alloy. This permits extensive forming. The key is to apply a subsequent lower-temperature anneal so that less carbon goes into solution. Some of it will thus form M23C6. This is in a sense deliberately sensitizing the alloy, but the sensitization takes place at such a high temperature that chromium deficits around precipitated carbides are minimized by diffusion. This causes a higher Ms temperature because of the lower amount of carbon, and chromium, in solution in the austenite. Depending on the temperature at which the anneal is done, the Ms temperature can be controlled so that a transformation to martensite can be raised to either room temperature or some attainable cryogenic temperature. Figure 3 shows a chart of these heat treatment options. The lower strength levels achieved by the condition T route in Fig. 3 reflect the lower carbon content of the martensite, while the highest strength of the condition C route reflects the compound influence of cold work and martensite hardness with similar subsequent contribution from age hardening. The main alloys of this group are listed in Table 4. Examination of the chemistries in this table shows that the first two alloys rely on the precipitation of Ni3Al for the hardening, while the last two have no apparent precipitating components. Their hardening is a more subtle

Chapter 10: Precipitation-Hardening Stainless Steels / 143

Fig. 3

Processing routes for S15700 Source: Ref 5

Table 4 Alloy

17-7 PH 15-7 PH AM-350 AM-355

Compositions of semiaustenitic precipitation-hardenable alloys Designation

S17700 S15700 S35000 S35500

C

Mn

Si

Cr

Ni

Mo

Al

N

0.1 0.1 0.1 0.1

0.5 0.5 0.8 0.9

0.3 0.3 0.4 0.4

17 15 17 16

7.1 7.1 4.3 4.3

... 2.2 2.8 2.8

1 1 ... ...

0 0 0.1 0.1

secondary hardening from the tempering of martensite rather than the classic precipitation hardening via precipitation of intermetallic compounds. In 17-7 and 15-7, aluminum rather than titanium is the precipitating agent because

titanium would preferentially deplete the alloy of carbon and nitrogen, precluding the action of the conditioning heat treatment, which relies on manipulating the amount of carbon in solution. In AM350 and AM355, it is the precipitation of

144 / Stainless Steels for Design Engineers

Table 5

Mechanical properties of semiaustenitic precipitation-hardenable alloys

Alloy

UNS

Condition

17–7 PH

...

15–7 PH

...

AM-350

...

AM-355

...

TH 1050 (565) RH 950 (510) CH 900 (480) TH 1050 (565) RH 950 (510) CH 900 (510) SCT 850 (450) SCT1000 (540) SCT 850 (450) SCT 1000(540)

Yield, MPa

1100 1380 1585 1380 1550 1720 1210 1020 1250 1035

(Cr,Fe)2N within the martensite phase that is responsible for age hardening. In addition, molybdenum produces a secondary hardening in carbon-bearing martensite. These two early alloys did not possess the hardening potential that alloys employing copper-, titanium-, or aluminum-based precipitates enjoy. The mechanical properties of these alloys are not greatly different from the martensitic alloys, as can be seen in Table 5. Their separate existence is due to the need for alloys that are more fabricable at room temperature than are the alloys that are martensitic at room temperature. This benefit is offset by the necessity to condition anneal before age hardening. If corrosion resistance is a high concern, then the alloy and heat treatment that yields the greatest amount of chromium in solution should be chosen. Thus condition CH is better than RH, which is better than TH, the order of ascending solution annealing temperature and ascending Ms. The martensitic PH grades have somewhat better strength because they have a more uniformly martensitic structure, and they employ a tougher, lower-carbon martensite. The fact that the semiaustenitic alloys are typically sheet products generally makes their service toughness requirements less onerous, so that their retained δ-ferrite is not a crippling drawback because of its detrimental effect on through thickness toughness. Corrosion Resistance. The semiaustenitic PH alloys tend to higher values of PREN than the martensitic alloys inherently since they are alloyed to have lower Ms temperatures. The thermal processing of these alloys causes a significant portion of the chromium to be removed from solution as chromium carbide. This lowers the corrosion resistance from what would be expected based on the bulk composition. From an engineering point of view, it is best to assume

Tensile, MPa

1310 1520 1655 1450 1650 1790 1420 1165 1510 1124

Elongation, %

10 9 2 7 6 2 12 15 13 22

HRC

42 46 49 45 48 50 46 40 48 38

that all carbon is present as chromium carbide, and that the chromium content is diminished by that amount before applying Eq 1. In addition, carbon can remove some molybdenum in the form of carbides, and nitrogen can remove chromium as a nitride. The effective corrosion resistance of these alloys thus is similar to ferritic 430. The designer is thus advised to consult with producers about corrosion resistance depending on the thermal processing that will be used, especially if double aging is performed, which can cause some degree of chromium depletion at grain boundaries. While the general and pitting corrosion resistance of the semiaustenitic PH alloys are never quite as good as most austenitic stainless, they have very good resistance to SCC compared to ordinary martensitic stainless steels. The δ-ferrite and the generally well-tempered martensitic matrix provide a crack-arresting feature and good inherent toughness that resist SCC at higher strength levels than in straight martensitic stainless steels. AM-355 in the SCC (850 °F) condition can withstand stresses of 75% of 0.2% offset yield strength in salt spray without SCC failure.

Austenitic Precipitation-Hardenable Stainless Steels The austinitic PH class consists of just one important commercial alloy, A-286. The importance of the alloy is that it is entirely stable austenite in both the solution-annealed and the age-hardened condition. This means it is very formable and nonmagnetic. And, because the precipitation takes place in an austenite matrix, the precipitation takes place at a higher temperature, around 700 °C. This gives the alloy the

Chapter 10: Precipitation-Hardening Stainless Steels / 145

Table 6 Alloy

A-286 Discalloy

Fig. 4

Austenitic precipitation-hardenable composition UNS

C

Mn

Si

Cr

Ni

Mo

Al

V

Ti

S66286 S66220

0.05 0.04

1.5 1.6

0.5 0.5

15 14

25.5 26

1.3 3

0.15 ...

0.3 ...

2.15 1.7

A-286 properties as a function of test temperature. Source: Ref 5

potential to be used to near 700 °C without fear of overaging. Thus, austenitic PH stainless represents a way to strengthen the austenite matrix, which has the following advantages: • High ductility and therefore high formability in the soft, unaged condition • High toughness at all temperatures and strength levels • Excellent creep and stress rupture properties • Excellent oxidation, corrosion, and SCC resistance This alloy is rightly considered an ironbased superalloy and is the root of the group that succeeds it in properties, the nickel- and cobalt-based superalloys. These last alloys can attain greater strength and creep resistance than A-286, effectively ending further development of austenitic PH alloys. The lower cost of the A-286 alloy, compared to nickel-base

PH alloys, makes it attractive for a variety of aerospace and nonaerospace uses. The hardening mechanism of A-286 is the precipitation of Ni3 (aluminum, titanium). Diffusion, even at the higher temperatures, is slower in austenite, so aging treatments are typically 16 h. Table 6 gives the typical composition of A-286, and Fig. 4 shows some typical properties as a function of temperature for a standard 980 °C solution treatment followed by a 720 °C, 16 h aging. The mechanical properties can be greatly enhanced by cold working prior to aging, as is shown in Fig. 5. The toughness of the austenitic matrix is abundant and quite temperature insensitive. Charpy V-notch values of over 60 J are typical from –200 to 800 °C. The corrosion resistance of A-286 is comparable to that of 304 and 316. It has slightly

146 / Stainless Steels for Design Engineers

Fig. 5

The influence of cold work on aging response in A-286. DPH, diamond pyramid hardness. Source: Ref 5

better resistance to SCC despite its higher strength level. REFERENCES

1. F.B. Pickering, Physical Metallurgical Developments of Stainless Steel, Stainless ’84, Goteborg, Sept 3–4, 1984, p 2–28. 2. M. Murayama, Y. Katayama, and K. Hono, Microstructural Evolution in a 17-4 PH Stainless Steel After Aging at 400 °C, Met-

allurgical and Materials Transactions A, Vol 30A, Feb 1999. 3. G. Aggen, Ph.D. thesis, Carnegie Mellon University 4. E.E. Denhard, “Stress Corrosion Cracking of High Strength Stainless Steels in Atmospheric Environments”, paper presented at the Twenty-fourth Meeting of the AGARD Structures and Materials Panel (Turin, Italy), April 17–20, 1967. 5. Allegheny Technology Blue Sheets

Stainless Steels for Design Engineers Michael F. McGuire, p 147-154 DOI: 10.1361/ssde2008p147

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 11

Casting Alloys Summary WITH TYPICAL ALLOY SYSTEMS, casting is often the most convenient method by which to produce components. This is true for stainless steels—both for corrosion-resisting and for heat-resisting applications. This chapter discusses primarily the alloys used for stainless steel castings and their metallurgy. Foundry methods are discussed to the degree they are specific to the stainless alloys.

Stainless Steel Casting Alloys Essentially any wrought stainless alloy composition can be modified to be made as a cast alloy. The systemic difference between cast alloys and their wrought equivalents is that cast alloys generally contain between 1.0 and 2.5% silicon. As with other ferrous alloys, this is done to increase the fluidity of the melt to make it cast more effectively. Silicon has strong metallurgical effects, both beneficial and detrimental, which should be understood by the user of cast stainless steels. These are explained. A second general observation is that the stabilized ferritic stainless steel alloys, which constitute almost half the tonnage of all stainless steel used, are notably absent from the cast alloys. This is because these alloys are single phase at all temperatures in the solid state and because they have large as-cast grain sizes that can only be refined by heavy cold work followed by annealing. This makes them quite lacking in toughness as cast. Since heavy cold work defeats the purpose of casting to achieve a nearnet shape, stabilized ferritic stainless steels are seldom used as castings. Also, the standard stabilizing alloy, titanium, is too readily oxidized for normal foundry practice to avoid the loss of this

essential element. Thus, the casting alloys listed in Tables 1 and 2 (Ref 1) are recognizable as approximate counterparts of the co-listed wrought alloys (AISI grade). This cross reference to wrought equivalents is helpful when looking for data about an alloy that may be more easily found for wrought alloys than for cast. The High Alloy Product Group of the Steel Founder’s Society of America employs a naming system (ACI, the Alloy Casting Institute) for cast alloys that is significant; these designations are currently assigned by ASTM as grades and are added to ASTM specifications. The first letter, “C” or “H,” indicates corrosion resisting. The second letter indicates the relative amount of nickel, from a minimum of 0 to 1% for “A” up to 30% nickel for “N” alloys. The number following the hyphen for “C” alloys designates the maximum carbon in hundredths of a percent. The suffix letters designate additional alloying elements, such as Cu for copper, M for molybdenum, N for nickel or nitrogen, F for free machining, and C for columbium (niobium). The heat-resisting, “H,” alloys have generally only a second letter designating relative nickel level on the same scale as “C” alloys but going past stainless steels all the way to nickel-based alloys. The inclusion of a number after the first two letters indicates the center of the carbon range expressed in hundredths of a percent by weight. To learn more about the influence of alloying elements, refer to the chapters on the individual alloy families; see Section 3. Here, we briefly summarize: • Pitting and crevice corrosion resistance, as well as general corrosion resistance, are enhanced by chromium, molybdenum, tungsten, and nitrogen and carbon in solution. • Localized corrosion is caused by chromium depletion, which occurs when precipitates

148 / Stainless Steels for Design Engineers

Table 1

Compositions of cast stainless corrosion resisting alloys Composition(a), wt%—maximum or range

ACI designation

Nearest AISI grade

Chromium alloys CA-15 410 CA-15M CA-40 420 CA-40F 420F CB-30 431,442 CC-50 446 Chromium-nickel alloys CA-6N CA-6NM S41500 CA-28MWV 422

UNS

%C

%Mn

%Si

%Cr

%Ni

J91150 J91151 J91153 J91154 J91803 J92613

0.15 0.15 0.40 0.2–0.4 0.30 0.30

1.00 1.00 1.00 1.00 1.00 1.00

1.50 0.65 1.50 1.50 1.50 1.50

11.5–14.0 11.5–14.0 11.5–14.0 11.5–14.0 18.0–22.0 26.0–30.0

1.0 1.0 1.0 1.0 2.0 4.0

J91650 J91540 J91422

0.06 0.06 0.20–0.28

0.50 1.00 0.50–1.00

1.00 1.00 1.00

10.5–12.5 11.5–14.0 11.0–12.5

6.0–8.0 3.5–4.5 0.5–1.0

CB-7Cu-1

17-4PH (AISI 630) J92180

0.07

0.70

1.00

15.5–17.7

3.6–4.6

CB-7Cu-2

15-5 PH (XM-12)

J92110

0.07

0.70

1.00

14.0–15.5

4.5–5.5

CD-3MN

2205 (S32205)

J92205

0.03

1.50

1.00

21.0–23.5

4.5–6.5

CD-3MCuN

255 (S32550)

J93373

0.03

1.20

1.10

24.0-26.7

5.6–6.7

CD-3MWCuN (S32760)

J93380

0.03

1.00

1.00

24.0–26.0

6.5–8.5

CD-4MCu CD-4MCuN

J93370 J93372

0.04 0.04

1.00 1.00

1.00 1.00

24.5–26.5 24.5–26.5

4.75–6.0 4.7–6.0

CD-6MN

J93371

0.06

1.00

1.00

24.0–27.0

4.0–6.0

J93404 J93345 J93423 J92500 J92800 J92700 J92600 J92710 J92900 J92590 J92901 J92971 J92972 J92701 J92602 J93790

0.03 0.08 0.30 0.03 0.03 0.03 0.08 0.08 0.08 0.04–0.10 0.04–0.10 0.10 0.10 0.12 0.16 0.20 0.06

1.50 1.00 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 7.00–9.00 1.50 1.50 1.50 4.00–6.00

1.00 1.50 2.00 2.00 2.00 1.50 2.00 2.00 2.00 2.00 1.50 1.50 3.50–4.50 2.00 2.00 2.00 1.00

24.0–26.0 22.5–25.5 26.0–30.0 17.0–21.0 17.0–21.0 17.0–21.0 18.0–21.0 18.0–21.0 18.0–21.0 18.0–21.0 18.0–21.0 15.0–18.0 16.0–18.0 18.0–21.0 18.0–21.0 18.0–21.0 20.5–23.5

6.0–8.0 8.0–11.0 8.0–11.0 8.0–12.0 8.0–12.0 9.0–13.0 8.0–11.0 9.0–12.0 9.0–12.0 8.0–11.0 9.0–12.0 13.0–16.0 8.0–9.0 9.0–12.0 9.0–12.0 8.0–11.0 11.5–13.5

CE-3MN CE-8MN CE-30 CF-3 CF-3M CF-3MN CF-8 CF-8C CF-8M CF-10 CF-10M CF-10MC CF-10SMnN CF-12M CF-16F CF-20 CG-6MMN

2507 (S32750)

CG-8M CG-12 CH-8 CH-10 CH-20 CK-3MCuN

317 308 309S 309H 309 254SMO™

J93000 J93001 J93400 J93401 J93402 J94653

0.08 0.12 0.08 0.04–0.10 0.20 0.025

1.50 1.50 1.50 1.50 1.50 1.20

1.50 2.00 1.50 2.00 2.00 1.00

18.0–21.0 20.0–23.0 22.0–26.0 22.0–26.0 22.0–26.0 19.5–20.5

9.0–13.0 10.0–13.0 12.0–15.0 12.0–15.0 12.0–15.0 17.5–19.5

CK-20 CN-3M CN-3MN CN-7M CN-7MS CT-15C

310 904L AL-6XN® 320

J94202 J94652 J94651 N08007 J94650 N08151

0.20 0.03 0.03 0.07 0.07 0.05–0.15

2.00 2.00 2.00 1.50 1.50 0.15–1.50

2.00 1.00 1.00 1.50 3.50 0.50–1.50

23.0-27.0 20.0–22.0 20.0–22.0 19.0-22.0 18.0–20.0 19.0–21.0

19.0–22.0 23.0–27.0 23.0–27.0 27.5–30.0 22.0–25.0 31.0–34.0

312 304L 316L 316LN 304 347 316 304H 316H 316H NITRONIC™ 60 316 303 302 NITRONIC™ 50

(a) Balance Fe for all compositions. Source: Ref 1

%Mo

%Other

0.50 0.15–1.00 0.50 0.20–0.40 Ss

0.4–1.0 0.9–1.25

0.9–1.25 W, 0.2–0.3 V 2.5–3.2 Cu, 0.2–0.35 Nb, 0.05 N max 2.5–3.2 Cu, 0.2–0.35 Nb, 0.05 N max 2.5–3.5 1.0 max Cu, 0.10–0.30 N 2.9–3.8 1.4–1.9 Cu, 0.22–0.33 N 3.0–4.0 0.5–1.0 Cu, 0.5–1.0 W, 0.20–0.30 N 1.75–2.25 2.75–3.25 Cu 1.75–2.25 2.75–3.25 Cu, 0.10–0.25 N 1.75–2.25 1.75–2.5 Cu, 0.15–0.25 N 4.0–5.0 0.10–0.30 N 3.0–4.5 0.10–0.30 N

2.0–3.0 2.0–3.0

0.10–0.20 N Nb

2.0–3.0 2.0–3.0 1.75–2.25 2.0–3.0 1.5 max

(10xC)–1.2 Nb 0.08–0.18 N 0.2–0.35 Se

1.5–3.0

0.1–0.3 Nb, 0.1–0.3 V, 0.2–0.40 N

6.0–7.0

0.5–1.0 Cu, 0.18–0.24 N

4.5–5.5 6.0–7.0 2.0–3.0 2.5–3.0

0.18–0.24 N 3.0–4.0 Cu 1.5–2.0 Cu 0.5–1.5 Nb

Chapter 11: Casting Alloys / 149

Table 2

Compositions of cast heat-resistant stainless and nickel base alloys Composition(a), wt%—maximum or range

ACI designation

HA HC HD HE HF HH HI HK HK-30 HK-40 HL HN HP HP-50WZ HT HT-30 HU HW HX

Nearest AISI grade

504 446 327 312 302B 309 310

330

UNS

J82090 J92605 J93005 J93403 J92603 J93505 J94003 J94224 J94203 J94204 N08604 J94213 N08705 N08605 N08603 N08005 N08006 N06050

%C

%Cr

0.20 max 0.50 max 0.50 max 0.20–0.50 0.20–0.40 0.20–0.50 0.20–0.50 0.20–0.60 0.25–0.35 0.35–0.45 0.20–0.60 0.20–0.50 0.35–0.75 0.45–0.55 0.35–0.75 0.25–0.35 0.35–0.75 0.35–0.75 0.35–0.75

8–10 26–30 26–30 26–30 19–23 24–28 26–30 24–38 23–27 23–27 28–32 19–23 24–28 24–28 13–17 13–17 17–21 10–14 15–19

%Ni

4–max 4–7 8–11 9–12 11–14 14–18 18–22 19–22 19–22 18–22 23–27 33–37 33–37 33–37 33–37 37–41 58–62 64–68

%Si max

1.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 1.75 1.75 2.00 2.00 2.00 2.50 2.50 2.50 2.50 2.50 2.50

(a) Balance Fe for all compositions. Manganese content: 0.35–0.65% for HA, 1% for HC, 1.5% for HD, 2% for the other alloys. Phosphorus and sulfur contents: 0.04 (max) for all but HP-50WZ. Molybdenum is intentionally added only to HA: 0.90–1.2%. Maximum molybdenum for other alloys is 0.5%. HH contains 0.2% N (max). HP-50WZ also contains 4–6% W, 0.1–1.0% Zr, and 0.035% S (max) and P (max). Source: Ref 1

form in the solid state. These precipitates are carbides, oxides, and sulfides as well as intermetallic phases richer in chromium, molybdenum, or nitrogen than the matrix. • General corrosion resistance follows the above guidelines but is also helped by copper and nickel, which do not assist in pitting resistance. • High-temperature oxidation resistance is enhanced by increasing chromium and silicon. Wrought alloys employ aluminum and rare earths to help oxidation resistance, but the difficulty of keeping these elements from being oxidized requires special techniques such as vacuum induction melting and inert refractories for molds. • Iron-chromium (ferritic) alloys have better thermal fatigue resistance but poorer creep resistance than iron-chromium-nickel (austenitic) alloys. The alloy designation system largely ignores the wrought alloy distinctions by microstructure (i.e., ferritic, austenitic, duplex, PH [precipitation hardening], and martensitic). One reason is that the most widely used wrought-stabilized ferritics (e.g., 409, 439) do not exist as common casting alloys, and nominally austenitic alloys in the cast form contain enough ferrite to be significantly magnetic. Thus, the distinctions based on phase are not as well defined for casting alloys. The high ferrite content in the nominally austenitic casting alloys is to avoid or at least

minimize solidification hot cracking or to allow weld repair of cracks that do form. It has been shown that the existence of ferrite can increase the resistance to stress corrosion cracking.

Metallurgy of “C” Alloys The corrosion-resistant “C” series have wrought counterparts from which they differ essentially only in silicon content. This silicon has no significant influence on corrosion resistance or mechanical or physical properties, so an understanding of these alloys by approximating them to their wrought counterparts is justified. The main difference between the cast and wrought product forms of these alloys is the grain structure. In wrought grades, the grain structure can be manipulated by deformation and heat treatment. The use of deformation is not an option in cast alloys; consequently, the opportunities for grain refinement in cast alloys are limited. To counteract the problems of lower corrosion resistance of cast grades, a homogenizing solution heat treatment is necessary to counteract the chromium depletion that occurs due to solidification segregation and precipitating phases. Representative mechanical properties for “C” alloys are listed in Table 3 (Ref 1). Martensitic. CA alloys are martensitic. The metallurgy is straightforward and equivalent to their wrought counterparts. The mechanical

Heat treatment(a)

>955 oC (1750 oF), AC, T 980 oC (1800 oF), AC, T 980 oC (1800 oF), AC, T 1040 oC (1900 oF), OQ, A 790 oC (1450 oF), AC 1040 oC (1900 oF), AC 1120 oC (2050 oF), FC to 1040 oC (1900 oF), WQ 1120 oC (2050 oF), FC to 1040 oC (1900 oF), A 1095 oC (2000 oF), WQ >1040 oC (1900 oF), WQ >1040 oC (1900 oF), WQ >1040 oC (1900 oF), WQ >1040 oC (1900 oF), WQ >1095 oC (2000 oF), WQ >1040 oC (1900 oF), WQ >1040 oC (1900 oF), WQ >1065 oC (1950 oF), WQ >1065 oC (1950 oF), WQ >1095 oC (2000 oF), WQ >1040 oC (1900 oF), WQ >1095 oC (2000 oF), WQ 1150 oC (2100 oF), WQ 1150 oC (2100 oF), WQ 1120 oC (2050 oF), WQ 669 531 600 531 586 531 552 621 552 531 531 565 607 524 770 476

97 77 87 77 85 77 80 90 80 77 77 82 88 76 112 69

130

896

ksi

120 115 150 190 95 97 108

827 793 1034 1310 655 669 745

MPa

Tensile strength

434 248 290 255 310 248 262 310 290 262 276 303 345 262 365 214

634

689 689 862 1172 414 448 558

MPa

24 22 10 14 15 18 25 20 18 60 50 55 50 50 55 45 50 39 52 45 38 37 50 48

92 63 36 42 37 45 36 38 45 42 38 40 44 50 38 53 31

Elongation % in 50 mm, 2 in.

100 100 125 170 60 65 81

ksi

Yield strength, 0.2% offset

...

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

...

60 55 30 54 ... ... ...

Reduction in area, %

Room temperature mechanical properties of corrosion resisting cast stainless alloys

(a) AC, air cool; FC, furnace cool; OQ, oil quench; WQ, water quench; T, temper; A, age. Source: Ref 1

CE-30 CF-3 CF-3A CF-8 CF-8A CF-20 CF-3M CF-3MA CF-8M CF-8C CF-16F CG-8M CH-20 CK-20 CN-3MN CN-7M

CA-6NM CA-15 CA-40 CB-7Cu CB-30 CC-50 CD-4MCu

Alloy

Table 3

190 140 160 140 156 163 150 170 170 149 150 176 190 144 185 130

305

269 225 310 400 195 210 253

Hardness, HB J

9.5 149.2 135.6 100.3 94.9 81.4 162.7 135.6 94.9 40.7 101.7 108.5 40.7 67.8 190 94.9

35.3

94.9 27.1 2.7 33.9 2.7 ... 74.6

7 110 100 74 70 60 120 100 70 30 75 80 30 50 140 70

26

70 20 2 25 2 ... 55

ft-lb

Specimen

Keyhole notch V-notch V-notch Keyhole notch Keyhole notch Keyhole notch V-notch V-notch Keyhole notch Keyhole notch Keyhole notch V-notch Keyhole notch Izod V-notch V-notch Keyhole notch

V-notch

V-notch Keyhole notch Keyhole notch V-notch Keyhole notch ... V-notch

Charpy toughness

150 / Stainless Steels for Design Engineers

Chapter 11: Casting Alloys / 151

properties are governed by the thermal processing, and strength, hardness, and toughness can be varied over a wide range. The CB 30 and CC 50 alloys are ferritic and, as such, have negligible toughness but effectively deliver corrosion resistance. The toughness of CB 30 can be improved by balancing the chromium and silicon to a lower part of the range and the carbon and nickel to the higher end to render the microstructure partly martensitic. Precipitation Hardening. The cast PH alloys include CB-7Cu-1, which behaves in a similar way to 17-4PH, which has an overlapping composition range. Note that most other major wrought PH grades rely on titanium and aluminum to form coherent strengthening precipitates and so do not have cast counterparts. Copper, which can harden ferrite but not austenite, is thus the only strengthener available. There is one cast PH alloy that has no wrought counterpart. It is CD-4MCu; however, it is rarely used in the precipitation-hardened condition and is most commonly classified as a duplex stainless steel in which the nitrogen level is closely controlled. This is a highly alloyed duplex grade that contains copper to precipitation harden the ferrite phase. Oil field CO2 corrosion is resisted by alloys that resemble the martensitic PH grades. These alloys are discussed in Chapter 22, “Petroleum Industry Applications” and can be considered castable alloys. Duplex. The cast equivalents of alloys 2205 and 2507, J 92205, and J 93380 have similar properties and corrosion resistance. Modern wrought duplex alloys rely on nitrogen to partition the alloy with uniform corrosion resistance in each phase and to suppress intermetallic phase formation. Cast alloys are effectively limited to 0.25% nitrogen before gas porosity becomes excessive. Porosity can be reduced by replacing some nickel with manganese, which increases nitrogen solubility. Doing so would expand the most promising area of stainless steel development, lean duplex alloys such as 2101 and 2003, to the cast grades. Alloy 2101 with 4 to 6% manganese provides the corrosion resistance of CF8M or 316L with total nickel plus molybdenum of only 2% versus the 12% required for the austenitic alloy. The duplex alloys also have greater strength and are nearly immune to stress corrosion cracking. These alloys represent significant cost-savings potential for the foundry and for its customers. CE-30 is duplex steel, which is fairly simple metallurgically and uses only chromium for corrosion resistance. However, its

high level of nickel negates much of the potential cost savings duplex alloys offer. All castings are solution annealed and quenched to eliminate embrittling intermetallic phases. The cast duplex alloys may offer a better engineering approach than the equivalent austenitic cast alloys because they have greater strength and lower alloy cost for the same level of corrosion resistance. They do not have the same problems of hot cracking that make casting austenitic steels difficult. The poor hot workability of duplex steel is not an issue for castings. It is important for designers to understand that cast duplex steels are totally compatible galvanically with wrought or cast austenitic alloys of the same corrosion resistance. Mixing components with different microstructures does not create a galvanic differential when corrosion resistance levels are similar. Reluctance to mix alloys for galvanic reasons can be an expensive error when their similar corrosion resistances makes them compatible, even if they are quite different microstructurally. Austenitic-Ferritic. The typical CF alloys, which make up about two-thirds of U.S. stainless steel castings, are nominally austenitic but always contain ferrite. This is not detrimental and improves resistance to stress corrosion cracking and sensitization. Homogenization annealing can reduce the amount of ferrite and result in lower yield and tensile strength and higher elongation and toughness. The composition balance is the main determinant of ferrite level. Increasing the nickel, nitrogen, manganese, or carbon content decreases ferrite. Increasing chromium, silicon, or molybdenum content increases ferrite. Increasing the solidification rate will increase the ratio of austenite to ferrite in duplex or austenitic-ferritic alloys. The predominantly austenitic matrix has a very high toughness even at cryogenic temperatures. Ferrite, if continuous, decreases toughness. Fortunately, it is seldom present as a continuous phase. The loss of toughness associated with high ferrite content can be aggravated by heating the ferrite above 475 oC (885 oF) for a sufficient time for the ferrite to decompose to the brittle α and α'. At higher temperatures, development of the σ phase would have a similar embrittling effect. These phases thus formed are quickly redissolved and removed by annealing. Note that sometimes copper is added to austenitic alloys to improve corrosion resistance in sulfuric acid environments. It has no precipitation hardening effect in austenite, as it does in ferrite. When

152 / Stainless Steels for Design Engineers

used as a precipitating hardening agent, copper does not increase corrosion resistance. Virtually any non-titanium-bearing, corrosion-resistant, austenitic, wrought alloy can have a cast counterpart. Curiously, the 2xx lownickel alloys are not found in most cast alloys lists. If a specific wrought alloy cannot be found to have a published cast counterpart, the designer should not avoid requesting a producer to supply a version that the foundry is confident of making. The designer must thoroughly understand the design of the alloy desired so that any alterations to its composition necessary to allow castability will not compromise expected performance.

Metallurgy of “H” Alloys The heat-resisting “H” alloys are principally austenitic. Alloying elements and impurities diffuse more slowly through the face-centered cubic (fcc) austenitic structure than the bcc ferrite structure, making the austenite more resistant to diffusion-controlled creep. Austenite

Table 4

Mechanical properties of heat-resistant cast stainless alloys at room temperature Tensile strength

Alloy

Standard grades HA HC HD HE HF HH, type 1 HH, type 2 HI HK HL HN HP HPNb(d) HPNbTi(e) HT HU HW HX

has higher thermal expansion and lower thermal conductivity than ferrite, which aggravates thermal fatigue and oxide spalling. Nevertheless, the better high-temperature strength of austenite generally is the predominant consideration, and most “H” alloys are austenitic. Tables 4 and 5 list properties of “H” alloys (Ref 1 to 3). Ferritic HA, HC, HD. Of the ferritic alloys HA, HC, and HD, HA with less than 10% chromium is not quite stainless but is useful to 650 oC (1200 oF) for petroleum refinery applications. HC and HD are very high chromium ferritics that have very low toughness and creep resistance but are quite oxidation and sulfidation resistant. They can be cost-effective materials when high-temperature strength is not an overriding concern. Austenitic HE-HP. The predominant hightemperature grades are the austenitic HE through HP, after which come the nickel alloys, which are not generally classified as stainless steels because they contain less than 50% iron. The high material cost of the nickel base alloys restricts their use to those specific environments where maximum carburization or nitriding

Condition

MPa

N + T(a) As-cast Aged(b) As-cast As-cast Aged(b) As-cast Aged(b) As-cast Aged(b) As-cast Aged(b) As-cast Aged(b) As-cast Aged(c) As-cast As-cast As-cast As-cast As-cast As-cast Aged(c) As-cast Aged(f) As-cast Aged(g) As-cast Aged(f)

738 760 790 585 655 620 635 690 585 595 550 635 550 620 515 585 565 470 490 450 450 485 515 485 505 470 580 450 505

Yield strength

ksi

MPa

ksi

107 110 115 85 95 90 92 100 85 86 80 92 80 90 75 85 82 68 71

558 515 550 330 310 380 310 345 345 380 275 310 310 450 345 345 360 260 275 220 220 275 310 275 295 250 360 250 305

81 75 80 48 45 55 45 50 50 55 40 45 45 65 50 50 52 38 40

70 75 70 73 68 84 65 73

40 45 40 43 36 52 36 44

Elongation, %

21 19 18 16 20 10 38 25 25 11 15 8 12 6 17 10 19 13 11 8 8 10 5 9 5 4 4 9 9

Hardness, HB

220 223 ... 90 200 270 165 190 185 200 180 200 180 200 170 190 192 160 170 ... ... 180 200 170 190 185 205 176 185

(a) Normalized and tempered at 675 °C (1250 °F). (b) Aging treatment: 24 h at 760 °C (1400 °F), furnace cool. (c) Aging treatment: 24 h at 760 °C (1400 °F), air cool. (d) ISO 13583-2 specification minima. (e) ISO 13583-2 specification limits for microalloyed grade. (f) Aging treatment: 48 h at 980 °C (1800 °F), air cool. (g) Aging treatment: 48 h at 980 °C (1800 °F), furnace cool. Source: Ref 1

Chapter 11: Casting Alloys / 153

Table 5

High-temperature mechanical properties of “H” alloys

Alloy

Temp

HA

1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 800 oC 800 oC 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF 1400 oF 1800 oF

HC HD HE HF HH TYPE 1 HH TYPE 2 HI HK HL HN HP HPNb(a) HPNbTi(a) HT HU HW HX

Yield strength ksi, MPa

Tensile strength ksi, MPa

Creep rate 0.0001%/h psi, MPa

1% in 100,000 h psi, MPa

16 8.7 2.1

10.5 2.5 36 15

35 17 6.3 18 7

33 9 35 11

50 18.7

1.3 3.6 3.5 1.0 3.5 1.0 6.0 1.6 (est) 3.0 1.1 7.0 2.1 6.6 1.9 6.8 2.7 7.0 2.8 (est)

Stress to rupture in 1000 h

27

4.4

2.0

6.3 0.9

1.3 0.6 7.0 2.5 11.0 2.5 9.1

8.0 1.6 8.5 2.6 12.0 2.8 15 5.2

2.4

... ... 26 8 6.2 23 8 19.5 6.9

... ... 35 11 40 10 32 10 42 10.7

2.1 51 MPa 54 MPa 8.0 2.0 8.5 2.2 6.0 1.4 6.4 1.6

Stress to rupture in 10,000 h

0.9

1.7

2.1

... ... 12 2.7 2.9 7.8 2.6

55 MPa 64 MPa 1.7 1.8

2.2

(a) Data from ISO 13583-2. Source: Ref 2, 3

resistance is mandatory. For oxidation and sulfidation resistance, the iron base alloys are preferred. These cast stainless steels derive their oxidation resistance from their chromium level. The chromium near the surface acts as a reservoir to replenish the protective iron/chromium oxide scale as explained in Chapter 6 in the section on oxidation. Silicon, another stable oxide former, assists in forming this protective scale and resistance to carburization. Other typical alloying elements do not aid in oxidation resistance. If it were possible to cast these alloys with aluminum or rare earth additions without them being lost to oxidation before solidification, there could be some impressive benefits. Such alloys exist in wrought form, for example, 153MA and 253MA. The metallurgical basis of the benefits from aluminum and rare earth

alloying also are discussed in the oxidation section of Chapter 6. The major problem that all producers of stainless steels face is that of transferring molten metal from the furnace to the mold cavity. This problem is heightened when the foundry makes complex shapes. Methods developed to protect the molten stream from exposure to air to prevent reoxidation have shown great promise and have been demonstrated by the wrought alloy producers who tend to produce much simpler shapes than the foundry. Protection of the molten stream could result in castings with much better hightemperature performance that could be used instead of some use of higher nickel alloys. High-temperature strength is modestly improved by higher levels of chromium and nickel. Molybdenum improves high-temperature

154 / Stainless Steels for Design Engineers

strength, but its detrimental effect on oxidation resistance and its promotion of intermetallic precipitation limits its use. Carbon is very effective for promoting high-temperature strength and suppression of intermetallic phase formation. All “H” alloys, therefore, employ much higher carbon levels than the “C” alloys. This does, however, directly imply that the corrosion resistance of “H” alloys, should it be an issue, is significantly degraded over otherwise similar “C” alloys. The HP grades have undergone significant development over the last 30 years. This development has come about through the addition of niobium to increase creep and rupture properties. The use of microalloying additions has delivered creep and rupture properties some 30% higher than the HP grade without niobium microalloying. It is unfortunate that these HP grades have not at this time found their way into ASTM standards; however, work is under way to remedy this omission. Currently, the most upto-date collection of these grades can be found in ISO 13583-2.

higher-performance grades. It is also possible to use AOD-refined master melt stock to achieve the same benefits as AOD refining while using induction melting. Welding of cast stainless alloys is a common practice and does not present problems when using approved weld procedures and qualified welders. Chapter 17 describes joining methods in detail. The same precautions about sensitization apply to castings. Welding of non-niobiumstabilized “C” alloys with carbon levels above 0.03% will require postweld annealing to redissolve chromium carbides, which will otherwise make the alloy susceptible to corrosive attack in the chromium-depleted regions of the heataffected zone. Iron and nickel base “H” alloys that are fully austenitic can suffer from hot shortness due to sulfide films that precipitate along grain boundaries even at low bulk sulfur levels. This makes them susceptible to hot cracking of welds. Alloys with some ferrite are less susceptible to hot cracking, so most “C” alloys are highly resistant to this problem.

Foundry Practice

REFERENCES

While the scope of this book does not extend to the production of castings, certain aspects are important to the user of castings. For the last 50 years, virtually all stainless has been refined in argon oxidation decarburization (AOD) vessels or versions thereof. For “C” alloys, this refining method should be considered a basic requirement for good quality where the carbon levels are restricted to low levels (e.g., CF3M). “H” alloys are less refined inherently and can be simply arc or induction melted; however it may be necessary to use refining techniques for the

1. Cast Stainless Steels, Metals Handbook, desk ed., J.R. Davis, Ed., ASM International, 1988, p 386–390 2. International Organization for Standardization, www.iso.org, ISO 13583-2 3. Steel Founders Society of America, online documents: http://www.sfsa.org/sfsa/pubs/ index.html SELECTED REFERENCE

• http://www.sfsa.org

Stainless Steels for Design Engineers Michael F. McGuire, p 155-160 DOI: 10.1361/ssde2008p155

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 12

Melting, Casting, and Hot Processing Summary THE PRIMARY PRODUCTION PROCESSES of melting, casting, and hot processing are invisible to the end user. The vast majority of stainless steel is made by arc furnace melting followed by argon oxygen decarburization (AOD) refining and continuous casting. It is not normal, and it is seldom beneficial for the end user to specify processing paths. The end user should, however, be knowledgeable and require the producer to document the process and the producer’s control of it.

Introduction The manner in which stainless steel is made at the producing mill can have a great impact on its final properties. These production methods have undergone a major evolution over the last 50 years and are mainly responsible for stainless steels becoming the practical, widespread engineering materials they are today. Traditional carbon and alloy steel-making methods are not suitable for stainless steels. The fundamental difference is that the basic decarburization step, which is common to all steel making, is thermodynamically very difficult in stainless steel because the essential element, chromium, reacts more strongly with the purifying agent, oxygen, than does carbon. Thus, early stainless steel making, done in an arc furnace, was a lengthy process that necessarily involved high chromium losses to the slag as carbon was removed. This process was not only very expensive, the carbon levels that could be achieved were not much below 0.10%, making most of today’s stainless steels, whose carbon levels range from 0.010% in stabilized ferritic alloys to about 0.07% in normal

austenitic alloys, impossible to produce. The advent of AOD, continuous casting, ladle metallurgy, and powerful hot rolling mills has led to stainless steels of much higher quality produced at lower cost. Ironically, the low processing cost of stainless steel has spurred demand and made some of its ingredients, such as molybdenum and nickel, which are relatively scarce and expensive commodities, even more costly, forcing the cost of many alloys to spike even higher than in earlier years.

Melting and Refining The arc furnace is nearly universally used for the first step in the production of stainless steel. The arc furnace is quite flexible in the types of charge materials it can accept. Since the charge materials for stainless steel are typically carbon steel and stainless steel scrap, this flexibility allows scrap of all types to be used. The necessary chromium is added as ferrochromium, whose cost is inversely related to its carbon content. The carbon content of the heat of steel is roughly 1.5 to 2.5% when it is melted and ready to charge into the separate refining vessel. It is this carbon whose removal is the primary focus of refining. In the 1960s, Union Carbide engineers perfected a method, the previously mentioned AOD process, of removing nearly all the carbon from molten stainless steel without significant loss of chromium. This process is based on the following chemical reaction: Cr3O4 (Solid) + yC = yCO (gas) + Cr

(Eq 1)

The equilibrium for this reaction is: Ln (K ) =

−ΔG 4575T

(Eq 2)

156 / Stainless Steels for Design Engineers

where K is the equilibrium constant, and G is the Gibbs free energy. Working through the thermodynamics yields the relationship that summarizes the important relationship among carbon, chromium, and CO (Ref 1): Log =

%Cr 13,800 =− + 8.76 − 0.925 log p CO T %C (Eq 3)

Thus, increasing the temperature works to increase the elimination of carbon as CO, which evolves from the melt. This is similar in principle to the basic oxygen furnace (BOF) process for carbon steel in which oxygen is injected into molten steel to remove carbon by oxidizing it. The key to the AOD process, though, is the injection of oxygen and argon into the bath to keep the partial pressure of CO (pCO) very low. This is done at a temperature consistent with economic refractory life. The injection is done through tubes called tuyeres in the bottom of the barrelshaped vessel. The injection and the reaction cause extremely thorough mixing, which would never happen in the flat, stagnant, arc furnace bath. This mixing not only allows the CO-producing reaction to reach equilibrium, but also the mixing of the slag and metal also permits desulfurization. By increasing the ratio of argon to oxygen in the injected gas as the refining proceeds, the carbon is selectively oxidized without concurrent chromium oxidation. A typical starting ratio is 3 to 1 oxygen to argon/nitrogen by volume. The ending ratio can be as low as 1 to 9, oxygen to argon/nitrogen. The choice of which inert gas to use, argon or nitrogen, is based on cost and final nitrogen content desired. Stabilized stainless steels require low carbon and nitrogen levels, for instance, so the more expensive argon must be used. It is possible to use a vacuum system to keep the partial pressure of CO low when refining with injected oxygen. This is the vacuum oxygen decarburization (VOD) process. The VOD process can achieve slightly lower carbon levels but does not achieve cleaner steel as some believe. In both processes, after final carbon content has been achieved ferrosilicon is added to reduce the chromium in the slag and have it return to the molten steel. The excellent mixing of the slag and metal in the AOD permits this to be done efficiently. The silicon plus the manganese in the

steel combine to reduce the oxygen content of the steel to around 100 ppm. This could be further reduced by aluminum, but aluminum-based inclusions are generally undesirable. The thermodynamic activity of aluminum is considerably reduced in iron as chromium levels increase, so its role as a deoxidizer is less valuable in stainless steels. Titanium, on the other hand, is enhanced as a deoxidizer in chromium-iron alloys, and consequently small amounts of it are sometimes used as a supplementary deoxidant in alloys even though an alloy specification may not call for any. Titanium is believed to reduce hot working defects. More active deoxidants, such as calcium and magnesium, can be used when required. Also note that even if no intentional addition of metallic calcium is made, strong deoxidation with aluminum or titanium can reduce small amounts of calcium from the CaO in the slag, producing measurable calcium content in the metal. Besides carbon and oxygen, other impurities can be removed from the molten stainless. Once the steel has been deoxidized, sulfur can be readily removed by contact with a basic slag. Sulfur can be reduced to less than 0.001% in the AOD, and this excellent purity level is commercially furnished without additional price premium. Sulfur, although a harmful impurity from a corrosion standpoint, is often deliberately kept at moderate levels (0.008 to 0.015%) for tungsten inert gas (TIG) welding penetration (see Chapter 17) and at high levels (0.15%+) for machinability (see Chapter 15). These trade-offs, which are beneficial to processors, should be viewed with skepticism by end users, whose product integrity is compromised. There are processing methods for which higher levels of sulfur are not necessary that are preferable to the end user while not compromising welding or machining costs. For example, machinability can be improved by calcium additions that produce malleable oxides to replace the deleterious sulfides (see Chapter 15), and welding methods, such as laser welding, can be used in many cases to eliminate the need for the weld penetration enhancement of sulfur while increasing welding speeds. Phosphorus is an impurity for which no practical removal technology exists in stainless steel. Any known process to remove it first removes chromium. Thus, it exists in almost all stainless steel at levels close to its normal specification limit, about 0.030% in austenitic alloys and 0.020% or less in ferritic alloys, which are

Chapter 12: Melting, Casting, and Hot Processing / 157

on interstitial solubility. The higher solubility of carbon, nitrogen, and oxygen in stainless steels is significant. A manganese/silicon deoxidized stainless steel will still have about 100 ppm of dissolved oxygen at the freezing temperature as opposed to the less than 10 ppm of oxygen found in aluminum-killed carbon steel. This oxygen precipitates as oxides in the solid state. Vacuum induction melting (VIM) is another method of melting stainless steels. This is a nearly slag-free process, and little refining is possible. Melt purity is largely controlled by the purity of the starting material, and use of AOD master melt stock for VIM remelting is common. Limited decarburization is possible via injection of oxides such as Fe3O4 or SiO2 to create CO evolution inside the vessel. Using this technique, very low carbon levels (less than 50 ppm) are achievable commercially. Use of VIM is generally limited to high-value, high-purity, or low-tonnage melts.

made from a higher percentage of low-phosphorus carbon steel scrap. The deleterious effects of phosphorus on corrosion are not avoided unless much lower levels are achieved. Consequently, its presence is tolerated since it has no differential effect over the range in which it is found. Heavy metals are eliminated by high-temperature AOD blowing, as is hydrogen. Care must be taken not to reintroduce such impurities after refining, which is a risk when using damp or contaminated scrap for coolant. Alloy adjustment can be done in the AOD or preferably in a treatment-and-transfer ladle. The tapped molten steel generally has excess heat from the highly exothermic refining process. This allows the composition to be measured and adjusted before it must be cast. This can be done very precisely by wire feeding of alloying elements through the slag into the heat, which can be stirred by argon bubbling via porous plugs. This technique is very effective for the finetuning of reactive elements such as titanium. The refining treatments used for carbon steel and stainless steel are very similar, but there are subtle differences because of the difference in the thermodynamics of dilute solutions like carbon steel and highly alloyed, nondilute solutions like stainless steel. Table 1 shows the factors by which additions of various elements to stainless steel (j) alter the thermodynamic activity of other alloying elements (i). Equation 4 is used to calculate the activity of elements in steel. The activity coefficient γ varies with the concentration of alloying element x by: RT lnγ i = RT lnγ i0 +

n

∑

RT

j =1... n

δ ln γ i δxj

Remelting Some stainless steels and related alloys are remelted to refine composition or ingot structure. There are two principal remelt processes: vacuum arc remelting (VAR) and electroslag remelting (ESR). In VAR, the material to be remelted is cast into a cylindrical electrode and placed inside a cylindrical water-cooled vacuum chamber. A high-current direct current (dc) arc is established between the electrode and a starter plate at the bottom of the chamber. The end of the electrode is melted, and the molten drops fall through the intervening vacuum. Volatile constituents escape from the molten drops, and the purified drops collect to form a molten pool on top of the starter plate. VAR parameters are adjusted to maintain a shallow pool, which solidifies in a bottom-up fashion. The shallowness of the molten pool produces a refined grain

(Eq 4)

This calculation is best left to computer programs such as Thermo-Calc that have been perfected for these lengthy procedures. It should be noted that chromium, which is always present in nondilute quantities, has a powerful effect

Table 1 Influence of alloying elements on the thermodynamic activity of carbon, nitrogen, sulfur, and oxygen J i

Al

O N S 0

.04 –.03 ... –.39

C

Cr

Mn

.14 .13 .11 –.45

–.02 –.05 –.01 –.04

–.01 –.02 –.03 –.02

Mo

–.01 –.01 .003 .003

N

.11 0.0 .01 .06

Ni

.01 .01 0.0 .006

O

–.34 .05 –.27 –.20

S

.05 .01 –.03 –.13

Si

.08 .05 .06 –.13

Ti

W

... –.53 –.07 –.6

–.005 –.001 .01 –.01

158 / Stainless Steels for Design Engineers

structure with less solidification segregation than found in typical cast product. In ESR, the material to be remelted is cast into an electrode of similar shape, but slightly smaller than the water-cooled mold. A gap between the electrode and a starter plate at the bottom of the mold is filled with a prepared slag. Typically, this slag is calcium fluoride-based with high lime (CaO) content. Additional ingredients control the basicity, fluidity, oxidizing potential, and other properties of the slag. A high current is used to melt the slag, which in turn melts the end of the electrode, and the molten drops fall through the slag. Reaction of the molten drops with the slag removes sulfur and some other impurities, and the purified drops collect to form a molten pool on top of the starter plate. ESR melting typically is done at a higher rate than VAR, and the molten pool is deeper. This deeper pool produces a grain structure between that of VAR and typical cast product, with commensurate intermediate segregation patterns.

Casting Continuous slab, billet, and bloom casting have become the standard methods of making stainless steel primary products, replacing the obsolete ingot method. There are some alloys that cannot be continuously cast, but these represent a miniscule percentage of stainless production. Continuous casting produces slabs directly, thus removing the costly soaking and slabrolling processes. In a well-executed continuous casting operation, slabs are of sufficient quality that they require no surface conditioning before being hot rolled. Slabs range in thickness from 13 to 63 cm (5 to 15 in.). The segregation in continuous casters is less than in ingots because of the smaller section size. It is not eliminated, however, and certain alloying elements concentrate at the centerline, where they defy homogenization. Carbon and molybdenum are examples of alloying elements with this tendency. In properly executed continuous casting, the ladle feeds by a slide gate, or preferably a stopper rod gate, into a ceramic tube into the large tundish situated over the caster mold. The metal in the tundish is covered with a protective slag cover, and flow patterns within the tundish are designed to minimize dead spots and encourage removal of inclusions by impingement with the slag cover. The metal feeds through another

ceramic tube, called the submerged entry nozzle, into the mold, which is covered with a consumable protective and lubricating slag cover, called a mold powder. The mold powder, which melts in the mold as it is added, contains ceramics, fluxes, and carbon. The level of the molten metal should be carefully controlled by ultrasonic measurement, or other methods, to prevent fluctuations in level that may entrap slag in the slab surface. The entire water-cooled, copper alloy mold oscillates in a precise pattern as the solidifying strand of steel is withdrawn from the mold bottom by pinch rolls and sprayed with water to cool it. The pinch rolls apply enough pressure to slightly deform the slab. This deformation has a crucial, seldom-recognized effect. It causes a beneficial recrystallization that improves hot working characteristics of austenitic and duplex alloys. In ferritic alloys, it can cause excessive grain growth, which detracts from hot workability. The initial portion of slab cast in a sequence is seldom of adequate quality to be used because of exogenous inclusions, entrapped mold powder, and non-steady-state solidification structure. The defective portion must be identified and scrapped or diverted to low-quality requirement end uses. The strand is bent from an initial slightly curved shape to flat and cut into slabs. More than one heat of steel may be cast sequentially without restarting the process. This is ideal economically and for quality reasons since initial and final segments of a casting can contain more inclusions and aberrant structure. Some end users stipulate that no first slabs be applied to their orders. Producers generally apply first slabs to less-critical uses or discard suspect sections of them. If casting conditions are not optimal, the result can be slabs with poor surface quality that must be surface ground. Slabs are sometimes quenched to avoid precipitation of phases; however, they may be held at high enough temperatures prior to hot rolling to stay above the temperature range in which embrittlement can occur or to stay above the temperature at which an embrittled slab can fracture. Ferritic and martensitic alloys are especially prone to these problems. There has been great interest for decades in producing stainless steel coils directly from the melt in so-called strip casters. Elimination of hot rolling could be quite valuable in stainless steel, whose hot rolling from slab can be both expensive and problematic. There are a

Chapter 12: Melting, Casting, and Hot Processing / 159

number of such machines in pilot or limited production. They have not had sufficient commercial or technical success to have become a factor in the industry. Since their development is only being undertaken by those large stainless steel producers who already have the hot rolling assets that strip casting would replace, it seems unlikely that strip casting will soon become a major factor even if it is perfected technically. Another method of shortcutting the casting/ ingot step has been perfected: the powder metallurgy approach. In powder metallurgy, the refined molten metal is atomized by gas or liquid and made to freeze into small particles. These particles, having been quenched extremely rapidly, are quite homogeneous. Powder technology methods allow for the design of alloys that would otherwise freeze with too much segregation and too coarse a structure with conventional production methods. Traditional powder metallurgy production methods are used to make small near-net shape components, avoiding most of the costly machining steps. More impressively, powder technology is also used to produce massive components. For example, very high carbon/vanadium stainless tool steel components can be made by encapsulating powder in an evacuated canister in which it can be sintered and hot worked to 100% density and virtually complete homogeneity. Chapter 9 on martensitic alloys discusses these materials.

Hot Rolling Hot rolling remains an essential process for the vast majority of stainless steel used. Hot rolling characteristics of stainless steels vary greatly. Ferritic stainless steels are extremely easy to hot roll since they have a soft, singlephase structure at hot rolling temperatures. Martensitic stainless steels roll like their carbon and alloy steel counterparts since their microstructure during hot rolling is a moderately alloyed austenite similar to alloy steels. The microstructure during hot rolling is the crucial factor. Austenitic stainless steels have high strength at hot rolling temperatures. Furthermore, the low diffusion rates in austenite slow recrystallization so that the steel does not always soften between stands in tandem mills. This increases mill loads, and lower reductions must be taken than for alloy steels. Powerful hot strip tandem

mills that routinely roll carbon steel to 1.5 mm (0.06 in.) can struggle to attain 4.5-mm (0.18-in.) thickness for 316 stainless. The high separating forces on the hot rolling mill stands also cause greater roll deflection and compression, which if not countered by roll bending or roll shifting schemes can lead to significant variation in thickness across the sheet, as much as 0.25 mm (0.01 in.). This variation as a percentage of thickness is not reduced by cold rolling and is a major cause of tolerance loss in sheet and strip. Hot-rolled bands vary in thickness along the length of the coil because the tail end of the slab is colder and harder to roll. Coil boxes (on reversing mills) address this problem to a degree by permitting the semirolled coil to equalize in temperature. Hot strip tandem mills powerful enough to successfully roll high-quality stainless steel hotrolled bands are massively expensive and are seldom justified for the tonnage of stainless steel rolling a given melt shop produces, although rolling stainless on hot tandem mills used primarily for carbon steel can be an excellent production method. Hot Steckel mills have become the favored method of hot rolling stainless steel because their throughput better matches stainless steel melt shop production outputs. This permits the melt shop and caster to be adjacent to the hot mill, which permits energy-saving hot charging of slabs. In hot Steckel mills, typically a fourhigh reversing rougher rolls slabs to about 3-cm (1.2-in.) thick. Then, the transfer band is rolled to final gauge on a separate reversing four-high finishing mill with coil boxes to preserve temperature. The economy of having only two mill stands makes these mills ideal for typical stainless production quantities and permits the cost of sophisticated mill capabilities, such as roll shifting, roll crossing, or roll bending, not to have to be duplicated among many stands. This is the same justification for using Sendzimir mills to cold roll stainless. In both cases, the logic applies more to austenitic alloys than to the easily rolled ferritic stainless alloys. In either case, the hot-rolled band carries a heavy, embedded scale that must be removed from the surface before further processing in most cases. Some alloys can be cold rolled in the “black band” state at a cost of coarser surface finish and greater rolling mill roll wear. If normal cold rolling or use as hot-rolled coil is foreseen, the hot-rolled band must then be annealed and pickled since the as-rolled hot-rolled band

160 / Stainless Steels for Design Engineers

has poor corrosion resistance, poor mechanical properties, residual cold work and hardness variations, as well as a heavy oxide layer.

Defects Stainless steel hot-rolled bands can contain many types of defects. These are seldom seen by the end user because they are removed when they are not prevented. They do have repercussions on delivery. The major categories are: • Hot mill defects • Inclusion-related defects • Hot ductility-related defects Stainless steel is less forgiving of hot mill faults because its surface is not removed by oxidation to the degree carbon steel’s surface is. Thus, a skid mark from a slab-heating furnace will remain through the hot rolling, annealing, and cold rolling processes. This is true of all hot mill scratches, gouges, digs, etc. Rolling stainless requires a different mindset than rolling carbon steel, which argues against the benefits of rolling stainless on a mill built and primarily used for carbon steel. Inclusion-related defects are all essentially avoidable by using state-of-the-art technology. Protecting metal from reoxidation and keeping precise mold-level control in the continuous caster prevents all inclusions of a size that can produce a defect. Hot ductility defects are more subtle. They arise from many causes and are manifest prima-

rily as edge cracks and slivers. Edge cracks are simply a lack of ductility at the colder strip edge. Stainless hot ductility often has a narrow temperature window, and many factors can affect the size of that window depending on alloy type. The most inherently challenging alloys for hot working are the duplex alloys and the alloys that solidify in the fully austenitic state. The former has a mixed-phase structure, and the phases can exhibit mechanical incompatibility at certain temperatures. The latter alloys reject sulfur and oxygen during solidification and slab reheating to the grain boundaries, where they form very weak films. But, even alloys such as 304 and 316 can have very poor hot ductility if they contain much sulfur and oxygen or if they are reheated for long times or at temperatures above 1250 °C (2280 °F), which facilitates diffusion of sulfur and oxygen to the grain boundaries and also encourages very large grains. This poor hot ductility manifests itself as “slivers,” which can require grinding of the entire hot band surface. These tendencies are fought by low oxygen and sulfur levels and minimal slab-reheating temperatures and times, as well as slab surface working in the caster pinch rolls. Sometimes, very poor hot working alloys are given a single hot reduction pass on a hot mill to produce a full recrystallization that disperses grain boundary-weakening elements on subsequent reheat. REFERENCE

1. D. Peckner and I.M. Bernstein, Handbook of Stainless Steels, McGraw-Hill, 1977, p 3–13

Stainless Steels for Design Engineers Michael F. McGuire, p 161-171 DOI: 10.1361/ssde2008p161

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 13

Thermal Processing Summary THE THERMAL PROCESSING of stainless steel is a topic the end user should approach with great respect. It is not simple in concept or in practice. Before attempting to carry out any thermal processing on stainless steel, the practitioner must understand the alloy design, composition, and processing history of the material in question. The thermal processing then must be designed and executed in a planned, controlled manner. The consequences of failure in thermal processing can become catastrophic to mechanical properties and corrosion resistance.

Introduction The thermal processing of stainless steels can have many purposes. Normally, the objectives are simple: heating for hot working, annealing to soften after cold working, solution annealing to homogenize, heating to temper martensite, or to stress relieve. However, even if the objective is simple, the processes that occur are anything but simple. Variations in temperature, times at temperature, heating and cooling rates, and atmosphere can have complex and easily unintended consequences. There is no substitute for understanding the processes that are occurring when stainless steels are heated for successful heat treating to be achieved. Stainless steels have many alloying elements in large amounts. Many of these elements are highly reactive thermodynamically. The practical consequence of this is that many phases are thermodynamically possible at different temperatures. Stainless also reacts with its environment at high temperatures, causing changes in surface alloy content. Some of the resulting phases are

desirable, and some are potentially very detrimental. Readers are encouraged to review the earlier chapters on phases in stainless steel (Chap. 6–10) to familiarize themselves with these phases. Each of the alloy groups of stainless steels has radically different thermal processing objectives and requirements; therefore, each is discussed separately.

Austenitic Stainless Steels Thermal processes applied to austenitic stainless steels include: • Soaking for homogenization and preparation for hot working • Annealing to remove the effects of cold work and to put alloying elements into solid solution (solution annealing) • Stress relieving The temperatures at which these processes are carried out are shown in Table 1 for typical austenite compositions. Soaking Because virtually all stainless steel is continuously cast, the older soaking process of holding ingots in soaking pits for many hours is rarely used. The soaking had two functions. The obvious one was to equilibrate at the right temperature for hot working. The second, less-obvious, one was to achieve greater chemical homogeneity. The lack of homogeneity comes from the solute segregation that occurs as solute elements are rejected from the material that was first to freeze. Solute segregation was exaggerated by the slow solidification of ingots, and continuous casting helped make the stainless

162 / Stainless Steels for Design Engineers

Table 1

Recommended thermal processing temperatures for austenitic alloys

Alloy

Standard alloys 201, 202, 201LN 301, 301LN, all versions 304, 304L, 305, all versions 316, 316L, 316N, 317, 317L 308, 309, 309S, 310, 310S Stabilized alloys 321 347, 348 20Cb-3 Moderately alloyed, Creq30, high N AL6-XN, 4565, 654SMO, 254SMO Highly alloyed, sigma-prone alloys, Creq>30, low N AL6-X

Annealing temperature, °C

Annealing temperature, °F

ASTM A480(a) 2006, °F

Stress relieving, °F

Stress relieving, oC

...

...

...

1010–1120 1010–1120 1010–1120 1040–1175 1040–1175 ... 955–1065 980–1025 925–955 1120–1175

1850–2050 1850–2050 1850–2050 1900–2150 1900–2150 ... 1750–1950 1800–1950 1700–1750 2050–2150

1900 min 1900 min 1900 min 1900 min 1900 min ... 1900 min 1900 min ... Various

1500–1600 non-L, 1000–1600 L grades ... ... ... ... ... 1000–1600 ... ... 925–1010 1500–1600

1120–1175

2050–2250

Various

Not recommended

...

1205–1230

2200–2250

Various

Not recommended

...

815–870 non-L, 540–870 L grades ... ... ... ... ... 540–870 ... ... ... 815–870

(a) Standard specification for general requirements for flat-rolled stainless and heat-resisting steel plate, sheet, and strip.

steel more homogeneous. Nevertheless, cast slabs and blooms must be soaked to eliminate as-cast segregations. This process, to the extent it is done, occurs as they are reheated to the appropriate temperature for hot working. Soaking dissolves the few percent of residual delta ferrite that are present on slab solidification. It is important to soak at the highest temperature at which delta ferrite is not a stable phase so that it will dissolve, about 1250 °C (2280 °F) for most austenitic stainless steels. Soaking at higher temperatures causes ferrite levels to increase, negating the homogenization and causing very poor hot workability. Longer times at temperature than the minimum required for thermal uniformity also cause problems as any sulfur and oxygen impurities are rejected from austenite and can diffuse to grain boundaries, where they form weak, plastic films that also degrade hot workability. Grain growth, by reducing grain boundary area, exacerbates this effect. Thus, soaking times are best minimized and closely controlled. Alloys are therefore designed to have only a slight amount of delta ferrite to be redissolved during soaking. Ferrite is useful because it has a high solubility for oxygen and sulfur. Having none would result in impurity rejection of these elements to grain boundaries during solidification, which results in the worstpossible hot working characteristics. The oxygen and sulfur trapped in the ferrite during solidification precipitates in the solid state as inclusions, which also must be equilibrated with the sur-

rounding matrix by sufficient soaking. Welds that are unannealed have such precipitated inclusions in an unequilibrated state, and the result is diminished chromium concentration and poorer pitting resistance. Annealing Annealing serves two main functions in stainless steel: It removes the effect of cold work by replacing strained microstructure with new strain-free grains, that is, recrystallization. New grains nucleate and grow. If stored strain energy is insufficient, as happens often with ferritic stainless steels, true recrystallization is difficult to achieve, and the annealing process may only produce recovery without recrystallization, leaving the same grains relieved of stress. This leaves the surviving grains with the same crystallographic orientation that deformation produced and may or may not be the desired outcome. Second, annealing returns into solution solute that has been precipitated as unwanted phases, principally carbides, but also intermetallic phases. Annealing also may help to reduce solute segregation remaining from the casting process, making the composition more homogeneous. The homogenization process is accelerated by the reduction in dimensions from hot and cold working. A reduction in dimension by a factor of two reduces the time to achieve a given degree of homogenization by a factor of four.

Chapter 13: Thermal Processing / 163

Annealing to recrystallize is fairly rapid. To a first approximation, it is instantaneous, and the results are merely a function of the maximum temperature attained. This may not be the case for continuous annealing lines, in which transit time can be short enough, less than a minute at temperature, to limit the grain size attained. The driving force for recrystallization is the strain energy stored in the lattice from deformation. The strain energy in a given material is proportional to the square of the material’s flow stress. As the material is heated, recovery occurs first. This is the change in physical and mechanical properties associated with dislocation annihilation and polygonalization that occurs before the nucleation and growth of new grains. The nucleation of new grains occurs at highangle grain boundaries and proceeds by the movement of roughly hemispherical growth fronts into strained areas. The percentage recrystallized, once a sufficient temperature is reached, grows sigmoidally. It is normal for the time to fully recrystallize to be rather less than the time to attain that temperature. Even at the lower range of annealing temperatures, times are generally less than 1 min. Recrystallization will not occur if the lattice contains insufficient strain energy. Thirty percent cold work should be used as a rough threshold for the required amount. Annealing after lower amounts of cold work is characterized by scarce nucleation sites and can result in very large and nonuniform grain size. Hot-worked material often has a composite structure that may have already had some recrystallization depending on the final reduction temperature. Annealing may not produce a clear recrystallized structure in this case. The relative rapidity of recrystallization annealing is due to the fact that it is rate controlled by short-range diffusion. Solution annealing requires longer-range diffusion and thus can require much longer times. Some studies have shown that welds, for instance, do not recover completely from their loss of corrosion properties that arise from local alloy depletion unless they have been annealed for times on the order of 1 h (Ref 1). Others have seen homogenization in as little as 10 min (Ref 2). Wrought materials can require much shorter times because reductions during hot working have reduced diffusion distances. It should be noted that precipitates can be redissolved and not apparent in the annealed microstructure without full homogeneity being achieved. For example,

carbides can be redissolved and carbon diffused away from the carbide, but this does not mean that all composition gradients have been reduced to zero. This means that precipitates may re-form more rapidly in such a material than they would in a completely homogeneous alloy. The annealing temperature for a given alloy is chosen based on the temperature required to put all alloying elements into solution. Higher carbon levels, for instance, require higher temperatures to dissolve all the carbon. This is one of the principle values of accurate phase diagrams. If it were simply a consideration of recrystallization, all alloys could be annealed at similar temperatures at the low end of the recommended range. Within the recommended range, the temperature selected should be determined by the desired grain size. End use determines whether a fine or coarse grain size is preferable. Table 1 lists recommended annealing temperatures for austenitic stainless steels. The overall interplay between prior cold work and annealing temperature on mechanical properties of annealed material can be summarized as (Ref 3): • Grain size of a given alloy is the most important parameter in characterizing mechanical properties. • Yield and tensile strength are essentially constant for a given grain size regardless of the amount of prior cold work; however, the elongation depends on the prior reduction. • Yield strength, tensile strength, and hardness are essentially linear functions of grain size. • Elongation decreases with finer grain size and at an increasing rate as grain size becomes finer as long as cross-section size of the test specimen is not extremely small. This is not true for very coarse-grained material. • Maximizing elongation comes from maximum prior cold work and medium annealing temperatures • Anisotropy coefficients, or plastic strain ratios r are constant up to about 40% reduction after, which r45 and rn increase sharply, while rt decreases. This leads to earing during drawing. • The increase in properties for a one ASTM grain size increment is: a. 13 MPa (2 ksi) for tensile strength b. 20 MPa (3 ksi) for yield strength c. 2 HRB for hardness

164 / Stainless Steels for Design Engineers

Atmospheres for annealing are important. Austenitic stainless steels heated in air, of course, form oxide scales. Beneath this oxide, the metal matrix becomes significantly depleted of chromium (Ref 4), often more than 5% lower in chromium and to a depth of as much as 10 µ (395 µin.). So, not only must any oxide be removed, so must the chromium-depleted layer. This requires aggressive pickling, which while done commonly, may not be practical for many stainless users. The chromium-depleted zone, however, does pickle rapidly precisely because it does have less chromium. To avoid oxide scale formation, vacuum, hydrogen, or inert gas atmospheres may be used. If vacuum is used, it should be less than 2 × 10–3 torr (0.3 Pa). If an inert gas or hydrogen is used, the key consideration is moisture content. The dew point must be –40 °C (–40 °F) or lower. More stringent levels may be required if mirror finishes are desired. Cool down must be rapid as oxidation potential increases as temperature decreases. Vacuum or inert gas is preferable to hydrogen for alloys containing stable oxide formers such as aluminum or titanium or for alloys containing boron. Austenitic alloys that are subject to sensitization must be cooled rapidly enough from annealing temperatures to avoid carbide precipitation during cooling. If forced air or water quenching are impractical or if section size prohibits rapid cooling, then using stabilized or low-carbon grades is indicated. Superaustenitic stainless steels, and even alloys like 317, present a special problem because these alloys have significant sigma-forming tendencies. Sigma forms initially because solidification segregation causes local enrichment of sigma-promoting elements, such as molybdenum. It can also form from slow cooling of slabs or hot bands. This latter sigma forms at grain boundaries and will cause embrittlement and reduced corrosion resistance, so it must not only be redissolved, but also the alloy must be homogenized to remove the residual concentration gradients from the sigma. If this is not done, chromium- and molybdenum-depleted regions will still exist, and sigma will re-form much more rapidly during subsequent exposure to high temperatures. For this reason, the higher ends of the annealing ranges are recommended, and annealing times should be generous. Newer alloys have higher nitrogen contents to suppress formation of sigma and other deleterious inter-

metallic phases. Use of high-chromium andmolybdenum alloys without enhanced nitrogen is no longer recommended, and the use of lowernitrogen alloys should be reexamined and questioned if specified. Last, stainless surfaces should be scrupulously clean before annealing. Even hard water deposits can cause differential oxide growth, which can cause etched spots on the surface, where the postanneal pickling attacks the different oxide more strongly. Carbonaceous materials left on the surface are even more objectionable because they can cause carburization and subsequent loss of corrosion resistance. Stabilizing anneals are sometimes conducted on stabilized alloys such as 321 and 347. This is useful when carbon levels are sufficiently high that significant dissociation of carbides occurs at annealing temperatures. A second anneal at lower temperature, about 900 °C (1650 °F), then is done to permit the carbon to combine with the stabilizing element rather than leaving it available to form chromium carbides. Current preferred practice for these alloys is to maintain carbon and nitrogen below 0.03% for corrosionresistant service, which renders this stabilizing unnecessary. Alloys used for high-temperature service benefit from the creep-resisting contributions of higher carbon levels. Stress Relieving Austenitic stainless steel weldments often contain residual stresses, which can cause distortion or lead to stress corrosion cracking in service. They are commonly stress relieved at temperatures slightly below the annealing temperature, so that residual stresses may be relieved by creep. One hour at 900 °C (1650 °F) reduces residual stress by about 85%. Lower temperatures require exponentially longer times for the same stress relief, with times doubling for each 100 °C (180 °F) decrement as decreasing diffusion rates, which govern creep, are encountered. Cold-worked austenitic stainless steels have a markedly diminished proportional limit, particularly in compression. This Bauschinger effect, which arises from the easy mobility of dislocations, can be eliminated by stress relieving at around 350 °C (660 °F) for 2 h, which provides the thermal energy for dislocation interactions to lock into place. This produces a sharp yield point without premature nonproportional elastic deformation.

Chapter 13: Thermal Processing / 165

Ferritic Stainless Steels Ferritic stainless steels, from an annealing point of view, must be discussed in two categories. First are the modern, stabilized alloys, which are ferritic at all temperatures. These alloys behave as interstitial-free (IF) alloys because the interstitial carbon and nitrogen are removed from solution as a stable precipitate. In the second category are the older ferritic steels, which have enough austenitizing elements, usually carbon, in solution to cause them to form austenite at what would otherwise be a good annealing temperature. This makes them truly quasi-martensitic alloys, and they must be treated accordingly. Table 2, which lists heattreating temperatures for ferritic stainless alloys, also shows which grades fit into which category. Soaking Heating of ferritic stainless for hot working is straightforward. Whether stabilized or not, these alloys are heated to the 1000 to 1100 °C (1830 to 2010 °F) range for hot working. The superferritics can be heated to up to 1300 °C (2370 °F). At this temperature, no debilitating phases occur, and ductility is good. The high diffusion rate inherent to the ferritic structure makes homogenization easy. As long as hot working is completed at temperatures above that at which austenite forms, good hot ductility is expected. This is not a concern with IF alloys, which do not form austenite. Annealing The IF ferritics do not undergo any phase change with temperature during the course of properly executed heat treatment. The objective of annealing is generally simply to remove the effects of cold work. This is because they do not need to have carbon put into solution and, except in rare cases, do not have intermetallic Table 2 Recommended annealing temperatures for ferritic alloys Alloy

Stabilized, Cr+Mo20, 446

Annealing temperature, oC

Annealing temperature, oF

870–925

1600–1700

705–790

1300–1450

1010–1065

1850–1950

760–830

1400–1525

phases that require dissolution. Alloys with high chromium and molybdenum contents can form σ and/or α', the brittle, ordered body-centered cubic (bcc) phase, at temperatures below annealing temperatures, so rapid cooling is prudent when chromium plus molybdenum exceeds 20%. The driving force for recrystallization in these alloys is limited by the lower stored energy from deformation inherent in the bcc structure. In addition, the pronounced deformation texture leads to annealing responses that are more accurately characterized as recovery and grain growth with diminished recrystallization. These alloys retain this texture after annealing, and this characteristic anisotropy is exploited for good drawability. The major concern is to avoid excessive annealed grain size, which greatly reduces toughness. Anneal at the higher end of the range only if the loss of toughness associated with large grain size is not a concern. Stabilizing anneals are normally unnecessary for stabilized ferritics as their high diffusion rates ensure freedom from knife-edge attack due to sensitization from free unbonded carbon combining with chromium at grain boundaries. The stabilizing additions of titanium and/or niobium tie up the carbon as stable TiC or NbC, which does not redissolve during annealing. The interstitial-bearing ferritic stainless steels must be annealed subcritically, or the formation of austenite at higher temperatures would make martensite formation on cooling virtually unavoidable. Thus, a typical primary anneal cycle for a typical alloy such as 430 would be nearly 24 h at 750 °C (1380 °F), the majority of which is thermal equilibration of the large coil mass. The actual time at temperature required is less than 1 h. Continuous annealing is not practical because the diffusion of carbon is too slow to occur in the dwell time at temperature typical in continuous annealing lines. This cycle also precipitates essentially all the carbon and nitrogen as mixed Cr/Fe carbides and nitrides and homogenizes chromium content. This necessarily slow process permits subsequent subcritical annealing for mechanical properties (to alleviate the effects of cold work) to be done in a few minutes since carbon has been eliminated from solution by the formation of fairly stable carbides. Since the material is generally purchased in the annealed condition, the user need never be concerned with such lengthy anneals.

166 / Stainless Steels for Design Engineers

Stress relieving is rarely a concern for any type of ferritic stainless. Unstabilized grades should not be welded, and if they are, full subcritical annealing is required. Stabilized grades have no need for postweld heat treatment. Lowtemperature heat treatment runs the risk of α' formation and is best avoided.

Martensitic Stainless Steels The martensitic stainless steels resemble the unstabilized ferritic stainless steels described. The martensitic stainless steels form essentially 100% austenite on heating and have very high hardenability, so their ability to be softened by annealing is limited. The traditional martensitic stainless steels are iron/chromium/carbon alloys, sometimes with a small amount of nickel and/or molybdenum. More recently, alloys have been developed for petroleum applications that contain high copper, nickel, and/or molybdenum and low carbon. The principles of heat treatment of the two alloy categories are the same. The more highly alloyed newer alloys are, in fact, simpler to heat treat because their low carbon and nitrogen levels alleviate the need to temper.

martensite cannot be avoided by furnace cooling from austenitic temperatures, then only subcritical annealing is feasible. But, even for nickel-free alloys the hardenability is so great that annealing by slow cooling is quite difficult. Martensitic alloys are put into the annealed condition for processing before they are quenched and tempered for their final use. Thus, the more economic subcritical anneal is the predominant annealing heat treatment. The nickel-bearing alloys have such high hardenability that annealing in the critical range cannot produce softening by any practical cooling rate, so subcritical annealing is always recommended for these alloys. Nickel reduces the temperature at which austenite is stable as shown in Chapter 9, Fig. 9. Other additions like vanadium, molybdenum, and tungsten promote secondary hardening and tempering resistance, and subcritical annealing of these alloys becomes a slow, difficult process. This is a characteristic of the so-called super 12Cr alloys. Martensitic alloys have lower corrosion resistance in the annealed condition than in the hardened condition because in this state they have the maximum amount of chromium tied up as chromium carbide. Austenitizing

Soaking Hot working should be carried out in the austenitic range. Temperatures for this are listed in Table 3. Forging and hot working should always be followed by annealing to avoid stress cracking due to the deep hardening of these alloys. Annealing Martensitic stainless steels can be annealed by subcritical anneal and sometimes by full anneal depending on alloy level. If the alloy level is such, as in the nickel-containing grades, that Table 3

Table 3 lists the austenitizing and tempering ranges for martensitic stainless steels. Full austenitizing is crucial to producing a microstructure that is fully martensitic. Only austenite transforms to martensite. If other constituents, such as δ ferrite or carbides, exist during the austenitizing heat treatment before quenching, they will not transform to martensite. Some alloys, such as the 440 group, have enough carbon that the austenitizing temperature determines how much carbon is put into solution. The carbon in solution in the austenite will become the carbon level in the martensite, which directly determines strength and corrosion

Recommended annealing, austenitizing, and tempering temperatures for martensitic alloys

Alloy

Subcritical anneal, o C (oF)

Full anneal, o C (oF)

Straight Cr, C1.5 mm, 0.06 in.) material or when the HAZ is drastically reduced, as in laser welding. The high thermal expansion of austenitic stainless steel can cause high residual stress around welds, which may require annealing to eliminate. Another serious threat posed by thermal stresses is hot cracking. This can occur to material that has just solidified when geometric constraints to contraction imposed by the surrounding material imposed act on weak grain boundaries. This weakness occurs when the steel solidifies in an austenitic mode. When austenite freezes, it strongly rejects sulfur to the intergranular areas, where it forms weak films. This is solved by balancing the composition so that alloys solidify first as ferrite, which does not reject the sulfur, forcing it to precipitate as sulfide inclusions within the grains. This approach is highly effective but cannot be used for

Fig. 1

The Schaeffler diagram. Source: Ref 1

some highly alloyed grades with compositions that do not permit a ferritic solidification mode. In such alloys, sulfur and other contaminants, such as phosphorus, oxygen, zinc, and copper, must be excluded from the weld zone. Welds of less highly alloyed austenitics, generally those with less than 20% chromium, which are balanced to freeze in a ferritic mode, retain some ferrite at room temperature, normally between 3 and 10%. This is not harmful since the ferrite is richer in chromium and in molybdenum, if present. The amount of ferrite expected can be measured by magnetic devices and estimated from the Schaeffler diagram, a useful empirical mapping of weld metal phase composition shown in Fig. 1. This diagram has an arbitrary cooling rate resembling that of tungsten inert gas (TIG; described in a separate section of this chapter) welds. Faster or slower cooling will change the relative amounts of ferrite and austenite because of the need for diffusion to achieve the most stable phase balance. Very rapid cooling, as with laser welding, tends to make austenitic welds less ferritic and has the opposite effect in duplex alloys. The Schaeffler diagram has been improved by the Welding Research Council’s adoption of the modification shown in Fig. 2, which super-

Chapter 17: Welding / 203

Fig. 2

Welding Research Council’s (WRC’s) 1992 constitution diagram

imposes the solidification mode as a function of the composition. The crucial line on this diagram is dotted-dashed line AF, which delineates those compositions that solidify in a primary ferrite mode, precluding the problem of intergranular solidification cracking. Another problem particular to the more highly alloyed grades is the formation of intermetallic phases from long cumulative exposure to temperatures in the 600 and 900 °C (1110 and 1650 °F) range (coincidentally, the same as for carbide precipitation). The slow diffusion of alloying elements in austenitics makes this a lesser problem than in ferritics or duplex. This adverse precipitation is largely prevented in the modern, nitrogen-alloyed grades, so these alloys are recommended if extensive welding is planned. The more highly alloyed grades also suffer from greater microsegregation during solidification. This causes austenitic dendritic cores to have lower chromium and molybdenum content and consequently lower corrosion resistance. Thus, the welds have lower resistance to localized corrosion. This is addressed by using more highly alloyed filler metal or by solution annealing the welds.

Restricting heat input to under 16 kJ/mm (400 kJ/in.) and interpass temperature to under 150 °C (300 °F) helps to minimize each of these risk factors inherent to the more highly alloyed austenitic grades. Note that the influence of microsegregation of alloying elements is separate from and in addition to the negative influence of sulfur on the corrosion resistance of welds. Austenitics at the alloy level of 316 and above should not have sulfur above 0.001% for these alloys to deliver the expected corrosion resistance. The austenitic stainless steel weld metal composition can be altered by the gases to which the molten base metal is exposed. Lack of shielding can lead to oxygen combining with chromium and other elements, creating slag and depleting the alloy of needed elements. Thus, oxygen-free gas mixtures are used to exclude the ambient atmosphere from the molten pool during electric arc welding. Inert gases provide the barrier, while the addition of 3 to 5% by volume of nitrogen gives the necessary partial pressure to ensure that welds will not be depleted of vital nitrogen content. Figure 3 (Ref 1) shows the influence of nitrogen content of the shielding gas

204 / Stainless Steels for Design Engineers

Fig. 3

Effect of weld shielding gas composition on crevice corrosion resistance of autogenous welds in AL-6XN alloy tested per American Society for Testing and Materials (ASTM) G-48B at 35 °C (95 °F)

on corrosion resistance of a highly alloyed austenitic grade. Excess nitrogen in the shielding gas (e.g., more than 10%) can cause porosity in the weld, and greater than 5% is detrimental to the life of the tungsten electrode. The heat from welding can produce a surface oxide composed mainly of iron and chromium. The underlying surface can be significantly depleted of chromium because of the loss of chromium to this scale and therefore significantly lower in corrosion resistance. Pits can start in this thin layer and propagate into sound metal beneath. For heat-tinted surfaces, the darker the tint, the stronger will be the effect. To fully restore corrosion resistance, the area must be ground to remove the oxide and any depleted base metal. This should be followed by acid pickling, which completes the removal of the oxide and depleted zone. Duplex stainless steels differ from austenitic stainless steels in their metallurgical response to welding mainly because their approximately 50% ferrite causes greater thermal conductivity at lower temperatures, and ferrite has greater diffusion rates. These alloys solidify in a completely ferritic mode, and since ferrite rejects little sulfur on solidification, hot shortness is not a problem. So, compared to austenitic stainless steels, duplex stainless steels have the following distinguishing factors: • The ferritic solidification mode of duplex stainless steels provides very good hot cracking resistance. The rapid cooling of welds produces welds and HAZ with more ferrite than the parent metal by quenching in the high-temperature ferrite.

• Duplex alloys are more sensitive to problems in the HAZ because their generally high chromium and molybdenum content plus their ferritic content make the precipitation of embrittling intermetallic phases more rapid than in austenitics, so minimizing the total time at high temperature is the overriding concern. • While carbide sensitization is not an issue with the duplex alloys, the formation of intermetallic phases can cause loss of corrosion resistance. • Duplex, like all stainless types, must be protected from oxidation by shielding gas, and since nitrogen is a crucial alloying element, especially in duplex alloys, it must be a component of the gas mixture. • Cleaning before and after welding is equally important in duplex as in austenitics. Modern duplex alloys derive their impressive strength, toughness, and corrosion resistance from their nearly equal percentage of ferrite and austenite. The nitrogen content of the austenite brings its corrosion resistance up to that of the ferrite phase, which is richer in chromium and molybdenum. Nitrogen additions partition to the austenite and thus both strengthens it and increases its corrosion resistance to close to that of the ferrite. The early duplex alloys had a tendency to form excessive ferrite when welded and formed embrittling intermetallic phases rather rapidly. The additions of larger amounts of nitrogen stabilized the austenite to higher temperatures, so welds did not become so ferritic. The nitrogen also decreased the speed at which intermetallic phases form, enlarging the time window for welding without their precipitation. And, by promoting greater austenite formation at high temperature, the addition of high (>0.12%) nitrogen actually reduces the tendency for chromium nitride precipitation. Despite these advances, the key precaution in welding duplex alloys is to prevent the formation of embrittling phases while preserving as close to a 50/50 austenite/ferrite structure as possible. Minimizing time at red heat temperatures (500 to 900 °C, 930 to 1650 °F) is the objective. But, sufficient time must be spent above about 1000 °C (1830 °F) to promote the formation of sufficient austenite. If the weld cannot be annealed, increased nickel filler metal (e.g., 2209 with 2205 base metal) should be used. Thus, joint preparation must be done correctly and not left to the welder to correct using timeconsuming remedial procedures.

Chapter 17: Welding / 205

Duplex stainless steels, because of their moderate thermal expansion and higher thermal conductivity, can tolerate relatively high heat inputs since these factors determine the stress intensity that will be generated by thermal gradients. However, excessively low heat inputs can result in fusion zones that are predominantly ferritic, with a resultant loss of toughness and corrosion resistance. At the other extreme, heat inputs that are too high lead to the formation of embrittling intermetallic phases. This issue concerns the HAZ, which must dwell in σ-forming temperatures for some period of time. The key is to limit the time at those temperatures by not permitting interpass temperatures to exceed 150 °C (300 °F) because workpiece temperature has the greatest influence on time at σ-forming temperatures. It is prudent to impose this limitation when qualifying the weld procedure and then monitoring the production welding interpass temperature electronically to ensure qualifying procedures are not more lenient than are those of production. Postweld stress relief is not needed for duplex weldments and indeed could be harmful because of the danger of embrittlement. Full annealing can be done and can restore the original phase balance and composition that gives the optimal toughness and corrosion resistance found in wrought material. Ferritic stainless steels can be split into two groups for purposes of welding: the older semiferritic group and the more prevalent stabilized ferritic group. The first group, in which chromium is between 16 and 18% with carbon up to 0.08%, is exemplified by the alloy 430. These alloys form appreciable amounts of austenite when heated above 800 °C (1470 °F). Unless they are cooled extremely slowly, more slowly than can be done in welds, the austenite transforms to martensite, which is very brittle. The stabilized grades commonly use titanium or niobium to combine with the carbon and nitrogen, which otherwise would cause the formation of the high-temperature austenite, rendering the alloys ferritic at all temperatures. The salient metallurgical characteristics for welding of the two groups are: • Both groups offer good thermal conductivity and low thermal expansion. • Both groups require protection from oxidation by shielding gases. The stabilized group should not be exposed to nitrogen. • The semiferritic group will form martensite, which requires annealing to eliminate.

• The stabilized group can lose toughness via excessive grain growth. • The grades more highly alloyed with chromium and molybdenum can form α' and σ, leading to embrittlement. The semiferritic alloys such as 430, 434, and 436 are seldom welded and often called unweldable. The reason is that the welds are invariably partially martensitic and thus normally brittle. Only very specially controlled compositions of 430 can be welded successfully, and these are not generally available commercially. While the technical remedy for this is simply annealing, it is seldom economically viable. It is rare to see any welding more extensive than spot welding of unexposed surfaces with these alloys. If for some reason they must be used and welded, then the techniques for welding martensitic stainless steels should be employed. The stabilized ferritic stainless steels are commonly welded. The levels of stabilizing elements required to prevent austenite formation and sensitization are well known and are reflected in the alloys’ chemistry specifications. For 409, the required titanium level is Ti > 0.08 + 8(C + N), while the requirement for the higher chromium 439 is 0.20 + 4(C + N). These are empirical relationships that take into account that some titanium oxidizes before it can stabilize carbon and nitrogen. Niobium can replace some titanium. This is discussed in detail in Chapter 8 on ferritic stainless steels. Because of the low toughness these alloys have in large cross sections, these alloys are only rarely seen with minimum section size of more than 3 mm (0.11 in.) and normally have sections less than 2 mm (0.08 in.). Thus, successful welding is simplified to making a sound, well-shielded weld without producing excessive grain growth in the HAZ. In practice, this can be achieved by limiting heat input to less than 6 kJ/cm. An empirical relationship between grain diameter D and heat input E (kJ/cm) has been reported (Ref 2). In the fusion zone, the relationship is: D = 206 × E – 585.6

(Eq 1)

In the HAZ, it is: D = 29.6 × E – 50.6 for up to 6.6 kJ/cm

(Eq 2)

and D = 75 × E – 350 above 6.6 kJ/cm

(Eq 3)

The light gauges ensure sufficiently short times at high temperature that precipitation of

206 / Stainless Steels for Design Engineers

intermetallic phases should not be a concern, even though they can form, especially in superferritic alloys. The impact properties of ferritic stainless steels are always a concern because their transition temperature can become elevated to ambient levels. It has been determined that there exists an optimum level of titanium around 0.10%, which ensures this minimum transition temperature (Ref 3). Because it is difficult to have low enough carbon plus nitrogen to stabilize at this titanium level, dual stabilization with titanium and niobium as well as not having excessive heat input are the best way to ensure weld toughness. Especially in the superferritics, maintaining the benefits of having the fairly precise balance of carbon plus nitrogen to the stabilizing elements titanium and niobium requires that neither carbon nor nitrogen come into contact with the weld pool. Likewise, oxygen must be rigorously avoided because it will quickly deplete the essential titanium, which is even more readily oxidized than chromium. Extraordinary surface cleaning at and near the weld will pay dividends in final quality. Martensitic stainless steels vary little in alloy content, ranging from 11 to 18% chromium with small amounts of nickel and molybdenum. Their carbon content ranges from 0.10 to over 0.30%. Thus, the major challenge they present is avoiding the potential cracking, which is most likely to occur in the HAZ from stresses caused by the austenite-to-martensite transformation on cooling. Since this transformation cannot be avoided, the desired approach is to start with a well-tempered or annealed material and then preheat and maintain high interpass temperatures. For low carbon levels, below 0.10%, preheat can be omitted, but between 0.10 and 0.20% carbon, preheating to 250 °C (480 °F) is advised and for higher carbon levels, 300 °C (570 °F). The problem becomes more severe with increasing carbon level because the transformation takes place at lower temperatures in more brittle material. Even with preheating, distortion may be encountered. For all normal uses of martensitic stainless steels, a final heat treatment is required to achieve the quenched and tempered properties for which these alloys are designed. Aside from the cracking consideration, martensitic welding considerations are similar to, but less stringent than, those of low-alloy stabilized ferritic stainless steels with regard to

cleanliness and shielding. If mechanical requirements permit, the use of austenitic (309L) weld filler metal should be considered. The soft joint may deform to accommodate thermal strains and thus minimize weld cracking. Precipitation-Hardening Stainless Steels. Last, precipitation-hardening (PH) stainless steels, while very complex metallurgically, are straightforward from a welding perspective. Obviously, any heat treatment to achieve the properties of which these alloys are capable must be a final step. The considerations in welding them are: • Shielding must be sufficient to prevent loss of oxidizable alloying elements such as titanium, aluminum, and, of course, chromium. • Filler metal must match the base metal if like properties are required. • Postweld heat treatment solution annealing must be adequate to homogenize weld solidification segregation. • Austenitic PH grades are fully austenitic and subject to hot short cracking. • The high aluminum or titanium contents of many PH alloys cause their welds to be “slaggy,” and these slaggy welds have are irregular with objectionable recesses, crevices, or prominences. These alloys are easily welded and not prone to cracking or developing embrittling phases. But, because these alloys are designed for extreme mechanical performance, it is essential to preserve their correct chemistry by shielding with a fully inert gas mixture. If mechanical properties equal to that of the base metal are not required in the weld, then austenitic filler, such as 309L, can be used. Table 1 summarizes the major metallurgically important parameters for the various types of stainless alloys. It is prudent to consult with the manufacturer’s data sheets for specific recommendations on alloys that they produce as they are often privy to test data and user experience that cannot be found elsewhere in the literature.

Material Selection and Performance Stainless alloys that are prone to precipitation of intermetallic phases require special prewelding consideration. Such alloys include duplex, superferritic, and superaustenitic alloys. Any amount of time for which these alloys have been exposed to temperatures at which inter-

Chapter 17: Welding / 207

Table 1

Welding parameters for various stainless steels

Alloy group

Filler

Heat input kJ/cm(max)

Shielding gas

Preheat

Interpass max

Postweld heat treat

Austenitic

...

20–40

Ar+2% O2, Ar/3% CO2/2% H2 He+7.5%Ar+2.5 CO2 Same Same Same Same Same Same Argon/helium or argon + 3–5% N2:no O2 Argon/helium

150 oC

150 oC

None or full anneal

301, 302, 304 304L 309 310 316L, 316Ti 321, 347 Superaustenitic

308, 308L 308L 309, 310 310 316L, 317L 347, 308L 22, 675, 276

Same Same Same Same Same Same 16

... ... ... ... ... ... 50 oC

... ... ... ... ... ... 100 oC

... ... ... ... ... ... None or full anneal

PH grades

Same as base alloy

20–40

no

...

Full solution anneal

Martensitic 410

410, 308, 309L

20–40

Ar+2% O2, He+7.5%Ar+2.5 CO2

250oC

250 oC min

Slow cool

420

420, 308, 309L, 310

20–40

Ar+2% O2, He+7.5%Ar+2.5 CO2

250 oC

250 oC min

Anneal

440

440, 308, 309L, 310

20–40

Ar+2% O2, He+7.5%Ar+2.5 CO2

250 oC

...

...

Supermartensiic

Same as base metal

20–40

Argon/helium

no

...

Full solution anneal

430

430, 309L

20–40

Ar+2% O2, He+7.5%Ar+2.5 CO2

no

...

Subcritical anneal

434

309 Mo L

20–40

Ar+2% O2, He+7.5%Ar+2.5 CO2

no

...

Subcritical anneal

409

410L, 308, 309L

6.0

Ar+2% O2, He+7.5%Ar+2.5 CO2

no

n.a.

none

439

439L, 309L, 316L

6.0

Ar+2% O2, He+7.5%Ar+2.5 CO2

no

n.a.

none

Superferritic 2003, 2101, 2304, 19-D 2205

29-4C 2209

6.0 5–25

Argon/helium Argon + 3% N2

no no

n.a. 150 oC

None or full anneal None or full anneal

2209

5–25

Argon + 3% N2

no

150 oC

None or full anneal

25 Cr duplex 2507 superduplex

25Cr-10Ni-4Mo-N 25Cr-10Ni-4Mo-N

5–25 5-25

Argon + 3% N2 Argon + 3% N2

no no

150 oC 150 oC

None or full anneal None or full anneal

Ferritic

PH, precipitation hardenable

metallic phases form without full subsequent homogenization anneal is time that the welder cannot use to complete a satisfactory weld before precipitation occurs. Thus, accurate knowledge of material history is vital. Likewise, variations within specification of nitrogen content influence the time it takes intermetallic phases to form. Once a welding procedure is qualified for an alloy with given nitrogen content, use of lower nitrogen alloys would not be prudent engineering practice. Austenitic stainless steels that are intended for autogenous welding are often specified with elevated sulfur levels, on the order of 0.005 to 0.015%. This is done to improve weld penetration through the so-called Marangoni effect.

This effect exploits the temperature-dependent surface concentration of sulfur in the weld pool, which causes a decreased surface tension toward the hotter center of the pool, causing the molten pool to flow toward the center on the surface and then flow downward, shooting the hottest metal to the bottom of the weld pool, as shown in Fig. 4. This speeds welding and minimizes weld and HAZ width, which is a good thing. The effect on corrosion resistance is less desirable since the abundant MnS inclusions that result from the higher sulfur levels decrease pitting resistance. This decrease in corrosion resistance can only be eliminated by a long anneal. Unfortunately, the pipe purchaser cannot know if the pipe has had a sufficient anneal.

Metal flow directions in a weld pool with (left) and without (right) sulfur. Source: Adapted from Ref 4

In-line induction annealing is insufficient for this purpose. Furnace anneals of about an hour are required. For alloys like 304L and 316L, the user should always require material chemistry certifications and assume that any sulfur levels above 0.003% are going to result in decreased pitting resistance of 1 to 5 PREN (pitting resistance equivalent number), which means up to 10 °C (18 °F) decrease in critical pitting temperature, roughly the difference between 304 and 316 in performance. This also applies to girth welds done by the pipe user. Welds are essentially a casting in the midst of wrought material. In addition to inclusions decreasing weld corrosion resistance as mentioned, solidification segregation can also cause microscopic regions to be poorer in corrosionresisting alloying elements chromium, molybdenum, and nitrogen. This effect is minimal for low-alloy material, but for highly alloyed austenitic grades, it is a major effect, as shown in Fig. 5. Eliminating this effect requires a thorough homogenization anneal. The use of filler metal with higher corrosion resistance does not totally offset the influence of welding on corrosion resistance because some of the base metal melts and is not altered in composition by the filler metal. This is called the unmixed zone. It is essentially a zone with properties equal to that which would occur in an autogenous weld, that is, the corrosion resistance is lower depending on total alloy level and sulfur content.

Welding Processes All stainless steels should be very clean prior to welding. The chemistries of both base metals and filler metals are carefully formulated to produce the mechanical and corrosion properties that these alloys have been designed to produce. Virtually any contaminant can either interfere with the welding procedure or detrimentally

90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 −5

Unwelded Welded

1

Fig. 5

194 185 176 167 158 149 140 131 122 113 104 95 86 77 68 59 50 41 32 23

2

3 4 5 6 Molybdenum, wt%

Critical pitting temperature in 6% FeCI3, °F

Fig. 4

Critical pitting temperature in 6% FeCI3, °C

208 / Stainless Steels for Design Engineers

7

The influence of molybdenum on critical pitting temperature. Source: Adapted from Ref 5

alter the composition of the welded joint, which in turn can alter corrosion and mechanical properties and compromise the entire structure. Moisture, paint, dirt or grease, oil, and oxides all can negate good material, good welding technique, and good procedural qualification. Cutting fluids, especially sulfurized oils, are especially detrimental and should be removed completely prior to welding. Preheating is never strictly forbidden since it is required to eliminate moisture. Joint design does not differ in principle from that of other steel weldments. There is, however an increased need for dimensional uniformity for the alloys susceptible to intermetallic precipitation since minimizing time at temperature is a priority, and variations in joint geometry impede the swift completion of the weld. This is also true for alloys that are susceptible to excessive grain growth, such as the stabilized ferritics, or to sensitization. Figure 6 shows some joint designs appropriate to stainless steels, including the more sensitive alloys. These, like all joint designs, aim to ensure full penetration without burn through. Gas tungsten arc welding (GTAW)/tungsten inert gas (TIG) is commonly used for the automated production of stainless steel pipe and tube, as well as manual short runs. It is versatile and generally used when thicknesses are less than 6 mm (0.2 in.). It can produce very highquality welds. A constant-current power supply is preferred. It is best performed with the DCSP (direct current straight polarity) electrode negative

Chapter 17: Welding / 209

Groove

t

Process

Thickness th, mm (in.)

Gap d, mm (in.)

Root K, mm (in.)

GTAW

3–5

1–3

...

...

GMAW

3–5

1–3

...

...

SMAW

3–4

1–3

...

...

SMAW

4–15

1–3

1–2

55–65

Bevel α(°)

d

a

GTAW

3–8

1–3

1–2

60–70

GMAW

5–12

1–3

1–2

60–70

SAW

9–12

0

5

60

SMAW

>10

1.5–3

1–3

55–65

GMAW

>10

1.5–3

1–3

60–70

SAW

>10

0

3–5

80

SMAW

>25

1–3

1–3

10–15

GMAW

>25

1–3

1–3

10–15

SAW

>25

0

3–5

10–15

GTAW

>3

0–2

...

...

GMAW

>3

0–2

...

...

SMAW

>3

0–2

...

...

SMAW

3–15

2–3

1–2

60–70

GTAW

25–8

2–3

1–2

60–70

GMAW

3–12

2–3

1–2

60–70

SAW

4–12

2–3

1–2

70–80

SMAW

12–50

1–2

2–3

10–15

GTAW

>8

1–2

1–2

10–15

GMAW

>12

1–2

2–3

10–15

SAW

>10

1–2

1–2

10–15

d

a

k

d

a

k

d

r = 6-8mm

t d

a

d

a

d

r = 6-8mm

GMAW, gas metal arc welding; GTAW, gas tungsten arc welding; SAW, submerged arc welding; SMAW, shielded metal arc welding

Fig. 6

Joint designs. Courtesy Ugine S.A.

technique. It is helpful to incorporate a highfrequency circuit to aid in establishing the arc. Thoriated electrodes containing 1.7 to 2.2% thoria are recommended because they have better emissive properties and provide better arc stability at higher currents. If consumable electrodes are used, the shielding gas precludes the need for coatings. The weld metal alloys are not necessarily the same as the parent alloys but are chosen based on their ability as weld metals to provide the most acceptable corrosion and mechanical properties. This sometimes means using austenitic filler with a ferritic base or higher nickel content in an austenitic or duplex base to compensate for the solidification rate or inherently lower corrosion resistance of the weld.

The shielding gas must replicate the controlled gas mixtures used to refine stainless steel and establish the original composition. The weld pool exposes a great deal of surface area to the atmosphere in a very turbulent manner. Gas flows, usually 12 to 18 L/min, must be adequate to prevent air infiltration by aspiration or turbulence before arc contact, ideally until temperatures cool to below oxidation temperatures. For manual GTAW using a filler wire, the wire should be fed continuously into the weld pool. Intermittent wire addition can lead to creation of zones of essentially autogenous weld, negating many of the benefits of filler metal addition. Moving the tip of the wire in and out of the protection of the gas shield is especially

210 / Stainless Steels for Design Engineers

bad. The hot tip can carry oxides and nitrides into the weld, defeating the action of the shield gas and impairing weld quality. Gas metal arc welding (GMAW) is arc welding in which a consumable electrode provides larger amounts of filler weld metal than practical in GTAW. There are three GMAW techniques: • Pulsed arc transfer • Spray transfer • Short-circuiting transfer Pulsed arc transfer employs a power source that is switched rapidly to provide transfer of weld metal droplets at regular intervals. Spray transfer uses a high current to form a stream of fine drops from the end of the electrode. This is done with high power, resulting in a large fluid weld pool, and therefore limits the technique to horizontal orientations and thick material. Short-circuiting transfer uses arc contact with the workpiece at low power to melt the electrode, after which the short circuit is broken, and material transfer ceases. The technique creates a minimal weld pool and is viable in many orientations. It is a low-heat process suitable for thin material but may cause lack of penetration defects if used for thick-section welding. For all GMAW processes, excessive protrusion of the wire should be avoided; otherwise, the full benefit of the inert gas shielding may be lost. Submerged arc welding (SAW) employs a consumable electrode immersed in a conductive flux that acts as a protective shield from the atmosphere. The arc is struck through the flux, and gravity deposits the molten metal to the workpiece. The large weld pool has high heat input and can deposit large amounts of metal relatively quickly. Thus, SAW may be preferable to multipass techniques for alloys such as duplex for which time at temperature is limited. It is restricted to horizontal orientations and requires postweld slag (flux) removal. Shielded metal arc welding (SMAW) is done manually with short lengths (“sticks”) of coated electrodes. This method has great versatility with some trade-off in cost and quality. This last aspect is arguable, but the lack of shielding gas may introduce oxygen to the weld metal, which can be detrimental to toughness. Flux cored wire (FCW) welding is a method that is able to accommodate a large range of thickness and orientations while providing high deposition rates. The equipment is the same as

for GMAW, but the consumable electrode, the FCW filler metal, has a flux core that supplements the shielding gas. Because of the flux, the shielding requirements are reduced; gases can be argon/25% carbon dioxide for horizontal welding with current and voltages from 150 to 200 amp and 22 to 38 V, respectively. Vertical welds can use 100% carbon dioxide with amperage of 60 to 110 amp and voltage of 20 to 24 V. Flow rates of gas are 20 to 25 L/min. It is possible to get high-carbon welds, which may not resist corrosion as well as desired, so as always, weld qualification, including corrosion evaluation, is critical. Oxyfuel gas welding (OFW), “torch” welding, uses oxygen to accelerate fuel (typically acetylene) combustion to produce temperatures that can melt steels. By controlling the fuel-air mixture, the flame can be made nonoxidizing for low-alloy steels. However, these “neutral” flames can simultaneously oxidize and carburize stainless steels. Thus, the OFW process is not suitable for use with stainless steels. Laser welding has become a major production method when it can be automated, as for pipe and tube or high-production manufactured items, such as air-bag canisters. Metallurgically, it resembles resistance welding in that both have minimal HAZ and very high quenching rates, both of which can have a pronounced effect on some types of stainless steel. The effect is to undercool the molten metal and suppress the transformation that would normally occur. So, an austenitic alloy that normally solidifies in a ferritic mode before transforming to austenite freezes directly as austenite. The freezing is so rapid that the normal hot shortness of austenitic solidification is avoided, so quality is not compromised. In fact, laser welds quench the material so rapidly that corrosion resistance is enhanced since inclusions cannot nucleate and grow. Duplex alloys, on the other hand, freeze in their high-temperature ferrite structure because the fast quench prevents the nucleation and growth of austenite. Unless this ferrite is heated to permit austenite to form, lower-toughness welds will result. Ferritic, martensitic, and PH alloys are not harmed by the rapid quench. Resistance welding is readily done on most types of stainless steel. Allowance must be made for the lower thermal and electrical conductivity of stainless steels compared to other common materials. Most resistance welds, including both seam and spot welds, have deep, tight crevices adjacent to the welds. The possibility of crevice

Chapter 17: Welding / 211

corrosion in these regions should be considered when contemplating the use of spot welds in stainless materials. The possibility of entrapment of foreign material and the difficulty of removing it from such crevices should also be considered, especially in equipment for food handling, pharmaceutical production, etc. High-frequency induction welding of stainless steel is more difficult than for lowalloy steel because of the refractory nature of chromium oxide, which has a higher melting temperature than does the stainless base metal. This is opposite from the situation in low-alloy steels, for which the iron oxide melts at a lower temperature than does the iron base metal. The presence of this refractory oxide on the surfaces to be joined makes it more difficult to obtain a defect-free weld. Thermal cutting of stainless steels is routinely practiced, but the processes and parameters used are determined by the refractory nature of the chromium oxides that form on stainless steels. The high temperatures attainable with lasers or plasma arc torches provide good cutting action, and these processes are frequently used. To expand the range of thicknesses that can be cut or to increase cutting speed, supplemental oxygen or nitrogen blast jets may be used. Stainless steels may also be cut using oxyfuel equipment if supplemental iron powder is used. Combustion of the iron increases the temperature, while the iron oxide helps flux the refractory chromium oxide. Thermally cut edges of stainless steel usually require subsequent cleaning, typically by grinding or milling. Chemical cleaning of all surfaces of cut pieces to remove heat tint, fume deposits, and other contaminants is advisable. Soldering and brazing are possible with all stainless steels. Soldering is done below 450 °C (840 °F), while brazing is done above 450 °C (840 °F). Solders are generally alloys of tin and bismuth, lead, silver, or antimony or combinations of several of these. Brazes are normally either silver based or nickel based. The chromium-rich oxide coating must be removed by a suitable flux for bonding to occur. Fluxes are typically acid type with chlorides. Thus, after the soldering or brazing, the flux must be thoroughly removed to prevent subsequent pitting corrosion. Brazing temperatures must be chosen to avoid ranges at which unfavorable phases form. The best range can be determined from examining temperature ranges to be avoided in the thermal processing chapter

(Chapter 13) of this book. Brazes and solders rarely match the corrosion resistance of stainless steels, and careful attention should be given to the potential for galvanic and other forms of corrosion when considering the use of soldered or brazed joints with stainless steels.

Welding Practices Safety must always be considered when welding. In addition to the normal hazards (which are not discussed here) associated with welding, welding of stainless steels presents a special hazard: hexavalent chromium. The fume created by welding stainless steel contains significant concentrations of chromium trioxide and other forms of hexavalent (Cr+6) chromium. Hexavalent chromium is a carcinogen and regulated by the Occupational Safety and Health Administration (OSHA). Exposure to and inhalation of stainless steel welding fumes must be avoided. The product exposure limit for hexavalent chromium is 5 μg/m3 as of December 31, 2008. Refer to OSHA for further updates on this limit. Use of fume extraction equipment is generally the preferred method of minimizing hexavalent chromium exposures. Positioning and operation of the fume extraction device must be done precisely to ensure effective fume removal while avoiding excess turbulence, which can cause loss of effective inert gas shielding of the weld pool. Thermal cutting of stainless steels also generates hexavalent chromium, and similar procedures are required to minimize exposure during such operations. Nondestructive Evaluation (NDE) is used almost universally to ensure weld quality. All of the standard NDE techniques used with other materials are applicable to stainless steel weldments. Allowance must be made for the differing physical properties of stainless steels, and appropriate reference defect standards must be provided. However, one technique—magnetic particle inspection—is problematic. The presence of bands of persistent austenite in martensitic or PH stainless steels can lead to spurious defect indications. For this reason, magnetic particle examination of stainless steel welds is best avoided. Recent developments in stainless steel have been made with weldability as a major consideration. Highly alloyed, low-carbon martensitic alloys for line pipe have been developed with

212 / Stainless Steels for Design Engineers

the express purpose of use in the as-welded condition. The low carbon makes welds of this material that are tough and do not require tempering, so girth welds in the field are possible. Likewise, the lean duplex alloys have very delayed precipitation of intermetallic phases because of their higher nitrogen and lower chromium and molybdenum contents. This makes welding of these alloys much more foolproof than with the early duplex alloys, such as S31803. The dual-stabilized ferritic alloys have tougher welds than those stabilized with only titanium or niobium. New developments in welding also have an impact on stainless steels. The friction stir welding (FSW) process offers the promise of reliable solid-state joining. By avoiding melting and resolidification, issues associated with solute redistribution are eliminated. The relatively low temperatures involved essentially eliminate

generation of weld fume (see the discussion of safety). Other new welding processes, such as multiple (GTA or GMA) torch welding, laserassisted GMA or GTA welding, etc. promise greater productivity. REFERENCES

1. D.J. Kotecki, Welding of Stainless Steels, Welding, Brazing, and Soldering, Vol 6, ASM Handbook, ASM International, 1993, p 677–707 2. B. Aziez and R. Feen, Sheet Metal Ind., 1, 1983, p 28–34 3. S.D. Washko and J.F. Grubb, Proc. Int’l Conf on Stainless Steel, 1991, Chiba, ISIJ 4. Stainless Steels, Les Editions de Physiques, 1992, p 786 5. A. Garner, Corrosion, 37, 1981, p 178

Stainless Steels for Design Engineers Michael F. McGuire, p 213-223 DOI: 10.1361/ssde2008p213

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 18

Architecture and Construction Summary STAINLESS STEEL IS OFTEN EMPLOYED as an architectural material; the material can itself be viewed as a metaphor for architecture, a discipline that must balance aesthetics, economy, and structural integrity. Stainless steel’s unique combination of beauty, strength, and economy makes it a remarkably appropriate material for uses as diverse as sculpture and concrete reinforcing bar. However, stainless steels are complex; they come in many different grades (chemical analyses), and these grades have varying strengths, appearance, resistance to corrosion, availability, and costs. The success of a building project involves careful planning for the appropriate use of materials. This chapter deals with the technology of stainless steel as it pertains to its proper use in architecture, art, and construction.

can choose from a variety of stainless alloys with sufficient corrosion resistance to withstand any environment. The ability of stainless steel to resist corrosion resides in its chromium-rich superficial passive layer. Stainless steel by definition must contain slightly more than 10% Cr. The passive layer forms spontaneously in air or water, and if it is removed, say by abrasion, it re-forms by itself. This is explained in greater technical

Corrosion Resistance Corrosion is the life-limiting factor for architectural metals. Steel, copper, aluminum, lead, bronze, and other alloys react with the environment and degrade over time, as does wood, stone, plastic, paint, and even glass. With stainless steel, it is possible to choose a material that can withstand attack from the environment indefinitely. One of the first major uses of stainless steel in architecture was in New York’s Chrysler building, completed in 1930. Despite the rather crude methods of early production, limited alloy options, and lack of application experience involved in its construction, the domed top of the Chrysler building still shines undiminished by the harsh coastal and urban climate (Fig. 1). With today’s technology, one

Fig. 1

The Chrysler building with its famous bright stainless details. Copyright © iStockphoto.com/Steven Allen. Used with permission

214 / Stainless Steels for Design Engineers

detail in this book in the chapters on corrosion, but the key aspects are that the chromium atoms on the surface of the metal react with oxygen in air and water to form with neighboring iron atoms into a tight, ionically nonconductive layer that prevents any further oxygen penetration. This layer is mere atoms thick and completely invisible. The strength of the layer in resisting corrosive attack is proportional to several key alloying elements: chromium, molybdenum, and nitrogen. They contribute according to the following formula: PREN = %Cr + 3.3(%Mo) + 16(%N)

(Eq 1)

PREN stands for pitting resistance equivalent number. This number can be related to resistance to the mildest form of corrosive attack that stainless steel undergoes, pitting corrosion. Pitting is a “weakest link” phenomenon in which corrosion begins in small, micron-size parts of the surface and then grows by virtue of the more aggressive media that form within them because of the corrosion reaction’s products. Pitting occurs in environments that contain chlorides. Chloride ions compete with oxygen and disrupt the integrity of the protective passive layer. As alloys become richer in the alloying elements mentioned, their ability to maintain the passive layer can overcome the chlorides’ ability to destroy it. The key is to choose an alloy rich enough in chromium, molybdenum, and nitrogen to withstand any environment the structure will experience. Pitting corrosion, and another similar form of corrosion called crevice corrosion, can be prevented by proper choice of alloy, finish, and design. Crevice corrosion occurs when recessed spaces are small enough to act like a corrosion pit. The acidity within the crevice increases because of restricted diffusion in and out of the crevice, just as happens within a pit. The buildup of iron and chloride ions makes a very corrosive medium that disables the passive film formation. A design that avoids crevices is the best defense. The decision criterion for material selection with stainless steel should then be, Which grade and finish will exclude the possibility of pitting corrosion at the lowest cost? Then, proper design should be used to exclude the possibility of crevice corrosion. Other factors involving strength and fabrication should also be considered. Table 1 ranks a number of stainless alloys by pitting resistance.

Balancing Corrosion Resistance, Processing Characteristics, and Economy The rule of thumb for grade selection is usually somewhat oversimplified to recommend the use of types 430 and 304 on interior applications, type 304 on exteriors where salt is not a problem, and 316 where road salts or seacoast effects make a more corrosion-resistant grade necessary. A leading architectural metals company (Ref 1) makes the following recommendations, which mirror these traditional views: • Type 304 should be used for most exterior applications. • Type 316 should be used within ten miles of saltwater bodies. However, if the building is subject to saltwater spray, a nobler grade of stainless steel, such as 2101 or 2003, should be specified. • In close proximity to deicing salt use, even on nearby roadways where vehicle traffic can create airborne particles, type 316 should be used. If periodic rinsing will not occur on all exterior surfaces, these areas must be washed each spring. If dependable maintenance is not predicted, a nobler grade of stainless steel, such as 2101 or 2003, should be used. • Specify types with low carbon, less than 0.030%, if welding will be employed. • Any grade, including type 430, may be used in interior applications. • In the most severe environments—high heat and humidity, low rainfall, and high salinity, such as are found in Middle Eastern countries—a grade with a PREN of 25 or above is recommended. These guidelines are based on the admittedly easy availability of these alloys and a lack of concern for cost during times of peak raw material prices. For projects where quick availability is not more important than cost, 439 and 201 should be considered as viable replacements for 304. Stainless steel 2003 (UNS S32003) or an equivalent lean-duplex grade can replace 316 at a cost advantage during times of high alloy cost Table 1 Ranking of common stainless steels by pitting resistance equivalent number (PREN) Alloy 430 PREN 15

439 17

201 17

304 19

316 24

2101 2003 2205 2507 26 28 35 38

Higher PREN values indicate greater pitting resistance.

Chapter 18: Architecture and Construction / 215

as have been experienced on occasion, such as during the period of 2004 to 2007. It should be noted that the leaner alloys suggested (439, 201, and lean-duplex alloys such as 2003) can be somewhat more difficult to form. If panel designs call for 90° bends, this is not an issue. However, for applications requiring severe forming, as in the case of a double-lock seam on a standing seam roof, these grades can provide a challenge to the fabricator/installer. Further, these leaner grades can pose challenges with certain finishing methods, such as abrasive polishing and embossing. To avoid unwanted complications related to grade selection, the specifier should consult a competent architectural metals supplier. This effort will ensure a viable specification is written that will balance cost with the necessary performance attributes to make the part as well as resist corrosion once installed. It is valuable to know, in times of high nickel prices, that both the low-nickel 201 and the nonickel 439 can be used in place of 304, while 2101 (UNS S32101) and 2003 (lean duplexes) can replace 316. To obtain these grades usually involves working with a producing mill since they are not typically stocked in service center inventories. However, any competent architectural metals supplier will not shy away from the use of specialty grades where appropriate.

Surface Finish and Corrosion Resistance Surface finish is usually an aesthetic choice, but it has a significant influence on corrosion resistance and must be factored into grade selection. Mill finishes such as 2B and 2D are inconsistent because they are annealed and pickled to remove oxides. These unattractive surfaces, however, have correct corrosion resistance for their alloy content. Welding or abrading the surface degrades the corrosion resistance by a significant amount. An un-heat-treated weld has lower resistance to corrosion in proportion to the alloy content of the grade. Type 316 welds have the corrosion resistance of wrought 304. Abrasion has a similar effect. Type 316 with a No. 4 polish behaves like 304 with a 2B mill finish. Because welds are abraded, this compounds the effect. Very smooth abrasively polished finishes mitigate this reduced corrosion resistance, as shown in Fig. 2. The effect of reduced corrosion

resistance in abraded stainless steel surfaces is not seen on finishes that are produced by patterns imprinted by hard-rolling mill rolls that have been engraved with the desired pattern in reverse; for this reason alone, this method of surface finishing is recommended. These architecturally useful surface finishes are produced by the preferred rolled-on, or embossed, method. In addition to their advantage in corrosion resistance, they are extremely uniform from batch to batch, unlike finishes produced by abrasive-coated belts, which change in grit coarseness with use.

Balancing Service Environment, Design Requirements, and Maintenance Considerations An expert system has been developed that enables designers and specifiers to analyze the trade-offs of climate, design requirements, and maintenance on grade selection (Ref 3). Answering the questions in Fig. 3 for a particular application yields a score that can be used to identify an appropriate alloy according to the scale shown Fig. 4. The Nickel Institute, formerly the Nickel Development Institute, also offers excellent publications on topics related to alloy selection for specific service environments and design requirements (Ref 4). Reviewing a map of the salinity of rainwater in the United States is instructive of the degree to which geography influences corrosion severity. The average atmospheric chloride levels collected in rainwater are shown in Fig. 5 (Ref 5). The highest levels occur along the coastlines of the Atlantic and Pacific Oceans and the Gulf

Fig. 2

The decrease in corrosion resistance with increasing surface roughness by abrasion. Source: Ref 2

216 / Stainless Steels for Design Engineers

Fig. 3

Stainless steel selection expert system. Source: International Molybdenum Association (Ref 3)

Chapter 18: Architecture and Construction / 217

Fig. 4

Grades recommended based on the expert system. Source: International Molybdenum Association (Ref 3)

Fig. 5

Average chloride concentration (mg/L) in rainwater in the United States. Source: Ref 5

of Mexico. The maximum corrosion rate is related to the maximum chloride in the atmosphere. This will be related to the distance inland, the height above sea level, and the prevailing winds (Ref 6).

Aesthetic Considerations A correctly chosen grade of stainless steel will have no degradation over time and, if prop-

erly maintained, will stay new looking indefinitely. Surface finish aesthetics are arguably more important architecturally than the influence surface finishes have on corrosion. Numerous finishes have been developed to try to meet various objectives. Finishes vary in reflectivity, directionality, and subtlety. Figure 6 shows some of the finishes that go beyond the familiar brushed look (Ref 7), while Fig. 7 shows special finishes created by one manufacturer.

218 / Stainless Steels for Design Engineers

Fig. 6

Various rolled-on stainless steel finishes. Source: Ref 7. Courtesy of Outokumpu

Fig. 7

Special finishes for 304/304L and 316/316L stainless steels available from one manufacturer. (a) Rolled-in low-glare finish (InvariMatte). (b) Rolled-in no. 4 finish (InvariBlend). (c) Rolled-in moderate-glare finish (InvariLux). Source: Contrarian Metal Resources (Ref 8)

The classic abrasively produced finishes are No. 3, 4, and 8. These are American Society for Testing and Materials (ASTM) designations for abrasively produced finishes, which are traditionally produced by abrading the surface with different grit size abrasives. Finish No. 3 calls for 80 to 100 grit abrasive; No. 4 calls for 120 to 150 grit abrasive. Finish No. 8 is a mirror finish obtained by final polishing with 800 grit abrasive. Finishes No. 3 and 4 are directional, with grit lines typically 1 cm (0.4 in.) in length. Finish No. 3 has a surface roughness average (Ra) of 0.4 to 0.8 µm (15 to 30 µin.). Specular gloss at 85° is typically 40 to 60 (per ASTM D 523,

“Standard Test Method for Specular Gloss”). The standards for appearance do not exist within specifications, only the method of producing them. There is considerable difference in appearance from sheet to sheet, coil to coil, and manufacturer to manufacturer. The greatest consistency of appearance comes from specifying a brand of rolled-on finish from a given manufacturer. Any of the traditional finishes can be replicated by a rolled-on finish with greater uniformity, with the possible exception of bright annealed having a difficult time matching the mirror quality of a No. 8 finish. This is crucial in architecture, where the discovery of

Chapter 18: Architecture and Construction / 219

unacceptable visual nonuniformity on large areas can be disastrous, especially when this appears late in the construction process, as is normally the case with exterior components. Reflectivity or gloss can be a major consideration in the choice of a surface finish. Mirror finishes are often used for high impact, but more diffusely reflecting surfaces are more common. Patterned surfaces provide consistent reflectivity from a moderately reflective 40 to 60 specular gloss at 85° to a dull matte of less than 20, the latter having been developed for airport roofing, such as at Reagan Airport in Washington, D.C., or the Pittsburgh Convention Center (Fig. 8). Flatness is a special consideration for panels where lack of flatness, such as by “oil-canning,” can cause a very shoddy appearance. Flatness is measured in I units. Flatness (Iunits) = 2 (πH / 2L) × 105

(Eq 2)

where H is the height of the deviation from flatness, and L is the distance between peaks of deviations, assuming a sinusoidal wave.

Fig. 8

Because stainless in sheet form is usually reflective, small deviations from flatness can be very visible. A good standard for flatness that precludes visible distortion is five I units. Steel producers have various means to produce this level of flatness, the most extreme of which is actually stretching the steel sheet or coil until all distortions are eliminated. Sometimes, rather than aiming for high flatness a controlled deviation from flatness is used, such as slightly concave panels or panels with a die-pressed design. Another option to ensure flatness is to back light-gauge stainless steel with a stiff material. Deviations of sheets from squareness and straightness (camber) are also objectionable because such deviations can cause gaps between panels. The degree to which this is objectionable is a function of design, and tolerances can be held tightly at a cost. Width tolerance is normally +1/16 in./–0 in 48 in., while length is held to +1/8 in./–0 in 10 ft or less. Maximum camber is 3/32 in. in 8 ft. Closer tolerances can be negotiated.

The Pittsburgh Convention Center with low-gloss finish stainless steel roof

220 / Stainless Steels for Design Engineers

Maintenance and Repair Maintenance is a significant cost of any structure. One of the great values of stainless steel is its low cost of ownership. Stainless steel can be abused, however, and it does benefit from proper maintenance. The main objective of the maintenance of stainless is keeping it clean. There are two reasons for this. The obvious first reason is that whatever is soiling the surface is probably not attractive. The second reason is that it may harm the surface by allowing corrosion agents to concentrate. Table 2 provides recommended practices for removing various substances from stainless steel surfaces (Ref 9). One of the most common complaints about maintenance of stainless is the work involved in removing fingerprints. The oil from fingerprints makes an easily visible interference film on the reflective stainless surface. The typical remedy is to clean stainless steel with a solution containing light oil and a detergent. If the oil from a hand contacts the uniformly thin film of cleaning oil, no visible mark is left. Alternatively, Table 2

• Designs that can collect dirt, such as horizontal surfaces and recesses, should be avoided. • Designs that create uneven flow or drainage patterns producing uncleansed areas should be avoided. • Sheltered areas and areas subject to splatter, especially roadside spatter, should be designed so that they are easily cleanable.

Cleaning methods for uncoated stainless steel

Requirement

Routine cleaning of light soiling

Fingerprints Oil and grease marks Stubborn spots, stains, and light discoloration; water marking; light rust staining Localized rust stains caused by carbon steel contamination

Adherent hard water scales and mortar/ cement splashes Heat tinting or heavy discoloration

Badly neglected surfaces with hardened accumulated grime deposits Paint, graffiti

polymer coatings are applied to stainless steel at some producing mills to permanently provide a film to which additional fingerprint oil cannot add a noticeable discoloration. The greatest ally of stainless on building exteriors is the cleansing action of rain. Rain does not completely clean the surface, but it does dilute any harmful contaminants and forestall corrosion from accumulated chlorides. Without the benefit of cleansing by rain, stainless exteriors should be washed during routine window washing operations. Given the importance of cleaning, the following design considerations are recommended:

Suggested method(a)

Comments

Soap, detergent, or dilute (1%) ammonia Satisfactory on most surfaces solution in warm clean water. Apply with a clean sponge, soft cloth, or soft-fiber brush, then rinse in clean water and dry. Detergent and warm water; alternatively, Proprietary spray-applied polishes available to hydrocarbon solvent clean and minimize re-marking Hydrocarbon solvent Alkaline formulations are also available with surfactant additions. Mild, nonscratching creams and polishes. Avoid cleaning pastes with abrasive additions. Apply with soft cloth or soft sponge; rinse Cream cleaners are available with soft off residues with clean water and dry. calcium carbonate additions. Avoid chloridecontaining solutions. Proprietary gels or 10% phosphoric acid Small areas may be treated with a rubbing solution (followed by ammonia and water block comprising fine abrasive in a hard rubrinses) or oxalic acid solution (followed by ber or plastic filler. Carbon steel wool and water rinses) pads that have previously been used on carbon steel should not be used. A test should be carried out to ensure that the original surface finish is not damaged. 10–15 vol% solution of phosphoric acid. Proprietary formulations available with Use warm, neutralize with diluted ammonia surfactant additions. Avoid the use of solution, rinse with clean water and dry hydrochloric acid-based mortar removers. Nonscratching cream or polish. Apply Suitable for most finishes. with soft cloth or soft sponge. Rinse off residues with clear water and dry. Nylon-type pad Use on brushed and polished finishes along the grain. A fine abrasive paste as used for car body May brighten dull finishes. To avoid a patchy refinishing. Rinse clean to remove all paste appearance, the whole surface may need to material and dry. be treated. Proprietary solutions or solvent paint stripper Apply as directed by manufacturer depending on paint type. Use soft, nylon or bristle brush on pretreated material.

(a) Cleaning agents should be approved for use under the relevant national environmental regulations and should be prepared and used in accordance with the company’s or supplier’s health and safety instructions. Source: Adapted from Ref 9

Chapter 18: Architecture and Construction / 221

• Contamination by rust from carbon steel is corrosive and must be removed. • Designs should facilitate easy access for cleaning. While any structure would benefit from these guidelines, they are especially valuable in maximizing the benefits of stainless steel. Repair of more severe damage done to surfaces, such as scratching, is difficult. If a decorative surface pattern has been damaged, the challenge is in trying to replicate it in the field. Very few surface finishes are wholly repairable in the sense that they can be repaired in a spot so that the repair is invisible. The reason for that is mainly that abrasive finishes are applied by rotating belts, and it is virtually impossible to match the pressure, grit size, and arc of contact that created the original surface. It can be more nearly done to a surface with a rolled-on finish, which has a very consistent grit length and depth. The exception to this rule are abrasively applied or rolled-on long-grain finishes. These have very long grit lines, so grit length is easily duplicated with a belt sander, the usual tool available for field repairs. Even welds can be removed and reblended to be indistinguishable from the surrounding original surface. The ability to be repaired should be a top criterion in the choice of a surface finish whose appearance is critical and that may be subject to damage.

Fabrication Considerations Fabrication and joining of stainless steel employ the same techniques as for carbon steel and other metals. The specifics of cutting, forming, joining, soldering, and welding are described in the processing section of this book and are not repeated here. The main distinction in the use of stainless steel in this regard is that its higher strength and corrosion resistance permit the use of lighter gauges. This in turn permits designs in hollow or rolled-formed sections, which have higher stiffness and low weight and potentially lower overall cost than using less-expensive metals. A second aspect of higher strength and lighter gauge is greater spring back in forming operations, such as press braking. Certain processing principles related to architectural and building applications of stainless steels should be emphasized:

• Separation of tools and work areas between those used for stainless steel and carbon steel is prudent. Contamination of stainless surfaces with carbon steel from welding, grinding, and cutting can stain the surface of stainless steel and result in corrosion. This can be remedied by passivation, but it is much better to avoid it in the first place. • Welding is better done in the shop than in the field. Correct filler metals must be used, and proper weld finishing is essential. The ability of contactors to produce sound, attractive welds is an indicator of their overall competence with stainless steel. • Fasteners used with stainless steel should also be stainless steel. Galvanized steel, carbon steel, and aluminum will corrode more readily than the stainless, and this corrosion is aggravated by galvanic contact. The resulting corrosion products are also harmful as well as unsightly. Fasteners should not be permitted to cause distortion of flat panels.

Additional Service Considerations Fire resistance is an important consideration in buildings. Stainless steel is the only common building material that remains strong and tough at temperatures encountered in fires. Ordinary carbon steel undergoes a phase change at about 760 °C (1400 °F). This change in atomic structure results in a sudden shrinkage of more than 1 linear percent. This can literally pull a building apart. When this occurs to a structure already weakened by heat, as carbon steel is, catastrophic failure ensues. Austenitic stainless steel keeps the same atomic structure and remains much stronger than carbon steel at elevated temperatures. Thus, austenitic stainless steel has great value as a material for structures that must retain structural integrity in a fire. Tests have been conducted on glass-reinforced plastic, aluminum, galvanized steel, and austenitic stainless steel ladders under load and exposed to flame temperatures of more than 1000 °C (1830 °F). The plastic and aluminum ladders failed in less than a minute. The galvanized carbon steel lasted 5 min, while the stainless steel remained intact (Ref 10). If the need for fire resistance is serious, stainless steel becomes the material of choice. It is used on offshore oil platforms for stairways, ladders, walkways, handrails, gratings, floor systems,

222 / Stainless Steels for Design Engineers

Fig. 9

Graphic depicting low release of metal ions from two grades of stainless steel (304 and 316) to the environment, based on a 4-yr multidisciplinary research project involving both field research and laboratory studies. Source: Ref 11

firewalls, blast walls, living modules, and so forth. Ecological considerations are never trivial when considering a construction material. Many materials used in buildings degrade environmentally, usually by corrosion, and enter the general environment. Asbestos, lead-based paints, lead coatings, and others are once-accepted materials whose long-term effects have been dangerous and costly. Stainless steel, because it does not corrode when properly used, does not enter the environment. While this seems obvious, it has been the object of interdisciplinary studies that have demonstrated its innocuousness even under conditions of heavy acid rain on freshly abraded surfaces (Fig. 9) (Ref 11). Stainless is a material that will never come back to haunt an architect years later. Its intrinsic raw material content value ensures that even with the end of the life of a structure, the stainless in the structure will be recycled.

Concrete Reinforcing Bar Concrete reinforcing bar is one of the least glamorous uses of stainless steel in structures.

In bridges, parking garages, and other concrete structures, saltwater can penetrate the cement over time. If the internal rebar corrodes, the expansion of the corrosion products spalls the concrete, leading to the failure of the structure. This can be delayed by treating the concrete to repel the incursion of water or by coating the carbon steel rebar with epoxy. These are less than 100% effective. A more certain approach is to use stainless steel rebar. The stainless steel for this duty need not resist pitting corrosion, which affects a tiny percentage of the steel volume. Therefore, an inexpensive, low-nickel grade, such as 409, 430, or 201, can be used. Most of the work to date has been with more expensive grades, such as 316 and 2205. The lean duplexes are ideal for this application because of their high strength, resistance to corrosion and SCC, and moderate cost. The use of even these alloys reduces the longterm cost of these structures, so the future adoption of less-expensive stainless steels holds great promise. The more enlightened transportation departments in the United Kingdom; Ontario, Canada; and Michigan, New Jersey, and Oregon in the United States have led this development.

Chapter 18: Architecture and Construction / 223

REFERENCES

1. Stainless Steel Selection Criteria, Contrarian Metal Resources, www.metalresources. net, accessed June 2008. 2. Bulletin of the National Dairy Federation 189, 1985, p 3–12 3. C. Houska, “Which Stainless Steel Should Be Specified for Exterior Applications?” International Molybdenum Association, www. imoa.info, accessed June 2008 4. The Nickel Institute, www.nickelinstitute. org, accessed June 2008 5. H. Guttman, Atmospheric and Weathering Factors in Corrosion Testing, Atmospheric Corrosion, W.H. Ailor, Ed., John Wiley and Sons, 1982, p 51 6. R.B. Griffin, Corrosion in Marine Atmospheres, Corrosion: Environments and In-

7. 8. 9. 10. 11.

dustries, Vol 13C, ASM Handbook, ASM International, 2006, p 42–60 Guide to Stainless Steel Finishes, 3rd ed., Euro Inox, 2005, www.euro-inox.org, accessed June 2008 Contrarian Metal Resources, www.metalresources.net, accessed June 2008 Care and Maintenance of Stainless Steels, Leda-Vannaclip, www.l-v.com.au, accessed June 2008 The Nickel Institute, www.nickelinstitute. org, accessed June 2008 D. Berggren et al., Release of Chromium, Nickel and Iron from Stainless Steel Exposed Under Atmospheric Conditions and the Environmental Interaction of These Metals, European Confederation of Iron and Steel Industries, Oct 2004, www.eurofer. org, accessed June 2008

Stainless Steels for Design Engineers Michael F. McGuire, p 225-232 DOI: 10.1361/ssde2008p225

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 19

Automotive and Transportation Applications Summary THE ADEQUATE DURABILITY and life span of cars, trucks, or any transport system requires freedom from corrosion. This has required subsystems, such as those for exhaust and fuel, to resist more corrosive environments for longer periods of time. The main result has been a strong growth in the use of the leaner ferritic stainless steels in many components. As more exotic propulsion systems and fuel, such as fuel cells and ethanol, emerge, stainless steels may be required to endure the corrosive environments.

the auto manufacturers have become the largest users of stainless steel. Exhaust systems constitute the largest use of stainless steel in the automotive market, but there are other important applications that cannot be ignored: valves and gaskets, hose clamps, seat belt and air bag components, tubing, hardware, and filters. And, there will be new applications that respond to new socioeconomic needs, such as for greater crash worthiness, lighter weight, or resistance to the corrosion of new fuels. But, since exhaust systems currently predominate, they are covered first.

Exhaust Systems Introduction The use of stainless steel in automobiles used to be mainly a story of decorative applications: wheel covers and trim with a minor amount used for valves and hose clamps. However, as automobiles became more sophisticated technically and as durability and environmental demands grew, the role of stainless became increasingly functional and less ornamental. Stainless alloys in common automotive use now are generally highly engineered for their specific application and represent some of the most highly evolved applications engineering in any use of stainless. An examination of the preferred practices in materials selection in automotive systems is an excellent example of the rule of using the simplest and lowest alloy content grade that can do the job. Because automotive and steel-producing engineers have collaborated so well, both parties have benefited greatly, as have consumers, and

Laws enacted in the United States in the 1970s mandated automotive emission standards that could be met only with catalytic converters. The only practical materials that could withstand the temperatures of the hot end of an exhaust system using a catalytic converter were stainless steels. From AlleghenyLudlum’s MF-1 evolved a succession of ferritic alloys that grew in sophistication to meet the increasing needs of corrosion resistance, oxidation resistance, creep, thermal fatigue resistance, and formability. Soon, entire exhaust systems were made of stainless; often, they would last the life of the vehicle, rendering obsolete an entire muffler replacement industry. The compelling need of the automotive industry for economy drove the widespread adoption of argon oxygen decarburization (AOD), the continuous caster, and other highvolume methods of the carbon steel industry. So, even while the traditional automotive uses

226 / Stainless Steels for Design Engineers

of stainless steels—wheel covers and other ornamental trim—faded to nearly nothing, the use of stainless in automobiles grew to about 30 kg (65 lb) per vehicle by the turn of the century. The stainless steel industry was changed from a boutique industry to a mass production industry by its embracing the needs of the automotive market. An exhaust system normally consists of a manifold to collect exhaust gases, a catalytic converter to reduce NOx and CO emissions, and a muffler; each of these are connected by piping. Each component of the system has different requirements for formability, resistance to oxidation, resistance to external corrosion, resistance to internal corrosion, and mechanical properties. At the outset, it should be noted that ferritic stainless steels, as opposed to austenitic, are optimal for oxidation resistance, especially cyclic oxidation. It is not a difference in the oxide scale. The reason is that the thermal expansion of ferritic stainless more closely matches that of the oxide scale than does that of austenite. This prevents the fracturing and spalling of the scale. The intact scale of ferritic stainless is thus protective up to the temperature at which oxygen diffusion through the scale becomes great enough so that “breakaway” oxidation occurs. At the breakaway temperature level, scale growth is no longer parabolic with time but becomes linear and therefore no longer protective. The temperature of this breakaway increases as chromium content increases. We will see that other alloying elements can also improve this performance. Not only are the steels in exhaust systems almost exclusively ferritic, they are also stabilized by titanium or niobium. This prevents sensitization and makes all the chromium content useful as alloy. Titanium stabilization greatly improves corrosion resistance by removing not just the carbon and nitrogen, but also the oxygen and sulfur from solution. This sharply improves resistance to pitting corrosion. Niobium is used to costabilize and fight creep. These alloys are therefore essentially interstitial free and have excellent formability, which the designs of exhaust system components require. Their formability is further enhanced, when necessary, by low additions of tramp substitutional alloying elements such as manganese, nickel, and copper. Special thermomechanical processing is also used to optimize texture and grain size.

The metallurgy of ferritic stainless is discussed in depth in Chapter 8. Rather than reexplain these concepts here, we revisit only the main points that are relevant to alloy selection for exhaust systems. • As chromium level increases, so does resistance to oxidation and corrosion, but yield strength also increases, and ductility decreases. • Alloying with silicon, aluminum, and molybdenum also increases oxidation resistance, but these elements have the same detrimental effect on ductility while increasing hot strength. • Niobium above that needed for stabilization is a powerful solid solution hardener and is effective at high temperatures. • Ferritic stainless steels have very anisotropic forming properties. They resemble highformability carbon steels in that they tend not to thin when stretched, which greatly assists in formability. • Ferritic alloys can form a hard, brittle phase called α’ in a process commonly called 885 °F (or 475 °C) embrittlement. This is only a factor in alloys with chromium of 18% or more, especially those containing molybdenum and aluminum. Cold work accelerates the formation of this phase. • The σ phase does not readily occur in alloys containing less than 20% Cr, so it is not a consideration for exhaust systems unless silicon or molybdenum are also elevated. • Coating ferritic steel with aluminum is effective in preventing oxidation and corrosion. All these factors come into play in the design of exhaust systems. Because the alloys have evolved so well to fit the individual requirements for each component, we discuss them segment by segment through the exhaust system. The exhaust manifold collects the hot, combusted gases from the engine and delivers them to the front pipe. The exhaust manifold must possess good high-temperature strength and resistance to thermal fatigue. It must also be able to resist oxidation at the exhaust temperature, which can reach 950 (C (1740 (F). Exhaust manifolds had previously been heavy castings but are generally now formed from stamped sheet stainless steel or formed from welded tubing that may have a double-wall structure to insulate the gases from heat loss, which could preclude successful catalytic conversion downstream.

Chapter 19: Automotive and Transportation Applications / 227

As the highest-temperature component of the exhaust system, the exhaust manifold must possess the greatest resistance to high-temperature oxidation damage. Risk of such damage is due to the intermittent use of vehicles, which causes cyclic oxidation and the ensuing spalling of the oxide scale. There are numerous alloying approaches for optimizing the ferritic stainless alTable 1

loys for spalling resistance. All approaches involve raising chromium content but use different techniques to enhance the effect of chromium. Table 1 lists a number of the grades of stainless steel commonly used in exhaust systems and where they are used. The alloys are listed in order of increasing severity of the requirements for each major system component.

Alloys normally used for the major elements of automotive exhaust systems Service temperature

Component

Exhaust manifold

ºC 750–950

ºF 1380–1740

Alloys currently used, common name (related designation)

Requirements

High-temperature strength, thermal fatigue strength, oxidation resistance, formability

• • • • • • • • • • •

T439HP (UNS S43035, dual-stabilized 439) 18CrCb (DIN 1.4509, 18CrCb) 441 (DIN 1.4509) 304/304L/304H (UNS S30400, S30403, S30409) 321 (UNS S32100) 309S (UNS S30908) 310S (UNS S31008) 332Mo (S35125) 600 (N06600) 601 (N06601) 625 (N06625)

• • • • • • • High-temperature • strength, thermal • fatigue strength, • oxidation resistance, • formability, salt • attack resistance • High-temperature • strength, salt • attack resistance, • formability • • •

409 ALMZ (aluminized 409) T439HP (UNS S43035, dual-stabilized 439) 18CrCb (DIN 1.4509, 18CrCb) 441 (DIN 1.4509) 436S (type 436S) 444 (UNS S44400, T441) 433 (T443) 304/304L (UNS S30400, S30403) T321 (S32100) 316/316L (S31600/S31603) 316Ti (S31635) 332Mo (S35125) 625 (N06625) 409HP (UNS S40930, dual-stabilized 409) T439HP (UNS S43035, dual-stabilized 439) 441 (DIN 1.4509) 18CrCb (DIN 1.4509, 18CrCb) 444 (UNS S44400, T441) 433 (T443)

Front pipe

600–800

1110–1470

Flexible pipe

600–800

1110–1470

Catalytic converter shell

600–800

1110–1470

1000–1200

1830–2190

Oxidation resistance, thermal shock resistance

• ALFA-IV (FeCrAl)

Center pipe

400–600

750–1110

Salt damage resistance

Muffler

100–400

210–750

Corrosion resistance, from inner and outer surface

• • • • • • • • • • • • • •

Tailpipe

100–400

210–750

Corrosion resistance, from inner and outer surface

Catalytic converter substrate

Source: Adapted from Allegheny Technologies Inc.

High-temperature strength, thermal fatigue strength, oxidation resistance, formability

409HP (UNS S40930, dual-stabilized T409) 409 ALMZ (aluminized 409) T439HP (UNS S43035, dual-stabilized T439) 441 (DIN 1.4509) 18CrCb (DIN 1.4509, 18CrCb) 444 (UNS S44400, T441) 433 (T443) 409HP (UNS S40930, dual-stabilized T409) 409 ALMZ (aluminized 409) T439HP (UNS S43035, dual-stabilized T439) 436S (T436S) 441 (DIN 1.4509) 18CrCb (DIN 1.4509, 18CrCb) Type 304/304L (UNS S30400, S30403)

• 409HP (UNS S40930, dual-stabilized T409) • 409 ALMZ (aluminized 409)

228 / Stainless Steels for Design Engineers

The best choice for a given design is not obvious. We attempt to simplify the choices. Thus, the basic alternatives for exhaust system alloys are: • Straight chromium alloying at 11 to 12% with stabilization by titanium or niobium, the basic type 409 (UNS S40920) • Straight chromium alloying at 17 to 18% with stabilization by titanium or niobium, the basic type 439 (UNS S43036) Either of these basic alloys can enjoy enhanced oxidation resistance by additional alloying with molybdenum, aluminum, or silicon. In addition, they can be coated with hot-dipped aluminum-silicon alloy to increase oxidation resistance. Chromium or molybdenum alloy additions increase corrosion resistance, whereas aluminum or silicon additions do not improve that trait. Aluminum coating is a powerful corrosion fighter, and it has the aesthetic benefit of not showing red rust. Use of molybdenum or niobium enhances high-temperature strength. Alloys with these additions are thus useful for manifolds with a design that constrains expansion and contraction, making thermal fatigue a problem. All alloying additions detract from formability and toughness, as well adding to basic material costs. Thus, the objective must be to use only those alloying elements that are indispensable to performance. The front pipe connects the exhaust manifold to the flexible joint and experiences nearly the same temperatures as the exhaust manifold, but not the same risk of thermal fatigue. To reduce exhaust noise, a double-wall pipe is sometimes used for this component. The flexible joint is the one segment of the exhaust system for which austenitic stainless steels are preferred. The function of the flexible joint is to prevent vibration from the engine from being transmitted to the rest of the exhaust system. It consists of a double-wall pipe in a bellows configuration with an outer covering of braided stainless steel wire. It must have very good high-temperature fatigue strength to withstand the cyclic stress of the vibration it absorbs. The material used must have exceptional formability to be formed into a bellows. The greater hot strength and formability of austenitic steels thus prevails. The flexible joint is also exposed to road salt in

some regions, so it must resist hot salt corrosion. This may force the use of 316L versus the normal choice of 304L. The catalytic converter is the next component of the exhaust system. It exposes the exhaust gases to noble metal catalysts, which complete the combustion of the gases to form less-noxious compounds. This is an exothermic reaction at temperatures equal to those in the exhaust manifold. Thus, the housing, while not requiring great hot strength, must resist oxidation. The catalyst itself is supported by a ceramic and ferritic stainless steel carrier that must resist thermal shock and possess low heat capacity for rapid heating. Exotic alloys of 20% Cr with 5% Al are used for the carriers. The housing is generally made of a 17% Cr ferritic stainless. The converter is usually directly beneath the passenger compartment, so a heat shield of type 409 is used to separate it from the floor. The center pipe conveys the converted gases to the muffler. The cooling exhaust gases no longer present a major oxidation threat, but the condensing water vapor creates an internal corrosion risk, and road salt presents an external one. However, a simple grade such as 409 should generally provide sufficient resistance to this environment. The muffler, next in line, presents only a corrosion issue. The muffler must withstand corrosion from the outside, the worst of which comes from road salt or coastal salt sources. Internal corrosion is also a major consideration because condensing exhaust gases create a hostile, acidic environment. After startup, the heating of the muffler to temperatures above 100 °C (212 °C) evaporates these condensates, and internal corrosion ebbs. On short runs, this may not occur. This represents a worst case for internal corrosion. The dual internal and external corrosive attacks require the use of aluminized stainless for best performance. The tailpipe is exposed to view in most vehicles, and its appearance is therefore important. For this reason, an austenitic such as 304 can be used, as can chromium plating or aluminizing. The object here is to avoid visible corrosion. Truck exhaust systems are beginning to require similar technical sophistication as their emissions come under increased regulation. However, they do not present any challenges not already confronted and solved for passenger vehicles.

Chapter 19: Automotive and Transportation Applications / 229

Structural Components The driving forces of durability, safety, and weight reduction have spawned other, more varied applications for stainless in automotive engineering. Across the board, the main distinguishing trait of stainless that qualifies it as the optimal material is its corrosion resistance, but this characteristic would be insufficient in many cases without considering mechanical properties. Indeed, even if stainless steel were not corrosion resistant, its superior strength and toughness would qualify it for many automotive applications. Austenitic stainless steels are the toughest and stiffest practical materials available to the automotive engineers. Common 301 can be cold worked to yield strengths anywhere from its annealed level of about 300 MPa (44 ksi) up to 2000 MPa (290 ksi). In this higher-strength condition, it has become the standard material for seat belt anchors and hose clamps. Type 301 in the annealed condition is actually the original transformation induced plasticity (TRIP) steel as it can be tailored to have a controlled level of austenite stability. This allows it to transform at a known rate to martensite during deformation, giving not only a very high work-hardening rate

but extraordinary resistance to localized thinning, necking, and therefore fracture. When crash worthiness becomes a prime consideration, then this characteristic makes 301, or its low-nickel counterpart 201, an ideal material for structural, energy-absorbing components since austenitic stainless can be rivaled for such applications only by heat-treated alloy steel, titanium, or aircraft aluminum alloys, all of which are more expensive, less durable, or less formable. Tables 2 and 3 show the properties of specific variations on basic 301 developed by Outokumpu and how they stack up against the most competitive carbon steels, dual-phase steels, and TRIP steels (Ref 1). The value of a material as an energy-absorbing structure (i.e., one that enhances crash worthiness) is measured by the energy it can absorb per unit of mass. The kinetic energy of a collision that a structure can absorb in deformation is proportional to its strength multiplied by the amount it can deform before fracturing. The superiority of metastable stainless steels (i.e., 201 and 301, those that most easily transform to martensite during deformation) is shown in Fig. 1. Even with its lower density, aluminum falls far short of austenitic stainless in energy absorption per unit weight.

Table 2 Comparison of tensile properties of carbon steels and stainless steels for automobile structural components Type

Thickness, mm

0.2% proof strength, MPa

Ultimate tensile strength, MPa

True stress at ultimate tensile strength, MPa

Uniform elongation, %

Total elongation, %

Carbon steels TRIP 700 DP 750 DP 800

1.58 1.48 1.44

473 513 573

703 811 896

818 920 976

16.4 13.4 8.9

17 18.8 9.9

1.16 1.55

306 639

937 1068

1429 1377

52.5 28.9

59.3 38.6

Austenitic stainless steels HyTens X HT 1000 Source: Ref 1

Table 3 Comparison of resilience and toughness of carbon steels and stainless steels for automobile structural components Type

Resilience, J/m3

Toughness, j/m3

Carbon steels TRIP 700 DP 750 DP 800

0.996 1.131 1.32

105 101 74

0.536 1.726

364 269

Austenitic stainless steels HyTens X HyTens 1000 Source: Ref 1

230 / Stainless Steels for Design Engineers

Fig. 1

True stress-true strain curves for 301 variants (HyTens X and HyTens 1000) versus two duplex carbon steels (DP750 and DP800) and a transformation induced plasticity (TRIP) steel (TRIP700). Source: Ref 1

These exceptional strength-to-weight and energy absorption-to-weight characteristics permit automotive engineers to reduce weight and increase crash worthiness while designing vehicles with greater life span—because corrosion resistance “comes along for the ride,” as it were. Some components in which these virtues are most readily exploited are bumper systems (Ref 2). Porsche uses austenitic stainless steel for front and rear side members, internal push rods on front and rear axles, and lower rear wishbones in its Carrera GT. Another manufacturer, Audi, engineered various components of austenitic stainless steel into its otherwise aluminum-intensive A6 series. The use of stainless steel in strategic components enables greater weight reduction than that which the vehicle would have in all aluminum. Volvo and Saab have designed austenitic stainless steel bumper systems that also serve to reduce overall vehicle weight. While it is probably apparent to the reader that essentially any body component can be made in stainless and be made better in stainless, the question of when doing so is a better engineering decision involves economic considerations. Large automotive companies generally have large fixed investments in painting and coating systems to protect entire bodies from corrosion. The incremental savings of eliminating coatings

on individual body components is thus essentially nil. However, if the entire system is stainless and the investment is avoided, then the initial cost of a stainless body actually can be lower than one in coated carbon steel. This is the experience of Italian bus manufacturers, who began in the 1980s using 304 stainless steel in buses. Now, buses are 80% stainless. Designers began the conversion to gain the normal advantages the stronger stainless gives: over 10% lighter weight and over 10% improvement in crash worthiness of the passenger compartment, the accompanying savings in fuel consumption, and the virtual elimination of body maintenance. With essentially the entire body now in stainless, coating and painting could be eliminated. A stainless bus body is shown in Fig. 2. This swung the economic pendulum to stainless in a major way. Now, not only was the long-term cost of operating the bus lower, but the initial cost of the bus was lower. The economic analysis is shown in Table 4 (Ref 2). The design key was to use rectangular 304 stainless structural tubing, which allowed strong, stiff sections to be welded into space frames. It is only a matter of time until this is improved on by the use of 201 (with 3 to 4% Ni instead of 8 to 9% Ni) to lower cost and cold working of the tubing to achieve higher strength levels.

Chapter 19: Automotive and Transportation Applications / 231

Microcars are now a familiar sight in Europe. These vehicles are prized for their ability to be driven and parked in very small or congested locations. Their economy of operation is also a major attraction. These considerations combine to make stainless the best material for many of their components. Figure 3 shows a stainless steel microcar frame. The design by the famous design house Pininfarina employs a stainless frame to give maximum torsional stiffness and crashworthiness while eliminating painting entirely.

Other Automotive Components Stringent emissions controls regulations, led in the United States by the state of California, have made manufacturers reexamine the suitability of polymeric fuel tanks. These tanks contribute more to the required maximum 2 g/day of hydrocarbon emissions than is tolerable, so a

Fig. 2

Stainless steel bus bodies. Source: Ref 2

few manufacturers, such as Volkswagen, have installed stainless steel fuel tanks in their vehicles (Ref 2).

Trucks Over-the-road trailers are an excellent example of stainless steel being used for utilitarian purposes. Trailers used for hauling foodstuffs or corrosive materials are now constructed almost entirely of stainless steel; lined carbon steel tanks are now largely obsolete. The engineering basis for this is the same as for buses: high strength, no coating costs, and a product with long life and low maintenance costs. Structural members in trailers are typically 304, while tanks may also be 316L for corrosion resistance when the transported material requires it. Tank wrappers are often made of bright annealed and buffed 304. Manufacturers of trailers would be well advised to consider upgrading to duplex grades such as 2003 or 2205 or to cold-worked austenitic stainless, which would permit major weight reduction. This weight reduction would directly translate into greater load-carrying capacity because the payload of liquid-carrying trailers is limited by total gross weight. The ability to add a few thousand more pounds of payload would quickly pay back a small premium in material cost. Normal cargo-carrying trailers also use some stainless where corrosion is problematic, such as in doors and door frames. Weight reduction is less important in these trailers, which reach maximum load at a volume limit rather than a weight maximum.

Table 4 Life-cycle cost calculation (LCC) for stainless steel versus carbon steel for a bus application Cost of capital Inflation rate Real interest rate Desired LCC duration Downtime per maintenance/replacement event Monetary unit Value of lost production Material costs Fabrication costs Other installation costs Total initial costs Maintenance costs Replacement costs Lost Production Material-related costs Total operating cost Total LCC cost Source: Ref 2

Stainless steel 3.331 25.322 2.185 30.838 0 0 0 0 0 30.838

10.00% 5.00% 4.76% 20.0 years 1.0 day U.S. $ 101 U.S. $/day Carbon steel 1.391 28.582 4.050 32.023 1.448 2.897 57 0 4.402 36.425

232 / Stainless Steels for Design Engineers

Fig. 3

Microcar frame fabricated from stainless steel. Source: Ref 2

Rail Transport Passenger trains have exploited the high strength-to-weight and toughness qualities of the 301 family of stainless steels for many years. The corrosion resistance of these alloys makes them corrosion free in long use, obviating the need for painting and lowering maintenance costs. As with any other major use of type 301, a 5 to 10% increase in economy could be achieved if type 201 were used instead of 301. No loss in performance would occur. The transition to 201 has not occurred simply because of inertia and resistance to change on the part of designers and producers. Hopper cars made of 12% Cr martensitic stainless steels, typically 409Ni and 3Crl2, have excellent abrasion and corrosion resistance as well as high strength and therefore greater loadcarrying capacity. Curiously, the use of coldworked austenitic stainless in railcars, which has

been successful for decades, has not been carried over into trucks and buses even though it is technically feasible to economically produce structural sections in very high-strength stainless.

REFERENCES

1. R. Andersson, E. Schedin, C. Magnusson, J. Ocklund, and A. Persson, The Applicability of Stainless Steels for Crash Absorbing Components, ACOM, No. 3–4, AvestaPolarit AB, 2002 2. F. Capelli, V. Boneschi, and P. Viganò, “Stainless Steel: A New Structural Automotive Material, Vehicle Architectures: Evolution Towards Improved Safety, Low-Weight, Ergonomics, and Flexibility,” paper presented at Florence ATA 2005, 9th International Conference (Florence), May 2005, www.centroinox.it, accessed June 2008

Stainless Steels for Design Engineers Michael F. McGuire, p 233-242 DOI: 10.1361/ssde2008p233

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 20

Commercial and Residential Applications Summary STAINLESS STEEL HAS BECOME the essential material for products related to food, health care, and laundry because it combines strength and durability with an unexcelled ability to be cleaned, disinfected, and sterilized. These qualities have long been apparent to commercial food, laundry, and health care professionals and have increasingly carried over into equivalent domestic areas as consumers have become more aware of the benefits of stainless.

Introduction The last 20 years have seen the long-standing pervasive commercial use of stainless steel for food preparation and serving; laundry; heating, ventilation, and air conditioning (HVAC); and other appliances penetrate the domestic market for the same types of goods. Whether this is a fad of an increasingly affluent consumer or a reflection of more design engineers and consumers being more interested in lasting value than they were in the “throwaway” society that preceded that period remains to be seen. Stainless has been increasingly identified with highquality, high-end products. But, the case for value rather than fad seems to be stronger if the lessons of the harshly pragmatic automotive industry, in which decorative use of stainless has virtually disappeared while utilitarian uses have mushroomed, are any indication. The case for using stainless in appliances of all types, whether they are commercial or residential, relates to stainless being able to provide the best value over the intended service life.

Stainless is without rival for ruggedness and durability. Steel and aluminum corrode. Glass, stone, and ceramics break. Plastic is weak. The second, even more important, reason is that stainless steel is essentially benign from a hygienic viewpoint. Stainless steel itself is inert, both chemically and biologically, with respect to food. Further, it provides minimal harbor for unwanted biologic growth as do more porous materials. Stainless also competes quite well esthetically with other materials, offering the designer numerous surface finishes. Last, stainless is very amenable to nearly all manufacturing techniques. Its lack of need of coatings often makes components made from stainless less expensive to produce than equivalent designs that must be coated with paint, porcelain, or metal.

Food Contact Qualifications Setting aside cost, esthetics, and manufacturing considerations, a food contact material must first meet three criteria: It must be chemically inert, biologically inert, and cleanable and able to be disinfected. Chemical neutrality is achieved by a material when the material does not enter into the food with which it comes in contact. This has become an increasing concern as the effects of ions or chemicals released from food preparation materials have been viewed as potential toxic or disease agents. Medical knowledge is not sufficiently advanced to convince consumers of the harmlessness of such contaminants, so it is preferable to demonstrate the absence of contamination if one is to win the public confidence in a food contact material.

234 / Stainless Steels for Design Engineers

Stainless steel contains many constituent elements. Were they to enter the food with which the stainless came in contact, then stainless would be a poor food contact material. The distinguishing characteristic of stainless, however, is the spontaneous passive film, which is so stable chemically. This film acts as a barrier to corrosion, which would result in metal release. Stainless therefore is effectively inert. Tests have been made of the rates at which metal ions can enter foodstuffs (Ref 1). Table 1 shows the vastly lower rates of metal ion release from stainless than from aluminum and carbon steel, both of which are permissible, if not optimal, food contact materials. Aluminum releases aluminum ions into solution of both cooking oil and 3% acetic acid at nearly equal rates of 15 mg/cm3 in 30 days. Carbon steel releases iron at over 100 mg/cm3 in the same period. Stainless, however, releases less than 0.010 mg/cm3 of iron. This is complemented by other studies showing that the transfer of ions from food contact vessel to food is diminishingly and negligibly small. Tests have been conducted on stainless steels, types 304, 439, and 444, that had both industrial finishes (2B and BA) as well as freshly abraded and air-aged finishes. These sample steels were subjected to boiling solutions of oils, alcohols, water, and 3% acetic acid. None caused the transfer of either chromium or nickel to exceed the statutory 0.1 ppm level (Ref 2). Nickel levels of various foods before and after cooking have been scientifically measured to assess the possibility of leaching of that ion from 304 stainless steel (Ref 2). No increase was noted from the natu-

rally occurring level of up to 0.3 μg/g in cereals to 1.1 μg/g in meat and fish. These negligible levels of leaching simply indicate that foodstuffs are a benign chemical to stainless steel. Nevertheless, it is necessary to apply the correct assessment of the corrosivity of the foodstuff in question. In food production, as opposed to preparation for serving, more extreme levels of acidity and salinity can be encountered. Nippon Steel reported (Ref 3) that materials used in the manufacture of soy sauce, which can have 15% salt, must withstand prolonged contact at 45 °C (115 °F). Under such conditions, 316 stainless pits in about 1 day, while higher alloy grades, the 6Mo alloys, of which their YUS 270 is one (equivalent to UNS S31254), are projected to last 20 years before pitting. This is significant because only pitting corrosion is likely to release metals ions into a food substance. So, while guidelines exist for the minimum alloy content permissible for normal food contact, such as those promulgated by the National Science Foundation (NSF), one must still verify the corrosion due to a particularly aggressive food ingredient. Choosing the proper grade of stainless, based on pH, salinity, and temperature, is the responsibility of the design engineer. Referred to chapters in this book on both corrosion and individual alloy families for guidance in choosing an alloy based on resisting pitting in a given environment. This having been said, no alloy greater in pitting resistance than 304 is required in residential or commercial cooking food contact. The higher alloy requirements come from the more aggressive

Table 1 Net metal migration into acetic solution (3%) Metal migration during indicated time, µg/cm3 Material

Austenitic stainless

Ferritic stainless

Aluminum

Carbon steel

Source: Ref 1

Time

30 min 10 days 20 days 30 days 30 min 10 days 20 days 30 days 30 min 10 days 20 days 30 days 30 min 10 days 20 days 30 days

Iron

2.4 4.2 2.7 2.3 3.0 7.3 8.6 6.6 4.9 18.2 17.9 31.3 8,430 57,700 62,900 112,000

Chromium

0.12 0.22 0.22 0.28 0.43 0.40 0.71 0.87 0.93 3.42 5.58 12.40 0.62 7.40 6.82 14.00

Aluminum

≤ 0.19 0.22 0.19 ≤ 0.19 ≤ 0.19 0.19 ≤ 0.19 ≤ 0.19 930 5,300 7,160 15,350 2.7 26.7 24.0 36.9

Nickel

≤ 0.12 ≤ 0.12 ≤ 0.12 0.31 ≤ 0.12 ≤ 0.12 ≤ 0.12 ≤ 0.12 ≤ 0.12 ≤ 0.12 ≤ 0.12 0.22 ≤ 0.12 ≤ 0.12 ≤ 0.12 ≤ 0.12

Chapter 20: Commercial and Residential Applications / 235

concentrations and exposure periods that can be found in food-processing plants. Certain abuses can even damage stainless cookware. Very high temperatures, such as can occur when unattended pans have their liquids boiled away, can damage stainless but are more harmful to less-rugged alloys such as copper and aluminum. Stainless steel is primarily composed of iron, chromium, and nickel along with small amounts of manganese, silicon, and molybdenum. It contains trace amounts of copper, aluminum phosphorus, and sulfur. Each of these elements is naturally occurring in food. Each can be found in a typical multivitamin/multimineral supplement. Stainless is essentially devoid of heavy metals, such as lead and mercury, which are vaporized at the temperatures at which stainless is refined. Even if toxic metals were somehow to be made to contaminate stainless, the passive film would prevent their release. All these factors combine to make stainless the most chemically neutral metal found in food contact. Alternative alloys, such as copper and aluminum, actively leach into foods. Copper and aluminum have been linked to but not demonstrated to cause Alzheimer’s disease. Biological Neutrality. Microorganisms adhere to solid surfaces. When a clean surface comes in contact with food, a surface deposit is formed from the food. The film may also contain molecules left from previous cleaning and disinfecting. The formation of this film is presumably influenced by material characteristics such as roughness, although there are no specific studies on this. However, microorganisms adhere to this film and, as colonies of them grow, form a biofilm. This film consists of layers of microorganisms that can produce an exocellular polymeric matrix, which protects the colony from cleaning and disinfecting. Geometric factors also can protect these colonies. Rough surfaces are intuitively more difficult to clean. The ability to maintain a microscopically smooth surface is an asset in stainless that polymeric, enamel, and mineral surfaces lack. Stainless steel is much less roughened by abrasion, keeping the surface smooth (Ref 4).This will be seen to influence its ability to be cleaned and disinfected. There is some technology to go beyond biological neutrality in the use of coatings that actively discourage or eliminate growth of microorganisms. Polymeric coatings impregnated with silver ions have been developed and com-

mercialized (Ref 5). Silver ions, like copper ions, are powerful antimicrobial agents. The combination of such a coating with stainless as a corrosion-proof substrate may represent the maximum in hygienic and chemical protection and is already being used in medical applications where such concerns exceed those in ordinary food contact situations. Cleanliness. A necessary quality in any material considered for food contact is the ability to be cleaned. This includes the removal of both organic and inorganic substances. The most important objective of cleaning is to remove the visible and invisible materials that can provide a growth medium for microorganisms. This process is distinguished from disinfection, which is the reduction of the microbial population to a satisfactory level. What this level is depends on the standards of hygiene in force. And, although cleaning can and does reduce the population of microorganisms, true bacteriological cleanliness is obtained only after disinfection. The combination of cleaning and disinfecting is important. Studies have shown that the efficacy of disinfectants is weaker on bacteria that have been established in a surface biofilm than on bacteria in suspension. The most complete form of disinfecting is sterilization, whose objective is the complete removal of all microbial life and viruses. The purpose of cleaning stainless steel is to rid it of contamination. Various stainless manufacturers and associations have identified a number of effective of cleaning products (Ref 2, 6, 7): • Alkalines, which dissolve fats and oils • Chelating or sequestering agents, which agglomerate contaminants. These are often organic acids such as citric acid or oxalic acid and amine acids such as sulfamic acid and ethylene diamine tetraacetic acid (EDTA) or salts of these compounds. • Hydrocarbon solvents • Water with soap, detergent, trisodium phosphate, or other surface active agents, which emulsify • Dilute oxidizing acids like nitric acid • Mild acids such as phosphoric acid The effectiveness of a cleaner relates mainly to the contaminant to be removed. Some trial and error may be required for a given contaminant. Some precautions are worth mentioning. Abrasive cleaners should be used with caution. The abrasive size and hardness must be chosen

236 / Stainless Steels for Design Engineers

so that the stainless surface finish is not affected in an unwanted manner. If the abrasive is harder than the stainless or coarser than the stainless surface roughness, the underlying finish can be disturbed. Care should also be taken to clean with the polish grain if a polished surface is being cleaned. Also, cleaners containing chlorides are common. Their use is not recommended on stainless. Use of hydrochloric (muriatic) acid is especially detrimental. If chloride-containing cleaners are used, then thorough rinsing should be conducted to avoid chloride concentration through evaporation, especially in crevices. Steel wool or steel brushes should not be used on stainless under any circumstances as iron residue interferes with the integrity of the passive film. The ability of stainless to be cleaned is best measured by the actual removal of bacteria colonies. This has been done to compare unabraded and abraded (to simulate new vs. used) stainless steel, enameled steel, mineral resin, and polycarbonate materials, which can be used for sinks, counters, food prep tables, etc. (Ref 4). Figure 1 shows that the reduction in bacteria count by the same cleaning technique is ten times more effective on stainless than on the other material types. Abrasion did not degrade the ability of stainless to be cleaned as it did softer materials. The surface of stainless, even with the seemingly protected recesses due to

Fig. 1

abrasive polishing, permits bacterial colonies to be removed. The greater roughness of the other materials may serve to protect the bacterial colonies from shear forces and provide greater specific surface area on which the colonies can bond. Disinfection. The ability of a surface to be disinfected is measured by the concentration of a given disinfectant required for a specific reduction in bacterial population. Numerous studies have been published (e.g., Ref 8–10) showing that glass and stainless steel have equal aptitude for disinfection, and that polyesters, polyurethanes, rubber, and aluminum all required about one to two orders of magnitude greater concentrations of disinfectant for the same result. These results indicate why stainless is so essential to the food industry. Stainless can be disinfected quite readily, which allows the great invisible liability of food-borne diseases to be minimized. The effectiveness of sodium hypochlorite as a disinfectant is inarguable, also. So, despite its potential corrosivity, it will be commonly used. Taking this into account requires that commercial and residential food equipment be able to withstand some chloride level greater than otherwise projected. Industry practice in the United States has shown that corrosion problems occur at an unsatisfactory level with mechanically polished 430 but not with bright-annealed 430.

Bacterial retention as a function of material and cleaning time. Source: Ref 4

Chapter 20: Commercial and Residential Applications / 237

Thus, alloys with less than 16% Cr should not be used unless corrosion can be accepted. Alloys containing 16% Cr can be used with optimal surface finish. Alloys as low in carbon as 12% are used for cutlery applications where slight corrosive attack can be accepted. This is a necessary trade-off required to achieve high hardness for good cutting edge retention. Higher chromium grades such as 304 can be used even with mechanically polished surface finishes. From a cost-effectiveness point of view, there is no reason to use more expensive alloys than 430 or 201 in the vast majority of commercial and residential kitchen and laundry applications from a corrosion standpoint as long as surface finishes that have not been produced by abrasive polishing are specified. Many such finishes are widely used. In North America, the rolled-on replicas of No. 4 finish, Koolline, Lustrite, etc., are quite common, while in Europe the bright-annealed finish has been preferred. Both of these are preferable to mechanically polished finishes. The food industry, an immense consumer of stainless steel, could do more than any other industry to help conserve nickel by specifying alloys such as 430, 439, and 201 as their standard alloys as well as by specifying nonabrasive finishes. This can be done with no loss of functionality or change of appearance and could save 23% to 50% in material cost.

Applications Cookware. Any interaction between a food contact material and the food is most likely to occur during the cooking process when temperatures are greatest. Only glass and stainless are excellent food contact materials. And, since cookware must be flexible enough to handle any potential food, the choice of material for cookware must be the most conservative. For this reason and because of the brittleness of glass, stainless is the material of choice. The qualities discussed make aluminum and copper less desirable. Both leach into food. Copper can be tinned to combat this. The tin also corrodes over time but has very low toxicity. A larger drawback is the expense of retinning copper utensils. Aluminum is known as a toxic metal, with its toxicity causing symptoms similar to those of Alzheimer’s and osteoporosis (Ref 11). These two metals do have one advantage over stainless, however: their thermal conductivity. High

thermal conductivity in a cooking utensil minimizes differences in temperature across the surface in contact with the food, permitting better control of the cooking process. The solution to the problem of thermal conductivity is to make composite materials. Stainless can be bonded to copper and aluminum, which allows the stainless to be on both the food contact surface as well as the exterior, with an inner layer of copper or aluminum effectively spreading the heat. Aluminum and copper are nearly equally effective as inner conductive layers. Premium cookware features them both. The “sandwich” is the optimal design because it optimizes heating uniformity even more than using aluminum or copper alone would since the high conductivity inner core functions as an isotherm. The uniformity is the more important consideration than the absolute thermal conductivity or even the thermal diffusivity. In a triple layer, the choice of the non-foodcontacting stainless is less stringent. Sometimes, the exterior is made of a ferritic stainless steel. The ferromagnetism of ferritic stainless steel makes it ideal for induction heating. Alloys such as 436 have been used for this application, while 304 is the pervasive choice for the food contact surface. This is despite the fact that 201 or 301 are quite adequate for this application. It is also possible to produce a magnetic carbon steel core with stainless bonded to both sheet surfaces. The exposed edges are rolled to shield them from corrosion. Nonstick coatings, such as polytetrafluorethylene (PTFE), are very popular because of their nonstick qualities. Above 350 °C (660 °F) these coatings give off toxic fumes. This is a danger for certain types of cooking, such as wok cooking or blackening, but more likely to be encountered by accidentally high temperatures above those intended. Since they can be scratched and are not impermeable, their use does not alter the choice of the material to which they are applied. Kitchen Appliances. Every type of commercial kitchen appliance can be, and usually is, made of stainless steel, as are premium domestic kitchen appliances. This choice is based on durability and ease of cleaning and disinfecting. And, because many commercial appliances are visible to the customer, aesthetics are also a driving force. Choice of alloy for a given appliance is a crucial cost factor. As was noted that 430 is marginal for kitchen use, because of the prevalence of chloride-containing cleaners, unless it has been bright annealed. All austenitics

238 / Stainless Steels for Design Engineers

are satisfactory under normal use. Designers seem to generally neglect the possibility that their equipment may be used in coastal climates. In the high ambient salinity of coastal climates, corrosion will occur unless 304 with a bright-annealed finish or a brushed finish rolled onto a bright-annealed 304 is used. Mechanically polished 304 stainless is inadequate for coastal environments. These are the same guidelines used for architectural applications. Figure 2 shows how different alloys withstand coastal conditions. The corrosion on 430 would be considered excessive, while that on the 201 and 304 is acceptable given that some routine cleaning would have prevented the corrosion that is present on these samples, which were exposed to coastal salt and humidity for 10 years in North Carolina (Ref 12). In the vast majority of ambient conditions, coastal salinity is not a problem. This applies to inland conditions or coastal conditions where interior environments are protected by air conditioning or adequate cleaning of the stainless is practiced. This is normally the case for commercial equipment. Under these conditions, 201 is quite adequate, and the use of 304 represents wasteful overengineering. This choice is supported by decades of use by the major manufacturers of commercial appliances. Many who are large enough to specify their desired grade on bills of materials rather than simply buying from service center inventories have routinely used 201 and realized an approximately 8% lower cost before surcharges. Use of 201 versus 304 reduces surcharges by almost 50%, which can be a much larger savings than the base price savings. Smaller manufacturers are often precluded from these savings because of the general, if inexplicable, practice of service centers not stocking 201 despite its being the most costeffective general-purpose stainless grade. The extended nickel price elevation from 2004 onward has a good chance of changing that situation as end users rebel against surcharges, which cannot be passed on to their customers. It has been pointed out that there is an array of 201-type grades, and that this is a drawback to their wider adoption. I recommend following American Society for Testing and Materials (ASTM) A240 and specifying UNS S20100 when substituting for 304 as this has very similar performance in forming, welding, and appearance to 304 and can be most easily interchanged without complications in manufacturing and field performance. For

Fig. 2

Stainless steel samples exposed on a North Carolina beach for 10 yr. Source: Ref 12

Chapter 20: Commercial and Residential Applications / 239

parts made by deep drawing, substitution is still very possible, but deep-drawn grades are more finely tuned to specific process paths and must be more tightly specified than general-purpose grades. The more commonly used alloys for appliances are listed in Table 2. The greatest savings comes, of course, from using ferritic grades, and they should be used whenever forming requirements permit, which is the majority of the time, since most appliance components experience little more than cutting, bending, and welding. There are important precautions, however. Mechanical polishing results in unacceptable corrosion resistance, and the low work hardening rate of ferritics causes the mechanical polish to take on a different color shade. This subtle difference can be magnified to objectionable levels when a mechanically polished ferritic stainless, such as 430, is put side by side with an austenitic such as 201 or 304. This can be solved by specifying rolled-on finishes, which look the same on ferritics and austenitics. These finishes also supply the added corrosion resistance that makes alloys such as 430 acceptable. It is still preferable to use a dual-stabilized grade such as 468, which can be welded without adverse corrosion effects and has high formability and corrosion resistance at as little as half the cost of 304 when alloy surcharges are factored in. Use of dual stabilization permits keeping titanium levels to a minimum, making it possible to avoid TiN-caused surface defects, which occur if significant TiN precipitation occurs before solidification in the original steel production. This occurrence is strictly a thermodynamic phenomenon related primarily to the titanium and nitrogen levels, which should be minimized so that the product of titanium times nitrogen is less than 0.0025 when Table 2

concentrations are in weight percent. This is difficult to achieve for 17% Cr alloys if stabilization is by titanium alone. Interior or working parts of appliances, to the degree they require high cleanability or contact food, are also often made of stainless. This is especially true of dispensing machines, such as for beverages, ice cream, and ice. Stainless interiors are often found in refrigerators and dishwashers. In the case of dishwashers, forming requirements are often severe enough to require the use of austenitic stainless. Rolled-on finishes are generally preferred. Not a small reason for this is that this finish requires only a single temper pass to both flatten and provide the finish. This yields very consistent forming characteristics, meaning much lower breakage during press-forming operations. Rolled-on finishes also have very high visual consistency, which is usually a very important quality criterion for appliance manufacturers. Canisters, chafing dishes, serving pans, etc. are generally made from austenitic stainless steel, which lends itself to the typical deepforming operations used in their manufacture. Coatings are rarely used. If antimicrobial coating were to be used in food contact, this would be an ideal application since already cooked food is most often in the intermediate temperature danger zone at which bacteria can multiply. Food preparation tables also fit into this category. Appliance facades are increasingly using stainless. These include refrigerators, stoves, microwaves, drawers, etc. Shelves and exhaust hoods also benefit from being made of stainless. The drivers here are cleanability, durability, and esthetics. There are important visual considerations in these applications. Consistent,

Stainless steels commonly used for appliances Composition, %

Alloy

UNS No.

C

N

Cr

Ni

Mn

Si

201 301 304 316 430 439

S20100 S30100 S30400 S31600 S43000 S43035

0.15 0.15 0.08 0.08 0.12 0.07

0.25 ... 0.10 0.10 ... 0.04

16.0–18.0 16.0–18.0 18.0–20.0 16.0–18.0 16.0–18.0 17.0–19.9

3.5–5.5 6.0–8.0 8.0–10.5 10.0–14.0 0.75 0.50

5.5–7.5 2.00 2.00 2.00 1.00 1.0

1.00 1.00 1.00 1.00 1.00 1.0

468

S46800

0.030

0.030

18.0–20.0

0.50

1.00

1.00

436 444, YUS190 29-4C

S43600 S44400

0.12 0.025

... 0.035

16.0–18.0 17.5–19.5

... 1.0

1.00 1.0

1.00 1.0

S44735

0.025

0.025

28.0–30.0

0.5

1.00

0.75

Mo

... ... ... 2.0-3.0 ... ...

Ti/Nb

... ... ... ... ... 0.20 + 4 × (C + N), to 1.10 ... Ti + Nb: 0.20 + 4 × (C + N), to 0.80 ... Nb + Ta: 5 × C, to 0.70 1.75–2.50 Ti + Nb: 0.20 + 4 × (C + N), to 0.80 3.5–4.5 Ti + Nb: 0.20 + 4 × (C + N), to 0.80

240 / Stainless Steels for Design Engineers

defect-free surface finishes are paramount. This again can really only be achieved by rolled-on finishes since abrasively polished finishes vary excessively in roughness, reflectivity, and color. Panel flatness is often very important and another benefit from rolled finishes. If visible welds are required, as is often the case for products such as hoods and counters, then special finishes with very long polish grains have a major advantage in that the weld can be ground and polished with a belt sander of the appropriate grit size so that the weld blends imperceptibly with the adjoining original surface. This is a practical impossibility with abrasively polished finishes and very difficult with rolled finishes. Freedom from fingerprinting can be another valuable attribute for faÁade applications. This can be obtained on stainless by the mill application of a thin, bonded polymer film. All bare stainless finishes show fingerprints. With uncoated stainless, it is best avoided by using mineral oil-based cleaners. Although very high alloy stainless steels are used for high-temperature kitchen applications, such as heating element sheathing (American Iron and Steel Institute [AISI] type 334), it is seldom used for oven interiors because it does take on a heat tint when exposed to temperatures above 300 °C (570 °F). Range tops, which see lower temperatures, are normally stainless. Outdoor cooking grills, because they must endure exterior environments without corrosion, are almost always 304 or a similar grade. Heat tint does not occur with these to a problematic degree. Gas burner manifolds are also stainless. In this case, ferritics are required because of the need for extreme high-temperature oxidation resistance and the desirability of a low coefficient of thermal expansion. The preferred alloys are those developed for automotive exhaust systems, variations on 409 and 439. Flatware and cutlery were among the original uses of stainless. Stainless filled the gap between carbon steel, which was hard but whose rusting was an obvious problem, and silver, which was soft and whose cost prohibited its use to all but the wealthy few. Cutlery is the domain of martensitic stainless steel. The corrosion resistance of martensitic grades cannot be improved above modest levels, never reaching that of 304, but this criterion is secondary to hardness because of the need to keep a sharp cutting edge. Maximum corrosion resistance is achieved in the as-quenched condition. But, some toughness is a valuable but not crucial

characteristic, so most cutlery is tempered at low temperatures. The vast majority of requirements for high-quality cutlery are satisfied by 420 stainless. If greater cutting edge retention is desired, then more or harder carbides are engineered into the martensitic matrix. This is done by adding more carbon and chromium, as is found in 440A and to a greater extent in 440C. The wear resistance added by carbides is proportional to their hardness and amount. The chromium carbides of these straight-chromium martensitic stainless steels are very hard, 1800 HV, versus the 1100 HV hardness of iron carbides. The addition of higher levels of carbon ties up chromium so that it cannot add to corrosion resistance, however, so that it can become barely rust resistant. Furthermore, at high carbon levels, carbides precipitate in the liquid and are much coarser. These large carbides can pull out during edge honing, making a ragged rather than a fine, smooth cutting edge. However, vanadium and tungsten have even harder carbides, 2800 and 2100 HV, respectively. Through conventional casting and hot working, only a small amount of these carbides can be introduced into the matrix. The problem is that if primary carbides form during solidification, they tend to be coarse and to embrittle the alloy. Hard particles are much more useful for wear resistance if they are small and widely dispersed. To a degree, this refinement of the primary carbides can be achieved by raising nitrogen levels. These problems can be circumvented by the use of powder metallurgy, which permits the solidification step on a macroscale to be skipped. Larger volume fractions of hard carbides such as vanadium carbide and tungsten carbide can be added and dispersed. Table 3 lists the martensitic alloys used for cutlery. It is reasonable to say that most of these alloys far exceed the requirements of food preparation. Improved corrosion resistance of these alloys is achieved by adding molybdenum at the expense of chromium, which would cause excessive δ-ferrite retention if it were raised. This can be seen in alloys above the basic 420. Flatware has no hardness requirement, so grade selection is based on the need for perceived quality. At the high end is 304, which has all the corrosion resistance that could be needed for flatware. However, type 301 is commonly used also, as are ferritic steels, such as 430, for low-cost flatware. Depending on the shape of the final utensil, material is stamped or forged and then finished.

Chapter 20: Commercial and Residential Applications / 241

Table 3 Stainless steels commonly used for cutlery Composition, % Alloy

420 4116 440A 440C BG-42 ATS-34 14-4 CrMo 154 CM CPM S30V CPM S60V CPM S90V

Designation

UNS S42000 DIN 1.4116, nominal UNS S44002 UNS S44004 Nominal composition Nominal composition Nominal composition Nominal composition Nominal composition Nominal composition Nominal composition

Form

Wrought Wrought Wrought Wrought Wrought Wrought Wrought Wrought PM PM PM

C

Mn

S

Si

Cr

Mo

Ni

Other

0.15 min 0.50 0.60–0.75 0.95–1.20 1.15 1.05 1.05 1.05 1.45 2.15 2.20

1.00 ... 1.00 1.00 ... 0.4 0.5 0.45 ... 0.40 ...

0.030 ... 0.030 0.030 ... ... ... ... ... ... ...

1.00 ... 1.00 1.00 0.3 0.35 0.3 0.3 ... ... ...

12.0–14.0 14.5 16.0–18.0 16.0–18.0 14.5 14.0 14.0 14.0 14.0 17.0 13.0

... 0.65 0.75 0.75 4.0 4.0 4.0 4.0 2.0 0.40 1.0

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

... 0.15 V ... ... 1.2 V ... ... ... 4.0 V 5.5 V 9.0 V

Many kitchen utensils are also made entirely or in part with stainless. Type 304 is the alloy most commonly used, but again any of the stainless steels with at least 16% Cr are adequate, and grade selection depends on forming and joining requirements. Laundry appliances have converted significantly to stainless. This trend began in Europe with the development of high spin speed, horizontal axis washing machines. These washers use far less water and energy to achieve higher levels of cleaning with less damage to clothing. These features have eroded the share enjoyed by vertical axis, agitator-type washers, whose low speed allows them to be constructed of lowstrength materials such as plastic or porcelaincoated steel. The stresses induced by the high spin speeds, which are necessary in horizontal axis machines to take water removal from 80% to 95%, require the strength of stainless steel. Porcelain-coated carbon steel obviously can be strong, but the coating is cracked by strains that the steel itself easily tolerates. An additional benefit to stainless over porcelain is that stainless starts smooth and becomes even smoother with use, while porcelain becomes quite abrasive over time as wear opens voids with edges that can be quite sharp and cause significant damage to clothing. Washer tubs and drums are made of both ferritic and austenitic stainless. The selection is based on forming requirements rather than corrosion or strength. If components can be made by bending rather than stretching, then the lowercost ferritics can be used. Ferritics should be a 17% stabilized grade, such as 439 or 468, and austenitics can be 201, 301, or 304. Unstabilized ferritic alloys, such as 430, should never be used in welded applications. Dryers are less challenging, and it is difficult to make a strong case for the functional value of stainless. Those designs that use stainless will last longer and be gentler to

clothing. Those, along with the implied quality of stainless, are the main drivers for its use. Heating and Water Heating. With the development of high-efficiency, natural-gas-fired, forced-air furnaces, stainless has come into domestic use as a heat exchanger material. These furnaces gain their extra efficiency by condensing water from combustion gas exhaust. This condensate can, depending on incoming air, contain corrosion elements, which has led to the use of very highly alloyed ferritic stainless steel in their construction. Alloy 29-4C (UNS S44735) was the original alloy used nearly universally in the United States. The worst consequence of perforation by pitting could be the release into the home of toxic gas, so pitting corrosion must not be allowed. The intermediate efficiency furnaces (80 to 90%) require the use of corrosion-resistant vent pipe to prevent corrosion from condensation in the flue. High-temperature plastics were tried, but failed joints in them caused their recall after several fatalities were reported. High-efficiency (90% or higher) furnaces can use low-temperature plastic pipe, but these units require the use of a corrosion-resistant secondary heat exchanger to recover the latent heat of vaporization of the water from combustion. Alloy 29-4C was the original choice for most secondary heat exchangers, but at least one used alloy 6XN (UNS N08367) alloy for formability. Some manufacturers have always used lower-alloyed stainless steels. The proper handling of combustion products is an interesting problem in materials selection. The variability of the use environments leads to a huge spread in corrosion conditions and materials performance. In the end, one has to balance materials selection between cost (fortunately, 29-4C alloy is nickel free) and probability of failure. Given the number of units produced and

242 / Stainless Steels for Design Engineers

the potentially serious consequences of failure, failure rates must be less than 10 – 4, while failure rates much less than 10 – 6 are impossible to verify and hard to justify. In any case, the competition is always between various stainless steels. Galvanized steel will not work. The issue is difficult enough for natural-gas-fired furnaces. Oil fired is a developing situation for which there is no good consensus. Wood burners and other unconventional furnaces (such as corn burners) present additional challenges, and answers are even less obvious. Water heaters are sometimes made of stainless steel. It is not uncommon for water to have a sufficient level of chlorides to lead to stress corrosion cracking if an austenitic stainless is used. Therefore, the recommended alloy for this application is UNS S44400. More recently, lean duplex alloys have been developed, such as 2101 and 2003, which can perform quite well without corrosion or stress corrosion cracking. More highly alloyed duplex alloys such as 2205 are more expensive but would work well.

REFERENCES

1. M.J. Julio, M.L. Martin, and J.M. Baena, Cation Migration Tests in Metal Containers, Innovation Stainless Steel (Florence), Oct 1993, p 1.221–1.226 2. “Stainless Steel in Contact with Food,” Document Ugine, June 1996

3. “The Application of High Corrosion Resistance Stainless Steel YUS270 in Food Processing Facilities and Equipment,” Nippon Steel Technical Report 87, Jan 2003 4. J.T. Holah and R.H. Thorpe, Bacteria Retention on Cleaned Surfaces, J. Appl. Bacteriol., Vol 69, 1990, p 599–608 5. Agion Technologies, www.agion-tech.com, accessed June 2008 6. Removal of Stains and Discolourations, Outokumpu, www.outokumpu.com, accessed June 2008 7. “The Care and Cleaning of Stainless Steel,” Specialty Steel Industry of North America, www.ssina.com, accessed June 2008 8. E.P. Kysinski et al., J. Food Processing, Vol 55, 1992, p 246–251 9. A.A. Mafu et al., J. Dairy Sci., Vol 73, 1990, p 3428–4332 10. P. Gelinas and J. Goulet, Can. J. Microbiol., Vol 29, 1983, p 1715–1730 11. R.A. Goyer, Toxicity of Metals, Properties and Selection: Nonferrous Alloys and Special-Purpose Materials, Vol 2, ASM Handbook, ASM International, 1990, p 1233– 1269 12. Allegheny Ludlum research, as presented in D.S. Bergstrom and C.A. Botti, AL 201HPTM (UNS S20100) Alloy: A HighPerformance, Lower-Nickel Alternative to 300 Series Alloys, Stainless Steel World, KCI Publishing, 2005

Stainless Steels for Design Engineers Michael F. McGuire, p 243-246 DOI: 10.1361/ssde2008p243

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 21

Marine Systems Applications Summary AS RECENTLY AS THE 1960s AND 1970s, handbooks on stainless steel were stating that “stainless steels are not stainless in seawater,” and “successful prolonged corrosion-free service of stainless steel in seawater requires sophisticated corrosion engineering, or enormous good fortune” (Ref 1). The advances in stainless steel made since then have thankfully made these statements obsolete. Not only have basic corrosion problems been solved, stress corrosion cracking also can be avoided. More impressively, this can be done with alloys with strengths much higher than those of the alloys, such as 316, that they replace and that have been only marginally successful in marine environments. The inertia in changing from the weaker, less-corrosion-resistant, more expensive austenitic stainless steels is large because of less availability of the newer, better alloys, and lack of familiarity with their benefits. Those who understand and use these newer duplex alloys will be rewarded. This chapter reviews the major marine applications of stainless steels, including desalination equipment, shipping containers, and heat exchangers that handle seawater.

Desalination At one time not long ago stainless steel was thought to be an inadequate to marginal material for use in seawater. Its use in heated seawater was therefore all the more suspect. This was first changed with the development of superferritic and superaustenitic alloys. The superferritic alloys such as Seacure (UNS S44660) and 29-4C (UNS S44735) are quite resistant to seawater, even at high temperature. Their low toughness

restricts their use to items of rather thin gauge, less than about 1.0 to 2.0 mm, depending on alloy. Thus, their use is limited to tubing. Superaustenitic alloys can be used at any thickness, although they are a costly material. The success story for stainless steel and seawater and therefore desalination is that of duplex stainless steel. With the same corrosion resistance as any super austenitic or superferritic alloy, it has nearly double the strength plus resistance to stress corrosion cracking. And while duplex stainless steel is not a cheap material, it does contain much less nickel than an equivalently corrosion resistant austenitic stainless steels, which is a major cost saving factor. Desalination technology is relatively new if one ignores the fact that distillation has been around for a very long time. Desalination in commercially viable quantities began with multistage flash technology in the 1950s. The underlying principle of this process is the evaporation of water vapor from salt water with the subsequent condensation of the salt-free water vapor. In the multi-stage flash (MSF) approach feedwater is heated and the pressure is lowered so that the water “flashes” into steam. A variation on this technology is multiple-effect distillation (MED), another low-temperature distillation process. The differences in all distillation-based systems reduce to the efficiency of the design in minimizing energy consumed per unit of pure water output. All distillation processes require heating of the input water and some process power. The other basic engineering approach to desalination is reverse osmosis (RO). The invention of polymer membranes that could separate the salt ions from the water made this technology possible. No thermal energy is required. The water is pumped at high pressure through these permeable membranes physically separating the

244 / Stainless Steels for Design Engineers

salt from the water. The change in salt concentration across the membrane is a function of the pressure and the membrane itself. A second treatment may be used to improve water quality. The distillation methods require about 5 kWh/metric ton of water output, while the RO methods require twice that. The distillation methods require another 20 kWh of thermal energy from some source for feedwater heating, while the RO method requires none. Thus the ability to find energy from cogeneration or a source such as solar, etc may determine which process is preferred. Materials Selection for Desalination Materials used for distillation processes have evolved from use of type 316 (UNS S31600) stainless steel, first as lining and then as cladding. The superaustenitic alloys, the 6Mo variations, came next because they truly solved the corrosion problem, but at a price. Then, separately the duplex alloys were developed, with the first market the petroleum industry, whose demands and research made these alloys possible. It was not a stretch to see that high-strength alloys that could withstand seawater in offshore applications could do well on land as well. To give full credit, the pulp-and-paper industry was also beginning to employ duplex stainless steels

Fig. 1

for their processes. Type 316 stainless steel has passed from consideration as a material for handling brackish water or seawater. In distillation systems, the rule of thumb is that 2205 alloy (UNS S32205), with its pitting resistance equivalent number (PREN) of 35, is sufficient for seawater up to 20 °C (70 °F); alloys 2507 (UNS S32750) or Zeron 100 (UNS S32760) should be used for seawater at elevated temperatures or high salinity. For the output of fresh water, lesser alloying is required. Stainless steel types 304 (UNS S30400), 316 (UNS S31600), 2101 (UNS S32101), 2003 (UNS S32003), or even 439 (UNS S43035) may be used depending on the combination of salinity and temperature of the output water. Besides their high corrosion resistance for lower total alloy cost, the duplex stainless steels have higher strength, which is a significant factor since distillation plants are large. The use of duplex allows wall thickness reductions that bring about larger savings than those based solely on their cost per unit weight. Figure 1 shows the difference among the candidate stainless steels in corrosion resistance (Ref 2). The viable materials for seawater are those that can withstand roughly 20,000 ppm Cl– level at the appropriate temperature. The strengths of the various candidate materials are given in Table 1. These are typical values.

Corrosion resistance (pitting) as a function of salinity and temperature. 1. 304L (UNS S30403); 2. 316L (UNS S31603); 3. 2205 (UNS S32205); 4. 904L (UNS N08904); 5. 254SMO (UNS S31254). Source: Ref 2

Chapter 21: Marine Systems Applications / 245

Table 1

Typical analyses and properties of major marine alloys Composition, %

Alloy

2101 2003 2205 2507 304L 316L 317L 6XN 254SMO Zeron 100

Yield strength

UNS

Cr

Mo

N

Ni

PREN(a)

S32101 S32003 S32205 S32750 S30403 S31603 S31703 N08367 S31254 S32760

21.5 20.5 22 25 18 16 18 21 20 25

0.3 1.5 3 4 0 2 3 6 6 3.5 (+0.75 W)

0.22 0.18 0.17 0.27 0.05 0.05 0.05 0.22 0.20 0.27

1.5 3 5 7 8 10 14 24 18 7

26 29 35 42 18 24 29 45 43 42

MPa

515 515 515 550 220 220 230 380 380 550

Tensile strength

ksi

MPa

ksi

Elongation, %

75 75 75 80 32 32 33 55 55 80

650 725 760 800 520 520 540 760 750 750

94 105 110 116 75 75 78 110 109 109

40 40 35 35 50 50 45 45 45 35

(a) PREN, pitting resistance equivalent number.

Refer to the appropriate design code for your particular application to find minimum properties. The reader is cautioned that duplex longitudinal properties are slightly lower than the transverse properties that testing requires. Pumps for seawater follow the same guidelines as piping, tanks, and all other components. Cast or wrought duplex are the alloys of choice.

Shipping The major uses of stainless steel in shipping are in bulk storage containment. Cargos range from food and beverages to chemicals and liquid natural gas (LNG). Practice in the past has been to use austenitic grades of stainless with cathodic protection when necessary to address inadequate corrosion resistance. However, since 2000 marine chemical tankers have become the largest consumer of duplex stainless steel. The reason for this is that cargo tanks ideally have the widest potential range of cargos possible. This range is defined by corrosion resistance. This factor alone is reason to choose duplex over austenitic alloys such as 316L (UNS S31603) or 317L (UNS S31703). An equally decisive factor is strength. With codes permitting the tank’s design to be based on yield strength, the use of duplex alloys—with strengths about double those of austenitic steels—permits significant weight reduction. This is a major economic factor for ship owners in that dead weight can be replaced by fee-paying cargo at the same operating cost. These incremental revenues, over the life of the vessel, are many times the original cost of the material. Based on the high value for strength in ship economics, it would seem that the higheststrength alloys, such as 2507 (UNS S32750),

may be justified based on strength alone; their exceptional corrosion resistance would be simply an excellent side benefit. Corrugated stainless bulkheads are positioned within the carbon steel hull. The stiff corrugated bulkheads are themselves structural strengtheners for the entire ship. The vertical corrugations also facilitate tank cleaning as internal stiffeners are eliminated. Cryogenic containers are still the bastion of austenitic stainless steels. As leaner austenitic alloys have become less expensive than 9% Ni alloy steel, a martensitic grade, they have become the material of choice. In this case, the 201 types are preferred to 304 because 201 has greater strength at the cryogenic operation temperature and is, of course, less expensive. The expanding market for LNG has made ocean transport increasingly important because large disparities in prices often are due to the difficulty in transporting it. The two best materials are UNS S20153 and S20400, which perform equally well. If higher strength is valuable to a design for cryogenic uses, then UNS S21904 (21-6-9 or Nitronic 40) could be used. This alloy has yield strengths of 460 MPa (65 ksi) at room temperature and 1200 MPa (175 ksi) at –196 °C (–320 °F). It is completely resistant to martensite formation. Other shipboard systems benefit equally from the use of duplex stainless steel. This extends to piping, hardware, propellers, shafts, etc.

Heat Exchangers Coolers for captive water systems such as for power plants often need to resist corrosion by brackish water or seawater. To the extent that these are thin-wall tubing, ferritic alloys such as

246 / Stainless Steels for Design Engineers

Seacure (UNS S44660) or 29-4C (UNS S44735) have been used quite successfully. If thicker tubes are required, then the equivalent duplex or austenitic alloys can be used. This would include types 2003 (UNS S32003), 2205 (UNS S32205), or 2507 (UNS S32750) duplex stainless steels, depending on salinity and temperature; for austenitics, the 6Mo alloys such as 254SMO (UNS S31254) and AL6XN (UNS N08367) may be used. The duplex alloys have the advantage of lower cost. Both are resistant to stress corro-

sion cracking to very high temperature and salinity.

REFERENCES

1. Peckner and I. Bernstein, Stainless Steel Handbook, McGraw-Hill, 1966, p 37-1 2. Stainless Steel for Desalination Processes, Feb 2006, Outokumpu, www.outokumpu. com, accessed June 2008

Stainless Steels for Design Engineers Michael F. McGuire, p 247-255 DOI: 10.1361/ssde2008p247

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 22

Petroleum Industry Applications Summary

The petroleum industry has driven large segments of the steel industry since both their be-

ginnings. Demand for steel for drill pipe, casing, and tubing has led to many developments, such as the technology for producing highquality seamless and welded pipe and tubing. Pipeline needs have fueled the market for highstrength, low-alloy plate. Offshore production in often-hostile environments has presented severe material challenges. And, as the light sweet crude that was easily found and produced on land is exhausted, future supplies of hydrocarbons are increasingly likely to contain sulfides, carbon dioxide, and saltwater in sufficient amounts to make corrosion a top priority in selecting materials. For reference in this chapter, Tables 1 through 5 list the relevant alloys for petroleum industry applications. Many, but not all, of these alloys are listed in the National Association of Corrosion Engineers (NACE) MR0175, Sulfide Stress Corrosion Cracking Resistant Metallic Materials for Oil Field Equipment; the tables in this chapter also include some newer alloys not in the NACE document.

Table 1 Ferritic stainless steels for petroleum industry applications

Table 2 Martensitic stainless steels for petroleum industry applications

THE PETROLEUM INDUSTRY has had to deal with increasingly hostile environments in its search for new supplies of oil. And that petroleum, when found, often contains harmful ingredients. The result is increasing demand for steels with greater strength and corrosion resistance. Martensitic and duplex stainless steels have provided the corrosion resistance and strength to deal with higher levels of hydrogen sulfide, carbon dioxide, chlorides, and acidity. This chapter reviews the selection of stainless steels for petroleum applications, including oil country tubular goods (OCTGs), line pipe, offshore platforms, and refinery equipment.

Introduction

UNS

S40500 S40900 S43000 S43035 S43400 S43600 S44200 S44400 S44500 S44600 S44626 S44627 S44635 S44660 S44700 S44735 S44800 S46800

Common name

405 409 430 439 434 436 442 444 (18-2) … 446 26-1 Ti, E-Brite 26-1 Cb 26-4-4, Monit Seacure, SC-1 29-4 29-4C 29-4-2 468

Note: See Appendix 1 for alloy compositions. Source: Adapted from NACE MR0175, “Sulfide Stress Corrosion Cracking Resistant Metallic Materials for Oil Field Equipment”

UNS

J91150 J91151 J91540 K90941 S14125 S41000 S41426 S41427 S42000 S42400 S42500

Common name

CA15 CA15M CA6 NM 9Cr 1Mo S/W 13Cr 410 13CRS … 420 F6NM 15Cr JFE KL-12G JFE KNHP12Cr Nippon NT-CRS Nippon NT-CRSS 420M L80 13 Cr

Hardness, HRC, max(a)

... ... ... ... 28 22 ... 29 22 23 22 ... ... ... ... ... ...

Note: See Appendix 1 for alloy compositions. (a) As specified in NACE MR0175. Source: Adapted from NACE MR0175, “Sulfide Stress Corrosion Cracking Resistant Metallic Materials for Oil Field Equipment”

248 / Stainless Steels for Design Engineers

Table 3 Precipitation hardening stainless steels for petroleum industry applications UNS

Common name

S13800 S15500 S15700 S17400 S17700 S35000 S35500 S45000 S45500 S46500 S66286

13-8 PH 15-5 PH 15-7 PH 17-4 PH 17-7 PH AM-350 AM-355 Custom 450 Custom 455 Custom 465 A-286 Custom 465 (275) Custom 475

Hardness, HRC, max(a)

... 33 32 33 ... ... ... 31 ... ... 35 ... ...

Note: See Appendix 1 for alloy compositions. (a) As specified in NACE MR0175. Source: Adapted from NACE MR0175, “Sulfide Stress Corrosion Cracking Resistant Metallic Materials for Oil Field Equipment”

The modern dilemma that makes stainless necessary is the addition presence of wet carbon dioxide, which is extremely corrosive to carbon and alloy steel. As if this is not a sufficient material problem, sometimes the wetness is from saltwater, which further aggravates corrosivity. This corrosion problem is compounded by the accelerating influence of high temperature in deeper formations. What is the answer to the corrosion problem? Inhibitors, coatings, cathodic protection, or more corrosion-resistant materials are the main responses. The first three responses are not always practical. They also represent an ongoing cost rather than a one-time cost. Each situation must be evaluated regarding which is the optimal solution.

Combating Corrosion in Alloys for Petroleum Applications Alloying steel with chromium, copper, molybdenum, and nickel can lower the corrosion rate of steel by a factor of 10,000. Figure 1 shows the influence of chromium alone, which produces a 100-fold reduction in corrosion of steel in seawater and carbon dioxide (Ref 1). Molybdenum is the most powerful alloying addition to magnify the benefit of chromium. The effects of copper and nickel are also very significant, as Fig. 2 (Ref 2) indicates. These additions must be made in a very balanced way if a tough, fully martensitic structure is to be maintained. Carbon must be kept low to avoid the formation of chromium carbides during tempering, which would counteract the benefit of the chromium. Nickel is necessary to prevent σ-ferrite formation, which reduces toughness.

Table 4 Duplex stainless steels for petroleum industry applications UNS

J93345 J93370 J93380 J93404 S31200 S31260 S31500 S31803 S32001 S32003 S32101 S32205 S32304 S32404 S32520 S32550 S32750 S32760 S32803 S32900 S32906 S32950 S32977 S39274

Common name

PREN(a)

Escoloy CD4MCu Z100 958 44LN DP3 3RE60 2205 (old) 19D 2003 2101 2205 (new) 2304 U50 52N+ 255 2507 Zeron 100 2803Mo 329 2906 7-Mo Plus AF 918 DP3W

31-47 30-34 38-46 39-47 30-36 34-43 27-31 30-36 20-24 27-31 25-29 34-38 23-27 27-32 37-48 32-44 38-44 40-46 33-41 26-35 36-45 32-43 39-46 39-47

Note: See Appendix 1 for alloy compositions. (a) PREN, pitting resistance equivalent number. Source: Adapted from NACE MR0175, “Sulfide Stress Corrosion Cracking Resistant Metallic Materials for Oil Field Equipment”

The chapters on martensitic and precipitation hardening stainless steels discuss this in detail. The martensitic stainless steels used for these applications are resistant to carbon dioxideenhanced corrosion up to partial pressures of 100 atm, after which further alloying is necessary. This cannot be achieved with a martensitic structure, but the duplex alloys have the corrosion resistance and strength to work in this regime. They have high annealed strength and can also be cold worked to higher strength levels. If hydrogen sulfide is present, the selection process can become more difficult. Highstrength martensitic steels are susceptible to brittle delayed failure in the presence of hydrogen sulfide. Being stainless does not by itself provide immunity. If localized corrosion occurs, hydrogen uptake ensues, and delayed failure follows. Only keeping hardness below wellestablished levels can render a martensitic alloy immune. If the localized corrosion can be prevented, however, then the stress corrosion cracking (SCC) cannot be initiated. Molybdenum alloying expands the pH and chloride range from which an alloy can be free of the pitting corrosion that initiates SCC, as shown in Fig. 3 (Ref 2). Martensitic steels of all types have a maximum in susceptibility to SCC via hydrogen embrittlement near room temperature. Duplex alloys and austenitic alloys become susceptible at higher temperatures and do not

Chapter 22: Petroleum Industry Applications / 249

Table 5

Austenitic stainless steels for petroleum industry applications

UNS

Common name

J92500 J92600 J92602 J92701 J92710 J92800

CF-3 CF-8 CF-20 CF-16F CF-8C CF-3M CF-12M CG-8M CK3MCuN CH-20 CN-3MN AL 22 CN-7M 20Cb-3 AL 20 20Mo-4 20Mo-6 Sanicro 28 Nicrofer 3127 hMo 20Mod AL-6X AL-6XN JS-700 332 25-6Mo Cronifer 1925 hMo URSB-8 904L 201 201LN 202 Nitronic 30 (204L) 204 205 Nitronic 50 Nitronic 60 Nitronic 40(219) 21-6-9 LC Nitronic 33

J93000 J93254 J93402 J94652 N06022 N08007 N08020 N08020 N08024 N08026 N08028 N08031 N08320 N08366 N08367 N08700 N08800 N08925 N08926 N08932 N80904 S20100 S20153 S20200 S20400 S20430 S20500 S20910 S21800 S21900 S21904 S24000

PREN(a)

... ... ... ... ... ... ... ... ... ... ... ... ... 28 28 38 ... 39 54 38 ... 49 36 ... 46 47 49 39 ... ... ... ... ... ... ... ... ... ... ...

UNS

S30100 S30153 S30200 S30215 S30300 S30400 S30403 S30409 S30415 S30453 S30500 S30800 S30815 S30900 S31000 S31008 S31254 S31266 S31600 S31603 S31609 S31635 S31700 S31703 S31725 S31726 S31753 S32100 S32109 S32200 S32654 S33000 S33400 S34565 S34700 S34709 S35125 S35315

Common name

301 301LN 302 302B 303 304 304L 304H 153MA 304LN 305 308 253MA 309 310 310S 254SMO B66 316 316L 316H 316Ti 317 317L 317LM 317LMN 317LN 321 321H NIC 25 654SMO 330 334 4565 347 347H 332Mo 353MA Cronifer 2328

PREN(a)

... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 46 59 ... 25 ... ... ... ... ... ... ... ... ... ... 64 ... ... 54 ... ... ... ... ...

Note: See Appendix 1 for alloy compositions. (a) PREN, pitting resistance equivalent number. Source: Adapted from NACE MR0175, “Sulfide Stress Corrosion Cracking Resistant Metallic Materials for Oil Field Equipment”

Fig. 1

Influence of chromium on the corrosion rate of steel in environments experienced by oil country tubular goods. Test conditions: synthetic sea water; CO2 partial pressure, 0.1 MPa; test temperature, 60 (C °140 °F); test duration, 150 h; flow velocity, 2.5 m/s; specific volume, 800 mL/cm2. SSC, stress corrosion cracking. Source: Ref 1

Fig. 2

Influence of copper and nickel on the corrosion rate of martensitic stainless alloys used for oil country tubular goods. Source: Ref 2

250 / Stainless Steels for Design Engineers

The NACE recommendations of suitable materials are defined by MR0175. Table 6 summarizes these recommendations. The reader is encouraged to refer to the latest version of this document for further details.

Oil Country Tubular Goods

Fig. 3

Influence of molybdenum on susceptibility to stress corrosion cracking in solutions containing (a) 3.5% NaCl and (b) 0% NaCl. Source: Ref 2

exhibit the same increasing susceptibility with strength. So, when hydrogen sulfide, which enhances hydrogen uptake, levels exceed about 10–2 atm, the martensitic alloys should no longer be used, and the duplex alloys are preferred. As temperatures and hydrogen sulfide partial pressures increase, alloying must also, until at 1 atm of hydrogen sulfide nickel base alloys are required. Figure 4 shows this progression with the alloy recommendations of Sumitomo. The requirements behind this diagram are generic. Any producer’s alloys must comply with this diagram’s regions, which have been defined by NACE. Stainless steels are required above a certain carbon dioxide level for all levels of hydrogen sulfide. Martensitic alloys, commonly called “13Cr,” are the first step up from alloy steels. At higher levels of carbon dioxide and hydrogen sulfide, duplex alloys are required, with the 22CR alloys such as UNS S32205 used at temperatures up to 200 °C (390 °F) and the 25CR alloys such as UNS S32507 at temperatures up to 250 °C (480 °F). Nickel-base alloys are required at hydrogen sulfide levels above 1 atm partial pressure.

Oil country tubular goods (OCTG) include the drill pipe, casing, and tubing and associated hardware used to construct oil and gas wells. Drill pipe is used to twist the drill bit and convey drilling fluids to the point of contact and flush away debris. Casing is put in place to stabilize the well walls, while tubing is placed within the casing to carry oil and gas to the surface. Each of these components sees significant stresses, and high strength-to-weight materials are needed. Drill pipe is in tension, torsion, and compression alternately throughout its life. Casing hangs from the wellhead under its own weight for distances from hundreds of meters to 7000 or 8000 m and must withstand very high collapse as well as burst pressures. Well strings, the exact sequence of size and strength pipe for each level of the well, are optimized for the conditions of each well. The variety of strengths and sizes are standardized by the American Petroleum Institute. The use of the highest strengths has always been limited by hydrogen embrittlement accelerated by hydrogen sulfide, so that the maximum hardness for a given material must be strictly adhered to when hydrogen sulfide is present. The terms 13Cr, 22Cr, and 25Cr are commonly used in the industry even though this greatly oversimplifies the alloying, and therefore performance, options that exist. The 13Cr alloys are a family of martensitic stainless steels. The 22Cr and 25Cr alloys are duplex grades. The former are used in the quenched and tempered condition, while the duplex alloys are used as annealed or cold worked. The 13Cr grades began as simply variations on 420, which is a straight-chromium martensitic often used for cutlery. This alloy, while far better (about 100 times) than alloy steel in corrosion resistance, has nearly the least corrosion resistance of all stainless steels. To achieve higher corrosion resistance molybdenum is added. Molybdenum at 1% increases resistance to general corrosion in a sodium chloride/ hydrogen sulfide/carbon dioxide environment by about tenfold. Another 1% increases it another

Chapter 22: Petroleum Industry Applications / 251

Fig. 4

Alloy suitability as a function of H2S and CO2 partial pressure. Source: Ref 1

tenfold. The 2% level of molybdenum also greatly reduces pitting, which in turn eliminates the initiation point of SCC. Simply adding molybdenum would cause the alloy to have excessive δ-ferrite, which cannot transform to martensite and would therefore reduce mechanical properties. Thus, nickel must be added to counter the ferrite stabilizing effect, unfortunately, but necessarily increasing the cost. The nickel does help lower the general corrosion rate. Carbon and nitrogen in these alloys are kept at low concentrations. These alloys are otherwise almost identical to precipitation hardening martensitic stainless steels without the precipitating phase.

Martensitic alloys are susceptible to SCC by a hydrogen embrittlement mechanism. This susceptibility is strongly temperature dependent. It decreases with temperature from a maximum at ambient to none at around 100 °C (210 °F). If the hydrogen sulfide level exceeds 0.03 atm, then 22Cr alloys should be used rather than 13Cr because of this risk. Hydrogen sulfide may be contained in the petroleum, or it may come from sulfate-reducing bacteria, introduced by flooding, for example. This can cause a sulfide-free system to become sulfide rich after the fact and make initial materials choice wrong after the fact. The 22Cr and 25Cr alloys have significantly higher resistance to chlorides and wet hydrogen

252 / Stainless Steels for Design Engineers

Table 6 Restrictions in use recommended by NACE MR0175 for selected stainless steels used for petroleum industry applications UNS

J91150 J91151 J91540 J93254 J95370 N08926 S15500 S15700 S17400 S20910 S41000 S41425 S41426 S41427 S41429 S41500 S42000 S42400 S42500 S45000 S66286

Common name

CA15 CA15M CA6NM ... ... ... 15-5 15-7 17-4 ... 410 ... ... ... ... F6NM 420 ... ... 450 A286 Austenitic A-2 Duplex Superaustenitic, type 3a Superaustenitic, type 3b

PREN(a)

Temperature, ºC (ºF)

pH, min

H2S, kPa

Hardness, HRC

Cl–, mg/L

... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 40 30

... ... ... ... 150 (300) 121 (250) ... ... ... 66 (150) ... ... ... ... ... ... ... ... ... ... 65 (150) ... 232 (450) 232 (450) ...

3.5 3.5 3.5 ... ... 3.5 ... ... ... ... 3.5 ... 3.5 3.5 4.5 3.5 3.5 3.5 3.5 ... ... ... ... ... ...

10 10 10 ... 700 700 3.4 ... 3.4 100 10 10 10 10 10 10 10 10 10 1 100 ... 10 20 ...

22 22 23 100 HRB 94 HRB ... 33 32 33 ... 22 28 27 29 27 23 22 23 22 31 35 22 ... ... ...

... ... ... ... 90,000 60,700 ... ... ... 35 ... ... ... 6,000 ... ... ... ... ... ... ... ... ... ... ...

121 (250) 149 (300) 171 (340)

... ... ...

700 310 100

... ... ...

5,000 5,000 5,000

>40

Notes: See NACE MR0175 for further use and processing restrictions. See Appendix 1 for alloy compositions. (a) PREN, pitting resistance equivalent number. Source: Adapted from NACE MR0175, “Sulfide Stress Corrosion Cracking Resistant Metallic Materials for Oil Field Equipment”

sulfide and can resist SCC at ten times higher concentrations than the 13Cr alloys. Besides the inherently greater corrosion resistance that derives from the chromium, molybdenum, and nitrogen levels of the 22Cr and 25 Cr alloys (see the chapters on corrosion in this book), the duplex alloys have very fine grain size and a roughly 50/50 mixture of ferrite and austenite. This acts as a crack arrestor should one phase be susceptible to cracking while the other is not. There have been no reported downhole failures of annealed or cold-worked duplex alloys. There was one instance of very high-strength tubing cracking after cathodic contact with carbon steel casing. This was after removal from the well and after handling damage had occurred. The affected microstructure was found to be high (70%) in ferrite (Ref 3). For corrosion resistance above that furnished by superduplex materials such as the 25Cr alloys, super austenitic alloys fill a gap before nickel base alloys are needed. These alloys achieve a tenfold increase in hydrogen sulfide resistance and very elevated SCC resistance. These are the so-called 6Mo grades. The more advanced of them contain high levels of nitro-

gen. The more common alloys are UNS S32654 and N08367. The recent development of lean duplex alloys has not yet made its way into OCTGs. These alloys offer an inherent alloy savings over the 13Cr grades in nickel and molybdenum content while offering better corrosion and SCC resistance. Their strength levels in the annealed condition, 450 MPa (65 ksi), are lower than those of the martensitic alloys, 600 MPa (87 ksi), so for most downhole applications they will require cold working. It is likely, however, that these alloys will see their first service as line pipe, where they will not need to be cold worked to higher strength levels to be widely used.

Line Pipe and Flow Lines With the awesome cost of corrosion, the case for stainless line pipe is easily made. Whether to use stainless depends on vulnerability of carbon steel. This evaluation is made based on the carbon dioxide, hydrogen sulfide, water, salinity, temperatures, pressures, flow conditions, and so forth. The normal basis for

Chapter 22: Petroleum Industry Applications / 253

these calculations follows that published by C. de Waard of Shell (Ref 3). The competing technology when corrosion dangers arise is the use of corrosion inhibitors, cathodic protection, or to line carbon steel with a protective coating. The use of inhibitors is subject to the risk of velocity limitations, temperature limitations, and simply of the inhibitor working appropriately, not to mention cost. Cathodic protection is costly and complex. Coatings can be damaged by numerous occurrences in acidity, mechanical damage, or fluctuations in temperature or pressure. Use of a stainless corrosion-resistant alloy can have well-defined and controlled costs and performance over the life of an installation. Line pipe differs from downhole in having strength requirements more in line with that of annealed duplex alloys. These requirements gave birth to modern duplex alloys, starting with UNS S31803 and evolving to UNS S32205 as the value of higher nitrogen became understood. Nitrogen not only enhances corrosion resistance, but also suppresses the formation of undesirable and embrittling intermetallic phases that might otherwise form at welding temperatures. It also keeps the desirable austenite/ferrite ratios in weld metal. Since the development of the first widely accepted duplex alloys, more alloys have emerged. Superduplexes, such as UNS S32750, have become accepted alloys. Then, the need to improve costs led in the 1990s to the use of martensitic alloys with high levels of nickel and molybdenum, which at the time were lower cost. The emergence in the early 2000s of lean duplex alloys provided strength and more corrosion resistance with lower nickel levels, giving them a cost advantage during periods of high nickel cost. The main attribute required by line pipe that is not as important in OCTGs is weldability. This is not an overwhelming challenge for duplex alloys, but for martensitic alloys, it requires a very low interstitial level so that the martensite is self-tempering and ductile in the as-welded condition. This can be achieved by stabilizing the alloy with small amounts of titanium. It would appear that under current conditions that alloy 2101 (UNS 32101) has a cost/performance edge over the martensitic competition and should for the long term. The main ingredients required in a duplex for strength and corrosion resistance are chromium

and nitrogen, both relatively inexpensive alloying elements. Alloys S32001 and S32101 are well formulated for medium and high levels of corrosion resistance required for wet carbon dioxide, hydrogen sulfide, and trace chlorides. The main precaution for duplex alloys is maintaining a nominally 50/50 mixture of ferrite and austenite with no embrittling intermetallic phases. The modern alloys have high (greater than 0.14%) nitrogen, which helps to preserve austenite levels after welding and suppress intermetallic formation. Nevertheless, minimization of time above 350 °C (660 °F) is important. This tendency increases with chromium and molybdenum content, which is another reason why the lean duplex alloys are so attractive. For subsea use, 22Cr duplex generally requires cathodic protection because of the risk of crevice corrosion. 25Cr duplexes are used without cathodic protection. Duplex pipelines have been in service in the North Sea since the 1970s.

Umbilical Tubing and Risers Increasingly, wells are located undersea. It is standard practice to control and monitor these wells via bundled umbilical tubing. The tubing can provide hydraulic and electrical power, control and adjust pressure, carry communications, and even introduce chemical to the well. The depth of wellheads can increase collapse pressures to levels beyond the capability of thermoplastics, which has led to the use of duplex stainless steel because of its strength and resistance to corrosion and SCC. When resistance to seawater is the main concern, the rule of thumb is that a pitting resistance equivalent number (PREN) of 35 or greater is required, whereas resistance to crevice corrosion requires a PREN of at least 40. This has made the superduplex UNS S32750 the standard. Such a critical item as an umbilical may seem like a poor application on which to economize, but again the lean duplexes offer possibilities to do so. By zinc coating lean duplexes such as alloy 19D (UNS S32001) and 2101 (UNS S32101), very long service lives can be safely extrapolated. These alloys are being promoted on their lower susceptibility to σ formation during welding, and if welding thermal cycles cannot be controlled that may be an issue, but superduplex seems to have become a pervasive choice because it is superbly reliable.

254 / Stainless Steels for Design Engineers

Risers are now produced in coiled tubing of over 100 mm (4 in.) diameter, so that very economical long lengths are feasible.

Platforms Platforms present a special case in which the costs of maintenance are high, the corrosion environment is severe, and the penalty for excess weight is also high. A savings of 1 ton in weight topside can save over $100,000 in steel in the subsea jacket. This leads to a rapid payback for the use of materials that are sufficiently resistant to corrosion such that corrosion loss allowance can be eliminated. Both titanium and stainless alloy UNS S32750 are equal candidates for this service, depending on availability and current alloy prices. Except in rare cases, stainless steel wins the cost battle between these alloy systems. Almost any structure is a candidate for stainless topside processing: piping, pumps, flanges, fittings, etc. Hardware of any type and construction materials benefit from being stainless. Seawater systems often employ 22Cr duplex with cathodic protection or unprotected 25Cr duplex. A wise preventive action is to paint stainless that is covered by insulation or similar material, which otherwise can result in concentration cells and consequent pitting. Table 7

Liquefied Natural Gas Vessels Liquefied natural gas (LNG) is becoming an increasingly important commodity as the value of stranded gas makes it economically desirable to convert it to a transportable state. Converting natural gas to a cryogenic liquid presents a material problem. Vessels to contain it must have strength and toughness at temperatures below –150 °C (–240 °F). The traditional material, 9% Ni martensitic steel, has become expensive compared to the lower-nickel austenitic stainless steels, such as 201LN (UNS S20153), which have no transition temperature and strengthen with decreasing temperature. Alloy 201LN is cheaper, easier to weld and fabricate, and of course is stainless, which 9% Ni steel is not. The extreme ductility of 201LN compared to martensitic steel gives it a decided advantage in terms of rupture resistance, which is a major design and political concern with this potentially explosive commodity.

Refinery Equipment Corrosion resistance is a major factor in the choice of materials in refinery operations. As we discussed, crude oil itself is sometimes a very corrosive fluid, but in refining the by-products,

Stainless steels used in various refinery processes

Process

Crude distillation

Corrosive agents

Applications

Alloys

Notes

Preheaters, distillation tower Towers

405, 409, 410

…

Vacuum fractionalization

Sulfur-containing acids (SCAs) SCA, chlorides

405, 410, 316

Coker

SCA, H2S

Condensers Coke drums

S44735, 2205 409

Gas plants

H2S, water, Cl–, ammonia

AL-6XN, 2205, 2507

Amine plant

Ammonia, MEA, DEA

410S, 316L 304L, 316L

… …

Sulfuric acid alkylation

Sulfuric acid Dilute sulfuric acid H2S, ammonia, PTA(a)

Compressor coolers, reboiler tubes Trays Reboilers, trays, filters, condenser tubing Contactor, mixer Effluent piping Hot sections

Depending on crude corrosivity Depending on chloride level Depending on crude corrosivity …

20Cb3 316L 321, 347

General Reactor internals

410S, 304 304

Low pH excursions possible … Long exposure at high temperature … HCl catalyst regeneration

Heat exchangers Trays Cyclones, vapor lines Tubing, heat exchangers Heat exchangers

2205 410 304 304 409, 321, 347, 2205, 6Mo 304, 20Cb3, 2205

Hydrotreating

Catalytic re-forming

Fluid catalytic cracking

High-temperature strength needed HCl residue High temperature

Hydrogen plant Hydrocracking

Sulfides, chlorides

Sour water stripping

Sulfuric acid, ammonium bisulfide, chlorides

(a) PTA, polythionic acid. Source: Ref 4

Stripper

… Condensers may need 6Mo … … Depending on temperature, risk of chlorides Severity depends on presence of sulfuric acid

Chapter 22: Petroleum Industry Applications / 255

chemicals used in refining and the temperatures used may further aggravate that corrosivity. The aggressive chemical agents that refinery materials must withstand include wet hydrogen sulfide and carbon dioxide, napthenic acids, polythionic acids, chlorides, sulfuric acid, and alkalines as well as simple oxidation. Sometimes, temperatures of use are such that embrittling or sensitizing phase transformations may occur. Table 7 lists some major refinery processes and the materials used in them (Ref 4). Most of these situations are discussed elsewhere in this book in detail. One that is quite specific to refinery applications is polythionic acid (PTA) attack. These acids usually form accidentally when sulfide corrosion products react with moisture and air. The attack is intergranular, and materials respond to it much as they do to the Strauss test. The remedies are to prevent the inadvertent formation of PTA and to avoid

using austenitics, which are prone to grain boundary chromium depletion by sensitization, and instead use low-carbon grades and stabilized grades.

REFERENCES

1. Sumitomo Products for the Oil and Gas Industries, www.sumitomometals.co.jp, accessed June 2008 2. H. Asahi et al., “Development of High Chromium Stainless Line Pipe,” Nippon Steel Technical Report 72, January 1997 3. C. de Waard and U. Lotz, “Prediction of CO2 Corrosion of Carbon Steel,” Paper 69, presented at Corrosion/93, National Association of Corrosion Engineers, 1993 4. C.P. Dillon, Corrosion Resistance of Stainless Steels, Marcel Dekker, 1995

Stainless Steels for Design Engineers Michael F. McGuire, p 257-263 DOI: 10.1361/ssde2008p257

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 23

Chemical and Process Industry Applications Summary

Single- and Dual-Environment Systems

ENGINEERS IN THE PROCESS industries must have materials that can contain a huge variety of chemical species at many temperatures, pressures, and flow rates. This is applied corrosion engineering combined with physics and structural design. It is obvious that this task depends on the availability of corrosion data, more than can be presented here. This chapter covers what data are necessary and how they can be found.

Under ideal conditions, a material may need to resist one single major corrosion threat. If the most potentially damaging species can be clearly identified, then candidate materials can be found by searching published data. These data are available freely online from Web sites (such as Ref 1 and 2) or for a charge from sources such as the National Association of Corrosion Engineers (NACE; Ref 3) and ASM International (Ref 4). It is difficult for any published data to keep up with the latest developments. The testing alone of new materials can take a long time, and then it must wait for publication. All materials are not covered, especially when a manufacturer publishes data on proprietary alloys and excludes competitive materials. That having been said, any improvements over standard alloys will first be reported by the developers of the alloy, and they will logically tout its strongest points. For this reason, dialogue with the primary steel producers is encouraged. No one has more exposure to the latest trends in applications. A single-environment system is typically one in which the aggressive chemical species is the only consideration. This is normally the case for piping, tanks, or reaction vessels holding the species or materials immersed in the aggressive species. A dual-environment system is typically encountered in heat exchangers, but it must also be extended to single-environment systems in which the exposure of the nonreactant side of the material to the ambient environment cannot be neglected, as in the case of marine ambient environments.

Introduction The need to work with hostile chemicals begins with the manufacture of those chemicals. It was in the production of nitric acid that stainless had its first industrial application. These are industries with purely need-driven material challenges. New processes are constantly in development, and they present new environments in which materials must perform. The choices are highly pragmatic. In an industrial environment, the costs of a poorly performing material can be well known by its effect on downtime, maintenance, liability, etc. The essential knowledge is which materials will work. The selection of materials for the chemical and power industries is first a study of corrosion resistance, including resistance to stress corrosion cracking (SCC). Strength plays a secondary role but can be an important cost factor. These considerations may occur at very high or very low temperatures, in which case corrosion resistance may become oxidation resistance and strength may mean creep strength.

258 / Stainless Steels for Design Engineers

The challenges that must be met are primarily ensuring adequate corrosion resistance and secondarily having acceptable mechanical properties. The corrosion issues run the full gamut of potential forms of corrosion: • General corrosion • Pitting corrosion and crevice corrosion • Intergranular corrosion • Stress corrosion cracking • Erosion corrosion In addition to these forms of corrosion associated with liquids, there are considerations of gas phase attack, which may be oxidation, sulfidation, or attack by other gases. Mechanical design considerations are normally limited to static stress allowances. Previously, handbooks dealt very lightly with this topic because all the normally recommended steels had similar strength. The proliferation of duplex stainless steels has changed that. Now, high-strength alloys of high corrosion resistance and SCC resistance are available and are making traditionally chosen stainless steels less than optimal.

Corrosion Types A designer wants to deal with general corrosion. Its rate can be predicted, and thickness can be chosen to allow for its occurrence. Corrosion data for general corrosion are normally presented in isocorrosion charts. These present the temperatures and concentrations for a given environment at which various materials will exhibit the same corrosion rate. This rate is most often 0.1 mm/yr, an amount that can be thought of as a tolerable level for many uses. Figure 1 shows an isocorrosion chart for stainless steels in sulfuric acid (Ref 1). The data are clear when presented in this fashion. It can further be appreciated that in general reducing alloy performance to a mathematical formula, such as the pitting resistance equivalent number (PREN) equation, would not be reasonable since the relative performance of alloys changes considerably with concentration. Thus, the design engineer must rely on experimentally developed data. Since these data are available both online and in print, no attempt will be made to reproduce them fully here. Examples are given in Tables 1 and 2 (Ref 1). Such tables are very useful, although the presentation is not visually compact. The inclusion

Fig. 1

Isocorrosion chart for sulfuric acid. Source: Ref 1

of carbon steel and titanium gives a valuable frame of reference for the engineer. If the forms of localized corrosion discussed next can be avoided, the corrosion tables are sufficient to guide the designer to a reasonable selection of candidate materials for any process in which the chemical species involved have been identified. If the data have not been developed for a certain environment, then the tables give a first approximation of which materials may be resistant from examination of similar environments, and a final decision can only be reasonably made though direct corrosion testing of candidate materials. Refer to the chapters on corrosion for a more thorough discussion of uniform corrosion. Pitting and Crevice Corrosion Stainless steel is unique among metals and alloys in that it derives its corrosion resistance from constituent alloying elements working together to form a thin passive layer that, when intact, is highly resistant to corrosion. The strength of the passive layer in resisting attack by halide ions, which are the most disruptive ions to the layer, is proportional principally to the chromium, nitrogen, and molybdenum contents of the alloy. This relationship follows the formula: PREN = %Cr + 3.3%Mo + 30%N

(Eq 1)

This formula is one of the commonly used versions, none of which is universally correct. Both tungsten and carbon can increase pitting resistance, while sulfur diminishes it. This is discussed in the corrosion section of this book.

Chapter 23: Chemical and Process Industry Applications / 259

The important consideration is that this formula assumes that the key alloying elements are homogeneously distributed in solution. This will only be true if correct thermomechanical processing occurs because, thermodynamically, these alloys are not used in an equilibrium condition. Were they to attain equilibrium, say by overheating, alloy segregation by precipitation could occur, causing localized loss of corrosion Table 1

resistance, which is what causes pitting. Chromium is very reactive: Its affinity for oxygen makes the passive film strong. Pitting has nearly always been associated with manganese sulfide inclusions, and although there is still debate over the precise mechanism, it appears that chromium depletion at the metal-inclusion interface is to blame. Eliminating inclusions by eliminating either manganese or sulfur improves

Corrosion table for sulfuric acid (H2SO4)

Concentration, % Temperature, °C

Carbon steel 13% Cr steel 18-2 (UNS S44400) 3R12 (UNS S30400) 3R60 (UNS S31600) 18-13-3 17-14-4 2RK65 (UNS N08904) Sanicro 28 (UNS N08028) 254SMO (UNS S31254) 654 SMO (UNS S32654) SAF 2304 (UNS S32304) SAF 2205 (UNS S31803) SAF 2507 (UNS S32750) Titanium

0.1 100 = BP

0.5 20

0.5 50

0.5 100 = BP

1 20

1 50

1 70

1 85

1 100 = BP

2 20

2 50

2 60

3 20

3 35

3 50

2 2 2 2 1 1 1 0

2 2 0 0 0 0 0 0

2 2 2 1 0 0 0 0

2 2 2 2 1 1 1 1

2 2 0 0 0 0 0 0

2 2 2 1 0 0 0 0

2 2 2 1 0 0 0 0

2 2 2 2 1 1 0 0

2 2 2 2 1 1 1 1

2 2 0 0 0 0 0 0

2 2 2 1 0 0 0 0

2 2 2 1 0 0 0 0

2 2 0 0 0 0 0 0

2 2 2 1 0 0 0 0

2 2 2 1 0 0 0 0

...

0

0

0

0

0

0

0

0

0

0

0

0

0

0

...

0

0

...

0

0

0

0

1

0

0

0

0

0

0

...

0

0

...

0

0

0

0

0

0

0

0

0

0

0

1

0

0

...

0

0

0

0

1

0

0

0

0

0

0

...

0

0

1

0

0

0

0

...

0

0

0

0

0

0

...

0

0

...

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

Concentration, % Temperature, °C

3 85

1

3 100 = BP

5 20

5 35

1 5 60

0 5 75

0 5 85

1

5 101 = BP

1 10 20

1 10 50

0 10 60

0 10 80

1

10 102 = BP

0

20 20

20 40

Carbon steel 13% Cr steel 18-2 (UNS S44400) 3R12 (UNS S30400) 3R60 (UNS S31600) 18-13-3 17-14-4 2RK65 (UNS N08904) Sanicro 28 (UNS N08028) 254SMO (UNS S31254) 654 SMO (UNS S32654) SAF 2304 (UNS S32304) SAF 2205 (UNS S31803) SAF 2507 (UNS S32750) Titanium

2 2 2 2 1 1 1 0

2 2 2 2 2 2 2 1

2 2 2 1 0 0 0 0

2 2 2 1 0 0 0 0

2 2 2 2 1 0 0 0

2 2 2 2 1 1 1 0

2 2 2 2 2 2 2 1

2 2 2 2 2 2 2 2

2 2 2 2 0 0 0 0

2 2 2 2 1 1 0 0

2 2 2 2 1 1 1 0

2 2 2 2 2 2 2 1

2 2 2 2 2 2 2 2

2 2 2 2 0 0 0 0

2 2 2 2 1 1 1 0

...

1

0

0

0

0

0

2

0

0

0

0

2

0

0

...

1

0

0

0

0

1

2

0

0

0

0

2

0

0

0

0

0

0

0

0

0

2

...

...

0

0

...

0

0

...

1

0

0

0

0

0

2

0

0

0

2

2

1

2

...

1

0

0

0

0

0

2

0

0

0

1

2

0

0

...

1

0

0

0

0

...

...

0

0

0

0

2

0

0

1

2

0

1

1

2

2

2

1

2

2

2

2

2

2

(continued) Notes: 0, corrosion rate of less than 0.1 mm/yr. The material is corrosion proof. 1, corrosion rate of 0.1–1.0 mm/yr. The material is not corrosion proof but useful in certain cases. 2, corrosion rate of more than 1.0 mm/yr. Serious corrosion. The material is not usable. BP, boiling solution. Source: Adapted from Ref 1

260 / Stainless Steels for Design Engineers

Table 1

(continued)

Concentration, % Temperature, °C

20 50

20 60

20 80

20 100

30 20

30 40

30 60

30 80

40 20

40 40

40 60

40 90

50 20

50 40

50 70

Carbon steel 13% Cr steel 18-2 (UNS S44400) 3R12 (UNS S30400) 3R60 (UNS S31600) 18-13-3 17-14-4 2RK65 (UNS N08904) Sanicro 28 (UNS N08028) 254SMO (UNS S31254) 654 SMO (UNS S32654) SAF 2304 (UNS S32304) SAF 2205 (UNS S31803) SAF 2507 (UNS S32750) Titanium

2 2 2 2 1 1 1 0

2 2 2 2 2 1 1 0

... ... ... ... ... ... ... 1

2 2 2 2 2 2 2 2

2 2 2 2 1 1 1 0

2 2 2 2 2 1 1 0

2 2 2 2 2 2 2 1

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

2 2 2 2 2 2 2 0

2 2 2 2 2 2 2 0

2 2 2 2 2 2 2 1

2 2 2 2 2 2 2 2

2 2 2 2 2 2 2 0

2 2 2 2 2 2 2 0

2 2 2 2 2 2 2 2

0

0

...

2

0

0

1

...

0

0

1

2

0

0

1

0

0

...

2

0

0

1

2

...

1

...

...

0

1

...

0

0

0

2

...

...

...

...

0

0

0

...

0

0

...

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

0

1

2

2

0

1

2

2

2

2

2

2

2

2

2

0

0

1

2

...

0

1

2

0

1

2

2

1

1

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

Concentration, % Temperature, °C

60 20

60 40

60 70

70 20

70 40

70 70

80 20

80 40

80 60

85 20

85 30

85 40

85 50

90 20

90 30

Carbon steel 13% Cr steel 18-2 (UNS S44400) 3R12 (UNS S30400) 3R60 (UNS S31600) 18-13-3 17-14-4 2RK65 (UNS N08904) Sanicro 28 (UNS N08028) 254SMO (UNS S31254) 654 SMO (UNS S32654) SAF 2304 (UNS S32304) SAF 2205 (UNS S31803) SAF 2507 (UNS S32750) Titanium

2 2 2 2 2 2 2 0

2 2 2 2 2 2 2 1

2 2 2 2 2 2 2 1

2 2 2 2 2 2 2 0

2 2 2 2 2 2 2 1

2 2 2 2 2 2 2 1

2 2 2 2 1 1 1 0

2 2 2 2 2 2 2 1

2 2 2 2 2 2 2 2

0 1 1 1 1 1 1 0

1 1 1 1 1 1 1 0

2 2 1 1 1 1 1 1

2 2 2 2 2 2 2 1

0 0 0 0 0 0 0 0

1 1 1 0 0 1 1 0

0

0

1

0

0

1

...

1

1

0

0

0

0

0

0

0

1

...

0

1

...

0

1

2

0

...

...

...

1

...

0

1

...

0

1

...

...

...

...

...

...

...

...

1

...

2

...

...

...

...

...

...

...

...

1

1

...

...

...

1

2

2

2

1

...

...

2

2

2

1

...

...

...

1

1

...

...

...

...

2

2

...

2

2

1

1

...

...

0

0

2

2

2

2

2

2

2

Concentration, % Temperature, °C

90 40

90 70

94 20

94 30

94 40

2 94 50

2 96 20

2 96 30

2 96 40

2 96 50

2 98 30

2

98 40

2

98 50

98 80

Carbon steel 13% Cr steel 18-2 (UNS S44400) 3R12 (UNS S30400) 3R60 (UNS S31600) 18-13-3 17-14-4 2RK65 (UNS N08904) Sanicro 28 (UNS N08028)

2 2 2 2 1 1 1 1

2 2 2 2 2 2 2 2

0 0 0 0 0 0 0 0

2 1 0 0 0 0 0 0

2 2 2 1 0 1 1 1

2 2 2 1 1 1 1 1

0 0 0 0 0 0 0 0

1 1 0 0 0 0 0 0

2 2 1 0 0 1 1 1

2 2 2 1 1 1 1 1

1 1 0 0 0 0 0 0

1 1 1 0 0 0 0 1

2 2 2 2 0 1 1 1

2 2 2 2 2 2 2 2

0

1

0

0

0

0

0

0

0

1

0

0

0

1

(continued) Notes: 0, corrosion rate of less than 0.1 mm/yr. The material is corrosion proof. 1, corrosion rate of 0.1–1.0 mm/yr. The material is not corrosion proof but useful in certain cases. 2, corrosion rate of more than 1.0 mm/yr. Serious corrosion. The material is not usable. BP, boiling solution. Source: Adapted from Ref 1

Chapter 23: Chemical and Process Industry Applications / 261

Table 1

(continued)

Concentration, % Temperature, °C

90 40

90 70

94 20

94 30

94 40

94 50

96 20

96 30

96 40

96 50

98 30

98 40

98 50

98 80

254SMO (UNS S31254) 654 SMO (UNS S32654) SAF 2304 (UNS S32304) SAF 2205 (UNS S31803) SAF 2507 (UNS S32750) Titanium

1

...

...

...

...

...

1

...

...

...

...

...

0

2

2

2

...

...

...

2

0

1

...

2

...

...

1

1

1

...

...

...

...

...

1

...

...

0

...

...

0

1

1

...

0

...

...

...

0

0

1

...

0

0

1

1

0

...

0

0

0

1

0

0

0

1

0

0

0

1

2

2

2

2

2

2

2

2

2

2

2

2

2

2

Notes: 0, corrosion rate of less than 0.1 mm/yr. The material is corrosion proof. 1, corrosion rate of 0.1–1.0 mm/yr. The material is not corrosion proof but useful in certain cases. 2, corrosion rate of more than 1.0 mm/yr. Serious corrosion. The material is not usable. BP, boiling solution. Source: Adapted from Ref 1

Table 2 Corrosion table for fuming sulfuric acid (oleum), H2SO4 + SO3 Conc. H2SO4, % Conc. SO3, % Temperature, °C

Carbon steel 13% Cr steel 18-2 (UNS S44400) 3R12 (UNS S30400) 3R60 (UNS S31600) 18-13-3 17-14-4 2RK65 (UNS N08904) Sanicro 28 (UNS N08028) 254SMO (UNS S31254) 654 SMO (UNS S32654) SAF 2304 (UNS S32304) SAF 2205 (UNS S31803) SAF 2507 (UNS S32750) Titanium

100 7 60

100 11 60

100 11 100

100 60 20

100 60 70

100 60 80

0 0 0

0 0 0

2 2 ...

... ... ...

... ... ...

... 2 ...

0

0

1

0

0

0

0

0

0

0

0

0

0 0 0

0 0 0

... ... ...

... 0 0

0 0 0

... ... ...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

...

2

2

2

2

2

2

Notes: 0, corrosion rate of less than 0.1 mm/yr. The material is corrosion proof. 1, corrosion rate of 0.1–1.0 mm/yr. The material is not corrosion proof but useful in certain cases. 2, corrosion rate of more than 1.0 mm/yr. Serious corrosion. The material is not usable. Conc., concentration. Source: Adapted from Ref 1

the potential at which passive film breakdown occurs. This is especially important for welds, which, if not annealed, can have maximum deleterious segregation by both inclusions and solidification segregation. All austenitic and duplex stainless alloys have best corrosion resistance when quenched from the solution annealing temperature. The precipitation hardening, martensitic, and ferritic alloys are more complicated but are less relevant to this topic. If information on them is needed, they are discussed in detail in their respective chapters.

When a crevice is permitted to exist, it mimics the pH-altering action found within pits in which transport restriction leads to a buildup of metal and hydrogen ions and oxygen depletion. All alloys undergo crevice corrosion under less-aggressive conditions than those required to induce pitting, so care must be taken to avoid crevices. Intergranular Corrosion Intergranular corrosion is a problem that can be avoided entirely by correct alloy selection and proper thermal processing. The principle cause of grain boundary attack is alloy depletion at the grain boundaries. The most familiar form of this problem occurs when austenitic alloys having carbon levels above 0.03% are welded. The region near the weld where temperatures reach 600 to 900 °C (1100 to 1650 °F) may have carbon migrate to and along grain boundaries, the fast diffusion paths, where it combines with less-mobile chromium atoms and precipitates as chromium carbide. This lowers the chromium level in solution, resulting in poor corrosion resistance only at the grain boundaries. This is easily prevented by selecting alloys with low carbon levels. Duplex alloys, curiously, undergo chromium carbide precipitation under the same conditions but do not undergo significant chromium depletion because the neighboring ferrite grains, in which chromium diffuses more rapidly, contribute chromium, mitigating the depletion. Precipitation segregation of all types, not just by carbides, must be guarded against. Sigma phase, nitrides, secondary austenite, and others can cause local breakdown of corrosion resistance if alloys are heated to a dangerous temperature for sufficient time. It is important to learn these potential

262 / Stainless Steels for Design Engineers

vulnerabilities by reviewing the metallurgy of any alloy selected for service.

Stress Corrosion Cracking The theory of SCC is still under debate. The reader will find the arguments confusing as the debate generates more heat than light. We will skip the theory; it can be found in the corrosion chapters. SCC, like excessive general corrosion or pitting, is avoided by referring to published test data from the corrosion tables. If a material must be used where a risk of SCC occurs, then stress levels must be managed to stay below the threshold stress for SCC. Figure 2 shows how various alloys resist SCC as a function of chloride concentration and temperature, the two most important aggravating factors. Material comparisons are made difficult because tests are normally run at a given fraction of a material’s yield strength. Thus, the data in Fig. 2 (Ref 1) must be interpreted. Higher-strength duplex alloys, while having better SCC performance than austenitics of equal corrosion resistance (e.g., 316 vs. 2304), have much better SCC resistance. Furthermore, the stress at which failure will occur is much higher since the yield strength at which the testing takes place is about twice as high for duplex alloys. SCC also exhibits a threshold stress below which failure does not occur. This is about 60% of tensile strength for duplex and about 30% for

Fig. 2

Stress corrosion cracking (SCC) resistance in neutral chloride solutions containing 8 ppm oxygen. Testing time, 1000 h. Applied stress equal to proof strength at testing temperature

austenitics. Designing within this limit is sensible practice. And, if alloy selection uses a rule of avoiding situations in which pitting can occur, SCC will also be avoided even if stress excursions occur since in general pitting is a necessary precondition for SCC.

Erosion Flow velocities can reach levels at which erosion becomes problematic, especially if hard particles are suspended in a fluid. Assuming that the material has sufficient corrosion resistance to survive well in the static environment, the best performance under erosive conditions is obtained by materials with higher surface hardness. Accordingly, the duplex perform better than austenitic alloys of the same corrosion resistance level.

Specific Environments The list of specific environments against which stainless steels are sufficiently resistant to select for use in the chemical process industries is too long to provide here. Some of the most important specific corrosives, such as nitric, sulfuric, phosphoric, hydrochloric, and organic acids and others, are covered in the chapter on corrosion. The main caution to the designer is to make sure that the source material from which design guidance is sought is current. Many otherwise excellent handbooks are somewhat obsolete in that they do not include the very importance duplex stainless steel family or only include the oldest alloys in the group, such as 2205 (UNS S32205). Many new alloys now exist that range in corrosion performance from that of 316 to that of the 6Mo-plus-N austenitics. These alloys are usable in all gauges, have high strength and toughness, resist SCC, and can achieve the corrosion resistance levels of any ferritic or austenitic alloy. They can also provide significant savings in alloy cost at the same corrosion level because they have lower nickel levels.

REFERENCES

1. Sandvik Materials Technology, www.smt. sandvik.com, accessed June 2008

Chapter 23: Chemical and Process Industry Applications / 263

2. Outokumpu Corrosion Handbook for Stainless Steels, www.outokumpu.com, accessed June 2008 3. P.A. Schweitzer, Corrosion Resistance Tables, 5th ed., National Association of Corrosion Engineers, NACE 37755, 2004 \aq2\

4. D.B. Anderson and B.D. Craig, Handbook of Corrosion Data, 2nd ed., ASM International, 1995

Stainless Steels for Design Engineers Michael F. McGuire, p 265-267 DOI: 10.1361/ssde2008p265

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

CHAPTER 24

Pulp-and-Paper Industry Applications Summary THE PULP-AND-PAPER INDUSTRY has seen more benefits from developments in stainless steel than any other industry. The harsh chemicals used in this industry called for better materials than the normal austenitic stainless steels without the expense of the 6Mo grades. This need has been met through the use of the duplex alloys, which have become the new standard.

Introduction The proximity of the Scandinavian paper industry to that region’s specialty steel industry has been symbiotic. As a result of the strong interaction between engineers having wellspecified needs for improved materials and metallurgists capable of providing them, the advances in materials in the pulp-and-paper industry have been a model of rapid technology transfer and innovation. Beginning in 1988, duplex stainless steels first went into production in kraft digesters, and there has been no turning back in the replacement of austenitic stainless steels by duplex. So, a discussion of the materials selection for the pulp-and-paper industry has changed from a fairly complicated analysis of which austenitic steel to use while guarding against stress corrosion cracking and pitting corrosion and when to use clad materials for cost savings, to a fairly simple discussion of which duplex stainless steel is most economical for a given piece of equipment. Since this revolution occurred in the 1990s before the latest surge in nickel prices, it is safe to say that future pulp-and-paper projects will be essentially the

exclusive domain of duplex stainless steels because of their lower cost per unit of corrosion resistance, high strength, and near immunity to stress corrosion cracking. Pricing changes mainly with alloying element costs, principally those of nickel and molybdenum. At prices between the highs and lows of the first decade of the 2000s, duplex costs have been roughly onethird less than that of an equivalent corrosionresisting austenitic stainless. This factored in with strength nearly double that of the equivalent austenitic make them an overwhelmingly superior choice for pulp-and-paper equivalent except if very special corrosion requirements differ from the norm, such as in bleaching.

Paper-Making Processes The kraft (German for “strong”) process was introduced in 1937, replacing the sulfite process. In the kraft process, the lignin-connecting wood fibers are dissolved under conditions of elevated temperature and pressure in acidic conditions of pH 2.0 to 4.0. This leaves a long fiber, which enables paper of high strength, hence the name kraft. Over the years the materials used for the vessels, called digesters, in which this process is carried out have been sequentially carbon steel, stainless steel, and stainless steel clad onto carbon steel. In the previous sulfite process, acidresistant brick vessels were used. Now, the digesters, essentially large vertical tanks, are constructed of 2205 (UNS S32205) as a rule (see Fig. 1). The digestion is typically carried out at 150 and 180 °C (300 and 360 °C) and 10 to 12 bar. The pH of the sulfate is around 2.0 to 4.0. In this environment, 316L can survive, but it

266 / Stainless Steels for Design Engineers

Fig. 1

The first kraft digester fabricated from alloy 2205. Courtesy of Outokumpu

requires maintenance and has a finite life. The 2205 is twice as resistant to corrosion, 0.005 mm/yr versus 0.011 mm/yr (Ref 1). In the nonchloride environment, molybdenum is not an essential alloying element, so the introduction of the use of 2101 (UNS S32101) or 2304 (UNS S32304) is a logical cost-saving move without strength or corrosion compromises. The reduction in wall thickness allowed by the higher-strength duplex depends on the engineering code required. The American Society of Mechanical Engineers (ASME) code requirement is based on tensile strength and permits only a 24% reduction in wall thickness, while the total kjeldahl nitrogen (TKN) code, based on yield strength, would allow a 46% reduction. This large a difference in strength levels required by codes is unfortunate and reflects an orientation to materials in which the yield/ tensile ratio is closer to unity, unlike either duplex or austenitic stainless steel. In the more unusual case of digesters using the sulfite process, the materials selected would be the same. As one proceeds downstream in the process, environments change greatly, but the optimal materials remain duplex for various reasons. The subsequent stage is blow tanks in which the

pulp suspension is injected at high velocity. The environment is a mixture of alkaline liquid, while the vapor phase can contain organic acids. The hardness of the duplex helps mitigate erosion, while the alloy level is beneficial against corrosion. 2205 is the alloy of choice here, but 2003 (UNS S32003) would suffice. The next step, washing and screening, has seen increasingly severe environments as closed systems required for pollution control have become more common. This has rendered the previous choice of carbon steel untenable. This stage also sees erosion potential from hard particles, such as sand, in the pulp. The optimum solution is a lean duplex such as 2101, 2304, or 2003. The delignification of the pulp comes next. This oxygen process dates from the 1970s. At first, highly alloyed austenitic alloys were used. Subsequently, it was found again that duplex performed better in that they were sufficiently corrosion resistant, but also offered freedom from stress corrosion cracking as well as materials savings because of their higher strength. The bleaching of the pulp is important for many types of paper, and this can be done by chlorine bleaches or ozone/peroxide bleaches. The chlorine bleaching now must generally be done in closed systems, which results in a buildup of chloride levels to a point at which corrosion levels are unacceptable unless very highly alloyed materials are used. The 6Mo grades have been successful, but now they can be replaced by duplex alloys such as 2507 (UNS S32750), which again save cost by virtue of their higher strength. Bleaching can be accomplished without chlorine in the so-called TCF, totally chlorine free, process. This reduces the corrosivity of the environment as the ozone and hydrogen peroxide used in the process are relatively harmless to stainless steel. Alloys such as 316 are adequate for this environment, but lean duplex, 2101 or 2304, offer cost reductions through their greater strength. In plants that use recycled paper and mechanical wood chip processing, the materials selection criteria remain the same. Duplex stainless has become the clear choice. Further downstream, containers and process equipment benefit equally from duplex down to the handrails and walkways. This wholesale use of duplex can make plants nearly maintenance

Chapter 24: Pulp-and-Paper Industry Applications / 267

free from a corrosion point of view, a dramatic change in an industry in which the thousand-fold greater corrosion rates of carbon steel presented operators with endless equipment downtime problems. Additional detail about corrosion challenges and the use of stainless steels in the pulp-andpaper industry can be found in Ref 2.

REFERENCES

1. A. Tuomi et al., Duplex America 2000 Conference, Houston, KCI Publishing, 2000 2. H. Dykstra et al, Corrosion in the Pulp and Paper Industry, Corrosion: Environments and Industries, Vol 13C, ASM Handbook, ASM International, 2006, p 762–802

Stainless Steels for Design Engineers Michael F. McGuire, p 269-278 DOI: 10.1361/ssde2008p269

APPENDIX 1

Compositions

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

0.15 0.03 0.03 0.15 0.15 0.08 0.12–0.25 0.0–3 0.08 0.08 0.08 0.04 0.06 0.10 0.12 0.12

0.15 0.08 0.03 0.15 0.03 0.15 0.08 0.15 0.15 0.15 0.15 0.08 0.03 0.04–0.10 0.08 0.08 0.03 0.08 0.04–0.06 0.12 0.018 0.16–0.24 0.08 0.05–0.10 0.20 0.08

S21500 S21600 S21603 S30100 S30153 S30200 S30430 S30215 S30300 S30223 S30310 S30400 S30403 S30409 S30451 S30452 S30453 S30424 S30415 S30500 S30600 S30615 S30800 S30815 S30900 S30908

C

S20100 S20103 S20153 S20161 S20200 S20300 S20500 S20400 S24100 S24300 S21900 S21904 S20910 S21800 S21400 S21460

Designation(a)

0.10–0.20 ... ... ... ... ... ... 0.10 0.10 ... 0.10–0.16 0.16–0.30 0.10–0.16 0.10 0.12–0.18 ... ... ... ... 0.14–0.20 ... ...

... 0.25–0.50 0.25–0.50

0.25 0.25 0.25 0.08–0.20 0.25 ... 0.32–0.40 0.15–0.30 0.20–0.40 0.20–0.40 0.15–0.40 0.15–0.40 0.20–0.40 0.08–0.18 0.35 0.25–0.50

N Cr

14.0–16.0 17.5–22.0 17.5–22.0 16.0–18.0 16.5–18.0 17.0–19.0 17.0–19.0 17.0–19.0 17.0–19.0 17.0–19.0 17.0–19.0 18.0–20.0 18.0–20.0 18.0–20.0 18.0–20.0 18.0–20.0 18.0–20.0 18.0–20.0 18.0–19.0 17.0–19.0 17.0–18.5 17.0–19.5 19.9–21.0 20.0–22.0 22.0–24.0 22.0–24.0

16.0–18.0 16.0–18.0 16.0–18.0 15.0–18.0 17.0–19.0 16.0–18.0 16.5–18.0 15.0–17.0 16.5–19.0 17.0–19.0 19.0–21.5 19.0–21.5 20.5–23.5 16.0–18.0 17.0–18.5 17.0–19.0

Composition of austenitic stainless steels

5.5–7.5 5.5–7.5 5.5–7.5 4.0–6.0 7.5–10.0 5.0–6.5 14.0–15.5 7.0–9.0 11.0–14.0 11.5–14.5 8.0–10.0 8.0–10.0 4.0–6.0 7.0–9.0 14.5–16.0 14.0–16.0 5.5–7.0 7.5–9.0 7.5–9.0 2.0 ... 2.0 2.0 2.0 2.0 2.0 2.5–4.5 2.0 2.0 2.0 2.0 2.0 2.0 ... 0.8 2.0 2.0 2.0 2.0 0.8 2.0 2.0 (continued)

9.0–11.0 5.0–7.0 7.5–9.0 6.0–8.0 6.0–8.0 8.0–10.0 8.0–10.0 8.0–10.0 8.0–10.0 8.0–10.0 7.0–10.0 8.0–10.5 8.0–10.5 8.0–10.5 8.0–10.5 8.0–10.5 8.0–10.5 12.0–15.0 9.0–10.0 10.5–13.0 14.0–15.5 13.5–16.0 10.0–12.0 10.0–12.0 12.0–15.0 12.0–15.0

Mn

3.5–5.5 3.5–5.5 3.5–5.5 4.0–6.0 4.0–6.0 5.0–6.5 1.0–1.75 1.5–3.0 0.5–2.5 2.25–3.75 5.5–7.5 5.5–7.5 11.5–13.5 8.0–9.0 0.75 5.0–6.0

Ni

... 2.0–3.0 2.0–3.0 ... ... ... ... ... 0.60 optional 0.60 optional 0.60 optional ... ... ... ... ... ... 2.0 ... ... 0.2 ... ... ... ... ...

... ... ... ... ... 0.5 ... ... ... ... ... ... 1.5–3.0 ... ... ...

Mo

Composition, % Si

1.00 1.00 2.00–3.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.75 1.00–2.00 1.00 3.75–4.25 3.2–4.0 1.00 1.4–2.0 0.75 0.75

1.20 1.00 1.00 1.00

1.00 0.75 0.75 3.00–4.00 1.00 1.00 1.00 1.0 1.00 1.00 1.00 1.00 1.00 3.50–4.50 0.30–1.00 1.00

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted. (a) Unified Number System, UNS numbers are S or N followed by 5 digits.

201 201L 201LN Gall-Tough 202 230 EZ 205 Nitronic 30 Nitronic 32 Nitronic 33 Nitronic 40 (219) 21-6-9 LC Nitronic 50 Nitronic 60 Tenelon Cryogenic Tenelon Esshete1250 216 216L 301 301LN 302 302Cu 302B 303 303Se 303 Plus X 304 304L 304H 304N 304HN 304LN 304BI 153MA 305 Cronifer 1815 RA 85 H 308 253MA 309 309S

Name

Table A1.1 P

0.040 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 ... ... ... ... 0.045 0.045 ... 0.045 0.045

0.060 0.045 0.045 0.040 0.060 0.040 0.060 0.040 0.060 0.060 0.060 0.060 0.040 0.040 0.060 0.060

S

0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.15 min 0.06 min 0.25 min 0.030 0.030 0.030 0.030 0.030 0.030 ... ... ... ... 0.030 0.030 ... 0.030 0.030

0.030 0.015 0.015 0.040 0.030 0.18–0.35 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.030 Nb 0.75–1.25 ... ... ... ... ... Cu 3.0–4.0 ... ... Se 0.15 min ... ... ... ... ... ... ... B 1.00–1.20 Ce 0.04 ... Cu 0.50 Al 0.8–1.5 ... 1.0 Al ... ...

... ... ... ... ... Cu 1.75–2.25 ... ... ... ... ... ... Nb 0.1–0.3 ... ... ...

Other

V 0.15–0.40 ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 0.03–0.08 Ce ... ...

... ... ... ... ... ... ... ... ... ... ... ... V 0.1–0.3 ... ... ...

Other

270 / Stainless Steels for Design Engineers

S30909 S30940 S30941 DIN 1.4828 S31000 S31008 S31009 S31040 S31041 S31042 S31050 DIN 1.4841 S31400 S31600 S31609 S31620 S31603 S31653 S31651 S31635 S31700 S31703 S31753 S31725 S31726 S32100

S32109

S33000 N08800 S35125 S33400 S34700

S34709

S34800

S34809

S37000 S38400

S35315

309H 309Cb 309HCb 309Si 310 310S 310H 310Cb 310HCb 310HCbN 310MoLN 310Si 314 316 316H 316F 316L 316LN 316N 316Ti 317 317L 317LN 317LM 317LMN 321

321H

330 332 332Mo* 334 347

347H

348

348H

370 384

353MA

24.0–26.9

0.12–0.18

0.08

12.5–14.5 15.0–17.0

0.005 ...

17.0–19.0

17.0–19.0

0.03–0.05 0.08

...

0.08

17.0–19.0

...

...

0.08

17.0–20.0 19.0–23.0 20.0–23.0 18.0–20.0 17.0–19.0

17.0–19.0

22.0–24.0 22.0–24.0 22.0–24.0 19.0–21.0 24.0–26.0 24.0–26.0 24.0–26.0 24.0–26.0 24.0–26.0 24.0–26.0 24.0–26.6 24.0–26.0 23.0–26.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 18.0–20.0 18.0–20.0 18.0–21.0 18.0–20.0 17.0–20.0 17.0–19.0

Cr

0.04–0.10

... ... 0.10 ... ...

0.10

0.40–0.10

0.08 0.03 0.10 0.08 0.08

... ... ... 0.11 ... ... ... ... ... 0.15–0.35 0.09–0.15 0.11 ... 0.10 ... ... 0.10 0.10–0.16 0.10–0.16 0.10 0.10 0.10 0.10–0.22 0.10 0.10–0.20 0.10

N

0.04–0.10 0.08 0.04–0.10 0.04–0.10 0.25 0.08 0.04–0.10 0.08 0.04–0.10 0.04–0.10 0.02 0.20 0.25 0.08 0.04–0.10 0.08 0.03 0.03 0.08 0.08 0.08 0.03 0.03 0.03 0.03 0.08

C

34.0–36.0

14.5–16.5 17.0–19.0

9.0–13.0

9.0–13.0

9.0–13.0

34.0–37.0 30.0–35.0 31.0–35.0 19.0–21.0 9.0–13.0

9.0–12.0

12.0–15.0 12.0–16.0 12.0–16.0 11.0–13.0 19.0–22.0 19.0–22.0 19.0–22.0 19.0–22.0 19.0–22.0 19.0–22.0 20.5–23.5 19.0–22.0 19.0–22.0 10.0–14.0 10.0–14.0 10.0–14.0 10.0–14.0 10.0–14.0 10.0–14.0 10.0–14.0 11.0–15.0 11.0–15.0 11.0–15.0 13.5–17.5 13.5–17.5 9.0–12.0

Ni

1.0 (continued)

1.65–2.35 2.0

2.0

2.0

2.0

2.0 1.5 1.0–1.5 1.0 2.0

2.0

2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 3.0–4.0 3.0–4.0 3.0–4.0 4.0–5.0 4.0–5.0 2.0

Mn

...

1.5–2.5 ...

...

...

...

... ... 2.0–3.0 ... ...

...

... ... ... ... ... ... ... ... ... ... 1.6–2.6 ... ... 2.0–3.0 2.0–3.0 1.75–2.5 2.0–3.0 2.0–3.0 2.0–3.0 2.0–3.0 2.0 2.0 2.0 2.0 2.0 ...

Mo

Composition, %

0.6–1.0

0.5–1.0 1.00

1.00

1.00

1.00

0.75–1.50 1.00 0.75 1.00 1.00

1.00

...

... 0.045

0.045

0.045

0.045

0.030 ... ... ... 0.045

0.045

P

0.045 0.040 0.040 0.040 0.045 0.045 0.045 0.045 0.045 0.045 0.020 0.045 0.045 0.045 0.045 0.20 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045 0.045

Si

0.75 1.00 1.00 1.50–2.50 1.00 1.00 1.00 1.00 1.00 1.00 0.5 1.50–2.50 1.50–3.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.75 1.00

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted. (a) Unified Number System, UNS numbers are S or N followed by 5 digits.

Designation(a)

Name

Table A1.1 (continued) S

...

... 0.030

0.030

0.030

0.030

0.030 ... ... ... 0.030

0.030

0.030 0.030 0.030 0.015 0.030 0.030 0.030 0.030 0.030 0.030 0.010 0.015 0.030 0.030 0.030 0.10 min 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.030

Other

Other

... ... Nb 10xC to 1.10 . . . Nb 10xC to 1.10 . . . ... ... ... ... ... ... ... ... Nb 10xC to 1.10 . . . Nb 10xC to 1.10. . . Nb 10xC to 1.10 . . . ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 5xC to 0.70 Ti . . . ... ... ... ... 0.030 max P ... ... ... ... ... Ti 5x(C+N) ... to 0.70 Ti 4x(C+N) ... to 0.70 ... ... Ti 0.15–0.60 Al 0.15–0.60 Nb 0.25–0.60 . . . Ti 0.15–0.60 Al 0.15–0.60 Nb+Ta 10xC ... to 1.10 Nb+Ta 10xC ... to 1.10 Nb+Ta 0xC Co 0.2 to 1.10 Nb+Ta 8xC ... to 1.0 Ti 0.10–0.40 Co 0.05 Nb+Ta 10xC Co 0.2 to 1.10 Ce 0.03–0.10 . . .

Appendix 1: Compositions / 271

... 0.15–0.25

0.15–0.25 0.35–0.60 ... ... ...

0.04 0.015

0.02 0.03 0.03 0.07 0.05

... N08031

N08932 S31266 S32200 N08007 20Mod

24.0–26.0 23.0–26.0 20.0–23.0 19.0–22.0 21.0–23.0

22.0–24.0 26.0–28.0

20.0–22.5 19.0–21.0

26.0–28.0 25.0–29.0 20.0–25.0 20.0–22.0 19.25–21.50 20.0–22.0 22.0–24.0 18.0–21.0 19.0–21.0 22.5–25.0 22.0–26.0 26.0–28.0 20.0–22.0 20.0–22.0 19.0–23.0 19.0–23.0 19.50–20.50 23.0–25.0 24.0–25.0 19.0–21.0

Cr

24.0–26.0 21.0–24.0 23.0–27.0 27.5–30.5 25.0–27.0

26.0–28.0 24.0–26.0

balance 24.0–26.0

31.0–33.0 32.0–37.0 30.0–38.0 3.25–4.50 1.50–2.75 10.5–12.5 7.9–9.0 8.0–11.0 32.0–38.0 35.0–40.0 33.0–37.0 29.9–32.5 23.5–25.5 23.5–25.5 24.0–26.0 23.0–28.0 17.50–18.50 16.0–18.0 21.0–23.0 32.0–38.0

Ni

2.0 2.0 1.0 1.5 2.5

0.75 2.0

... 2.0

1.0 1.5 1.0 8.0–10.0 7.0–9.50 1.0–1.5 1.5–3.5 0.75–1.50 0.75–1.50 1.0 1.0 2.5 2.0 2.0 2.0 2.0 1.0 5.0–7.0 2.0–3.0 2.0

Mn

4.5–6.5 5.0–7.0 2.5–3.5 2.0–3.0 4.0–6.0

2.5–3.0 6.0–7.0

12.5–14.5 6.0–7.0

... ... 4.0–4.8 ... ... ... ... 1.0–1.75 2.0–3.0 3.5–5.0 5.0–6.7 3.0–4.0 6.0–7.0 6.0–7.0 4.3–5.0 4.0–5.0 6.0–6.5 4.0–5.0 7.0–8.0 2.0–3.0

Mo

Composition, % Si

0.40 1.0 0.5 1.5 1.0

0.75 0.30

0.08 0.50

0.03 1.0 0.6–1.0 0.25 0.25 0.70–1.25 0.69–0.90 0.03–0.80 1.00 0.50 0.50 1.00 1.00 1.00 1.00 1.00 0.80 1.0 0.5 1.0

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted. (a) Unified Number System, UNS numbers are S or N followed by 5 digits.

0.15–0.25

0.015 0.02

N06022 N08926

AL 22 Cronifer 1925 hMo Cronifer 2328 Nicrofer 3127HMo URSB-8 B66 NIC 25 CN-7M N08320

... ... ... 0.28–0.50 0.20–0.40 0.15–0.25 0.28–0.38 ... ... ... ... ... ... 0.18–0.25 ... ... 0.18–0.22 0.40–0.60 0.45–0.55 ...

0.04–0.08 0.06–0.10 0.08 0.48–0.58 0.50–0.60 0.15–0.25 0.28–0.38 0.28–0.36 0.07 0.03 0.03 0.02 0.035 0.030 0.04 0.02 0.02 0.03 0.02 0.07

S33228 S25045 S35135 S63008 S63012 S63017 S63018 S63198 N08020 N08024 N08026 N08028 N08366 N08367 N08700 N80904 S31254 S34565 S32654 N08020

AC66 Incoloy 803 Incoloy 864 21-4N 21-2N 21-12N 23-8N 19-9DL 20Cb-3 20Mo-4 20Mo-6 Sanicro 28 AL-6X AL-6XN JS-700 904L 254SMO 4565 654SMO AL 20

N

C

Designation(a)

Name

Table A1.1 (continued) P

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

0.045 ...

... ...

... ... ... 0.045 0.050 0.045 0.045 ... 0.045 0.0035 0.030 0.030 0.030 0.030 0.040 0.045 0.030 ... 0.030 ...

S

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

0.035 ...

... ...

... 0.015 0.015 0.015 0.030 0.030 0.030 ... 0.035 0.035 0.030 0.030 0.030 0.030 0.030 0.035 0.010 ... 0.005 ...

Other

3.0–4.0 Cu ...

1.0–2.0 Cu 0.5–3.0 Cu

2.5–3.5 Cu 1.0–1.4 Cu

2.5 Co,0.35 V 0.5–1.5 Cu

0.05–0.10 Ce Ti 0.15–0.60 Ti 0.4–1.0 ... ... ... ... 0.1–0.35 Ti 8xC to 1.00 Nb Nb 0.15-0.35 8xC Nb ... 0.030 P 0.040 P 8xC to 0.5 Nb 1.0–2.0 Cu 0.5–1.0 Cu 0.10 Nb 0.3–0.6 Cu 3.0–4.0 Cu

Other

... W 1.0-3.0 ... ... ...

Ti 0.4-0.7 ...

Al 0.025 0.15–0.60 Al ... ... ... ... ... 0.25–0.60 Nb Cu 3.0–4.0 Cu 0.5–1.5 2.0–4.0 Cu 0.6–1.4 Cu ... ... 0.5 Cu ... ... ... ... 8xC to 1.00 Nb+Ta W 2.5-3.5 ...

272 / Stainless Steels for Design Engineers

S40910 S40920

409 409

...

0.01

0.030

0.020

S43035

S43932

439

439LT

ATI, AK 439 HP alloys 439 ultraform S46800 468

18 Cr–Cb

13.0

...

0.07

0.12 0.10

...

0.03

0.04

18.0

18.0–20.0

17.5

17.0–19.0

17.0–19.0

16.0–18.0 16.0–19.5

... ... ...

... 0.04

14.0–16.0 16.0–18.0 16.0–18.0

...

0.08

0.12 0.12 0.12

13.5

...

13.0

12.0

0.015

10.5–11.7 11.35

0.025

0.025

0.020

0.03 0.015

10.5–11.7

10.5–11.7 10.5–11.75

10.5–11.7 10.5–11.75

Cr 11.5–14.5 12.0–13.0 10.5–11.75

...

0.50

0.2

0.50

0.50

... 1.00

0.75 0.75 ...

...

...

...

...

0.5–1.0 0.20

0.50

0.50 0.05

0.50 0.50

Ni 0.60 ... 0.50

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted

AK alloy typical

0.01

0.03

S43023 S43036

430Se 430Ti

429 430 430F

4724

Alfa II

Alfa I

12 SR

0.03 0.010

S40975 AK alloy typical AK alloy typical ATI alloy typical ATI alloy typical Outukumpu typical S42900 S43000 S43020

409Ni 11 Cr–Cb

...

0.06

S40940

409Cb

0.020 0.020

0.030 0.030

N ... ... ...

0.02 0.02

0.03 0.03

C 0.08 0.05 0.08

409 ultraform AK alloy S40930 466

405 400 409

designation S40500 AK alloy S40900

UNS

Composition of ferrite stainless steels

Name

Table A1.2

0.30

1.00

0.35

1.00

1.00

1.25 1.00

1.00 1.00 1.25

0.70

...

...

...

...

...

... ...

... ... ...

...

...

...

...

... ...

...

... ...

... ...

0.45

1.00

0.45

1.0

1.00

1.00 1.00

1.00 1.00 1.00

1.0

0.03

0.03

...

1.0 1.30

...

0.040

...

0.040

0.040

0.060 0.040

0.040 0.040 0.060

...

...

...

...

0.040 ...

...

... 0.040

1.00 1.00 1.0

0.040 0.040

P 0.040 ... 0.045

1.0 1.00

Si 1.00 1.00 1.00

Composition, % Mo ... ... ...

(continued)

0.035

0.035

...

1.00 0.25

1.00

0.75 1.00

1.00 1.00

Mn 1.00 1.00 1.00

...

0.030

...

0.030

0.030

0.030 0.030 0.150 min 0.060 0.030

...

...

...

...

0.030 ...

...

... 0.030

0.030 0.030

S 0.030 ... 0.045

Ti+Nb:0.20+ 4x(C+N) to 1.10 0.25

... 0.20+4x(C+N) to 1.10 0.20+4x(C+N) to 1.10 0.20+4x(C+N) to 0.75 Ti+Nb 0.35

... ... ...

...

0.40

0.40

0.30

Ti ... ... 6x(C+N) to 0.75 6x(C+N)to 0.5 8x(C+N) to 0.15–0.50 8x(C+N) 0.8+ 8x(C+N) Ti+Nb 10xC to 0.75 Nb ... ...

0.55

...

...

...

...

... ...

... ... ...

...

...

...

0.60

... 0.35

...

... ...

0.17 ...

Nb ... ... ...

...

...

...

Al 0.15

...

SE 0.15 AL 0.15

... ... ...

Al 1.0

AL 4.0

AL 3.0

AL 1.2

... ...

...

... ...

... ...

Other Al 0.10–0.30 Al 0.25 Al ...

Appendix 1: Compositions / 273

S44735

S44600

29-4C

446

0.25

...

0.035

23.0–27.0

28.0–30.0

25.0–27.0

24.5–26.0

25.0–27.5

22.0

24.0

20.0

17.5–19.5

18.0–23.0 17.3

17.5–18.5

16.0–18.0 16.0–18.8

18.0

17.30

Cr

0.6

0.5

1.5–3.5

3.5–4.5

0.50

0.3

...

0.25

1.0

0.6 0.3

...

... ...

...

Ni 0.25

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted.

0.20

0.025

0.025

S44660

Sea–cure

0.015

0.01

0.035

...

0.03

0.025

...

0.08

S44635

...

0.035

... 0.015

0.01

Monit

433

0.025

0.20 0.01

S44200 ATI alloy typical S44400

...

... ...

0.12 0.12

0.030

...

N . . ..

0.08

C 0.015

S44100

designation AK alloy typical Outukumpu typical S43400 S43600

UNS

(continued) of ferrite stainless steels Composition

ATI alloy typical Outukumpu 4762 typical 453 ATI alloy typical E-Brite, 26-1 S44627

444, YUS 190-EM

441, 4509, 430J1L 442 436S

434 436

4742

18SR

Name

Table A1.2

1.50

1.00

1.00

1.00

0.40

0.3

0.7

0.30

1.0

1.0 0.20

1.00

1.0 1.0

0.7

Mn 0.30

...

3.5–4.5

2.5–3.5

3.5–4.5

0.75–1.25

...

...

...

0.75–1.25

... 1.2

...

0.75–1.25 0.75–1.25

...

1.00

0.75

1.00

0.75

0.40

0.3

1.4

0.4

1.0

1.0 0.4

1.0

1.0 1.0

1.3

Si . . ..

Composition, % Mo . . ..

0.040

0.040

0.040

0.040

0.020

...

...

...

0.040

0.040 ...

0.040

0.040 0.040

...

P . . ..

0.030

0.030

0.030

0.030

0.020

...

...

...

0.030

0.030 ...

0.030

0.030 0.030

...

S ...

... ...

...

Ti

... Ti+Nb:0.20+ 4x(C+N) to 0.80 Ti+Nb:0.20+ 4x(C+N) to 0.80 Ti+Nb:0.20+ 4x(C+N) to 0.80 ...

0.02

...

Ti+Nb:0.20+ 4x(C+N) to 0.80 ...

... 8x(C+N) min

0.1–0.6

0.25

...

...

...

...

0.5–0.20

...

...

10x(C+N)

...

... ...

... Nb+Ta 5xC:0.70 9xC 0.3–1.0

...

Nb ...

...

...

...

0.60 Al 0.10 REM 0.2 Cu 0.5 Cu+Ni ...

Al 1.5

...

...

... ...

...

... ...

Al 1.0

Other Al 1.7

274 / Stainless Steels for Design Engineers

designation S40300 S41000 S41003 S41008 S41040 S41003 S41400 S41425 S41500 S41600 S41623 S41800 S42000 DIN 1.4116 nominal S42020 S42023 S42200 S42400 S42500 ... ... JFE nominal JFE nominal Nippon nominal Nippon nominal JFE nominal JFE nominal S43100 S44002 S44003 S44004 S44020 S44023 Nominal PM Nominal PM Nominal PM Nominal PM Nominal PM Nominal PM Nominal PM

UNS

0.15 min 0.15 min 0.20–0.25 0.06 max 0.08–0.20 0.50–0.55 0.30 max 0.025 0.025 0.03 0.02 0.01 0.01 0.20 max 0.60–0.75 0.75–0.95 0.95–1.20 0.95–1.20 0.95–1.20 1.15 1.05 1.05 1.05 1.45 2.15 2.20

C 0.15 max 0.15 max 0.03 0.08 0.18 max 0.030 max 0.15 max 0.05 0.05 max 0.15 max 0.15 max 0.15–0.20 0.15 min 0.50

... ... ... ... ... ... ... ... ... 0.040 0.015 0.010 0.010 ... ... ... ... ... ... ... ... ... ... ... ... ...

N ... ... ... ... ... ... ... 0.06 0.12 ... ... ... ... ... ... 12.0–14.0 12.0–14.0 11.0–13.5 12.0–14.0 14.0–16.0 13.0–14.0 12.0–14.0 13.0 13.0 12.7 12.3 11.0 12.0 15.0–17.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 16.0–18.0 14.5 14.0 14.0 14.0 14.0 17.0 13.0

Cr 11.5–13.5 11.5–13.5 10.5–12.5 11.5–13.5 11.5–13.5 10.5–12.5 11.5–13.5 12.0–15.0 11.5–14.0 12.0–14.0 12.0–14.0 12.0–14.0 12.0–14.0 14.5

Composition of martensitic stainless steels

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted

420F 420FSe 422 424 425 425 mod Trinamet HP13Cr-1 HP13Cr-2 NT-CRS NT-CRSS KL-12Cr KL-HP 12Cr 431 440A 440B 440C 440F 440FSe BG-42 ATS-34 14-4 CrMo 154 CM CPM S30V CPM S60V CPM S90V

403 410 410S 410 410Cb 412 414 414 mod 415 416 416Se 418 420 4116

Name

Table A1.3

... ... 0.50–1.00 3.50–4.50 1.00–2.00 0.50 ... 4.0 5.0 4.5 5.8 2.4 5.5 1.25–2.50 ... ... ... 0.75 0.75 ... ... ... ... ... ... ...

Ni ... ... 1.5 ... ... 1.5 1.25–2.50 4.0–7.0 3.50–5.50 ... ... 1.80–2.20 ... ... 1.25 1.25 1.00 0.50–1.0 1.00 1.00 1.00 0.45 0.45 1.45 2.0 ... ... 1.00 1.00 1.00 1.00 1.25 1.25 ... 0.4 0.5 0.45 ... 0.40 ...

1.00 1.00 0.75 0.30–0.60 1.00 1.00 1.00 ... ... ... ... ... ... 1.00 1.00 1.00 1.00 1.00 1.00 0.3 0.35 0.3 0.3 ... ... ...

Mo 0.50 1.00 ... 1.00 1.00 1.00 1.00 0.60 0.60 1.00 1.00 0.50 1.00 ...

Composition, % Mn 1.00 1.00 1.00 1.50 1.00 1.50 1.00 0.5–1.0 0.50–1.0 1.25 1.25 0.50 1.00 ... 1.00 1.00 0.75 0.30–0.6 1.00 1.00 1.00 ... ... ... ... ... ... 1.00 1.00 1.00 1.00 1.00 1.00 0.3 0.35 0.3 0.3 ... ... ...

Si 0.50 1.00 ... 1.00 1.00 1.00 1.00 0.60 0.60 1.00 1.00 0.50 1.00 ... 0.040 0.040 0.040 0.040 0.040 0.040 0.040 ... ... ... ... ... ... 0.040 0.040 0.040 0.040 0.040 0.060 ... ... ... ... ... ... ...

P 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.040 0.060 0.060 0.040 0.040 0.040 0.15 0.06 0.030 0.030 0.010 0.030 0.030 ... ... ... ... ... ... 0.030 0.030 0.030 0.030 0.10–0.35 0.060 ... ... ... ... ... ... ...

S 0.030 0.030 0.030 0.030 0.030 0.030 0.030 0.005 0.030 0.15 0.30 0.060 0.030 0.030 ...

... Se 0.15 min 0.75–1.25 W ... ... ... CU 2.0–3.0 ... ... 1.5 Cu 1.5 Cu 0.5 Cu ... ... ... ... ... ... Se 0.15 min 1.2 V ... ... ... 4.0 V 5.5 V 9.0 V

Other ... ... ... ... Nb 0.05–0.30 ... ... Cu 0.30 ... ... Se 0.15 min W 2.50–3.50 ... ...

Appendix 1: Compositions / 275

0.21

0.02 0.01 0.09 0.09 0.07–0.11 0.10–0.15 0.08

0.08 0.07 0.07 0.05 0.05 0.05

...

... ... ... ... 0.07–0.13 0.07–0.13 ...

... ... ... 0.010 ... ...

10.0

11.0–12.50 10.5–11.50 16.0–18.0 14.0–16.0 16.0–17.0 15.0–16.0 13.5–16.0

15.0–17.0 115.5–17.5 14.0–15.5 12.25–13.25 14.0–16.0 11.0–12.50

Cr

5.5

10.75–11.25 7.5–8.5 6.5–7.75 6.5–7.75 4.0–5.0 4.0–5.0 14.0–27.0

6.0–7.5 3.0–5..0 3.5–5.5 7.5–8.5 5.0–7.0 7.5–9.5

Ni

0.10

0.25 0.50 1.00 1.00 0.50–1.25 0.50–1.25 2.00

1.0 1.0 1.0 0.2 1.0 0.50

Mn

S32900 S31200 S31260 S31500 S31830 S32001 S32003 S32101 S32205 S32304 S32520 S32550 S32750 S32760 S32906 S32950 S39274 S39277

Designation

0.08 0.03 0.03 0.30 0.03 0.03 0.03 0.04 0.03 0.03 0.03 0.04 0.03 0.03 0.03 0.03 0.03 0.025

C

... 0.14–0.20 0.10–0.30 0.05–0.10 0.08–0.20 0.05–0.17 0.14–0.20 0.20–0.25 0.14–0.20 0.05–0.20 0.20–0.35 0.10–0.25 0.20–0.30 0.20–0.30 0.30–0.40 0.15–0.35 0.24–0.32 0.23–0.33

N

23.0–28.0 24.0–26.0 24.0–26.0 18.0–19.0 21.0–23.0 19.5–21.5 19.5–21.– 21.0–22.0 22.0–23.0 21.5–23.5 24.0–26.0 24.0–27.0 24.0–26.0 24.0–26.0 28.0–30.0 26.0–29.0 24.0–26.0 24.0–26.0

Cr

Composition of selected duplex stainless steels

2.5–5.0 5.5–6.0 5.5–7.5 4.25–5.25 2.5–3.5 1.0–3.0 3.0–4.0 1.35–1.70 4.5–6.5 3.0–5.0 5.5–8.0 6.0–8.0 6.0–8.0 6.0–8.0 5.8–7.5 3.5–5.2 6.0–8.0 6.5–8.0

Ni

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted

329 44LN DP3 3RE60 2205 (old) 19 D 2003 2101 2205 2304 Uranus 52N+ 255 2507 Zeron 100 2906 7-Mo Plus DP3W AF 918

Name

Table A1.5

2.0

0.75–1.25 4.5–5.5 ... 2.0–3.0 2.5–3.25 2.5–3.25 1.0–1.5

... ... ... 2.0–2.5 0.5–1.0 0.5

Mo

Composition, %

1.0 2.0 1.0 1.2–2.0 2.0 4.0–6.0 2.0 4.0–6.0 1.0 2.5 1.5 1.5 1.2 1.0 0.8–1.5 2.0 1.0 0.8

Mn

1.0–2.0 1.2–2.0 2.5–3.5 2.5–3.0 2.5–3.5 ... 1.5–2.0 0.1–0.8 3.0–3.5 ... 3.0–5.0 2.9–3.9 3.0–5.0 3.0–5.0 1.5–2.6 1.0–2.5 2.5–3.5 3.0–4.0

Mo

Composition, %

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted. (a) Nominal value

...

S46500 S17600 S17700 S15700 S35000 S35500 S66286

Custom 465 Custom 475 17-7 PH 15-7 PH AM-350 AM-355 A-286

Ferrium S53(a)

S17600 S17400 S15500 S13800 S45000 S45500

Stainless W 17-4 PH 15-5 PH 13-8 PH Custom 450 Custom 455

N

Designation

Name

C

Composition of selected precipitation–hardenable stainless steels

Table A1.4 Si

Si

0.75 1.0 0.75 1.4–2.0 1.0 1.0 1.0 1.0 2.0 1.0 0.8 1.0 0.8 1.0 0.5 0.6 0.8 0.8

0.10

0.25 0.50 1.00 1.00 0.50 0.50 1.00

1.00 1.00 1.00 0.10 1.00 0.50

P

...

0.015 0.015 0.040 0.040 0.040 0.040 0.040

0.040 0.040 0.040 0.010 0.030 0.040

P

0.040 0.045 0.030 0.030 0.030 0.040 0.040 0.040 0.030 0.040 0.035 0.040 0.035 0.030 0.030 0.035 0.030 0.030

S

...

0.010 0.010 0.030 0.030 0.030 0.030 0.030

0.030 0.030 0.030 0.008 0.030 0.030

Other

0.030 0.030 0.020 0.030 0.020 0.030 0.030 0.030 0.020 0.040 0.020 0.030 0.020 0.010 0.030 0.010 0.020 0.020

S

Cu

Other

... ... 0.1–0.5 ... ... ... ... ... ... ... ... ... ... 0.5–1.0 ... ... 1.5–2.5 0.8–1.2

W

... Al 1.0–1.5 Al 0.75–1.5 Al 0.75–1.5 ... ... Al 0.35 B 0.001–0.010 W 1, V 0.3

Al 0.4 Nb .015–0.45 Nb 0.15–0.45 Al 0.90–1.35 Nb 8XC Nb +Ta 0.1–0.5

0.75 1.0 0.75 1.4–2.0 1.0 1.0 1.0 1.0 2.0 1.0 0.8 1.0 0.8 1.0 0.5 0.6 0.8 0.8

Ti 0.4–1.2 Cu 3.0–5.0 Cu 2.5–4.5 ... Cu 1.25–1.75 Cu 1.5–2.5, Ti 0.08–1.4 Ti 1.50–1.80 Co 8.0–9.0 ... ... ... ... Ti 1.9–2.35 V 0.10–.050 Co 14

276 / Stainless Steels for Design Engineers

Wrought equivalent(a)

410 ... 420 420F 431,442 446 ... S41500 422 17–4PH 15–5PH 2205 (S32205) 255 (S32550) (S32760) ... ... ... 2507 (S32750) ... 312 304L 316L 316LN 304 347 316 304H 316H 316H Nitronic™60 316 303 302 Nitronic™50 317 308 309S 309H 309 254SMO™ 310 904L AL–6XN®

J91150 J91151 J91153 J91154 J91803 J92613 J91650 J91540 J91422 J92180 J92110 J92205 J93373 J93380 J93370 J93372 J93371 J93404 J93345 J93423 J92500 J92800 J92700 J92600 J92710 J92900 J92590 J92901 J92971 J92972 ... J92701 J92602 J93790 J93000 J93001 J93400 J93401 J93402 J94653 J94202 J94652 J94651

0.15 0.15 0.40 0.2–0.4 0.30 0.30 0.06 0.06 0.20–0.28 0.07 0.07 0.03 0.03 0.03 0.04 0.04 0.06 0.03 0.08 0.30 0.03 0.03 0.03 0.08 0.08 0.08 0.04–0.10 0.04–0.10 0.10 0.10 0.12 0.16 0.20 0.06 0.08 0.12 0.08 0.04–0.10 0.20 0.025 0.20 0.03 0.03

C

... ... ... ... ... ... ... ... ... 0.05 0.05 0.10–0.30 0.22–0.33 0.20–0.30 ... 0.10–0.25 0.15–0.25 0.10–0.30 0.10–0.30 ... ... ... 0.10–0.20 ... ... ... ... ... ... 0.08–0.18 ... ... ... 0.20–0.40 ... ... ... ... ... 0.18–0.24 ... ... 0.18–0.24

N

11.5–14.0 11.5–14.0 11.5–14.0 11.5–14.0 18.0–22.0 26.0–30.0 10.5–12.5 11.5–14.0 11.0–12.5 15.5–17.7 14.0–15.5 21.0–23.5 24.0–26.7 24.0–26.0 24.5–26.5 24.5–26.5 24.0–27.0 24.0–26.0 22.5–25.5 26.0–30.0 17.0–21.0 17.0–21.0 17.0–21.0 18.0–21.0 18.0–21.0 18.0–21.0 18.0–21.0 18.0–21.0 15.0–18.0 16.0–18.0 18.0–21.0 18.0–21.0 18.0–21.0 20.5–23.5 18.0–21.0 20.0–23.0 22.0–26.0 22.0–26.0 22.0–26.0 19.5–20.5 23.0–27.0 20.0–22.0 20.0–22.0

Cr

1.0 1.0 1.0 1.0 2.0 4.0 6.0–8.0 3.5–4.5 0.5–1.0 3.6–4.6 4.5–5.5 4.5–6.5 5.6–6.7 6.5–8.5 4.75–6.0 4.7–6.0 4.0–6.0 6.0–8.0 8.0–11.0 8.0–11.0 8.0–12.0 8.0–12.0 9.0–13.0 8.0–11.0 9.0–12.0 9.0–12.0 8.0–11.0 9.0–12.0 13.0–16.0 8.0–9.0 9.0–12.0 9.0–12.0 8.0–11.0 11.5–13.5 9.0–13.0 10.0–13.0 12.0–15.0 12.0–15.0 12.0–15.0 17.5–19.5 19.0–22.0 23.0–27.0 23.0–27.0 (continued)

Ni

0.50(b) 0.15–1.00 0.50(b) ... ... ... ... 0.4–1.0 0.9–1.25 ... ... 2.5–3.5 2.9–3.8 3.0–4.0 1.75–2.25 1.75–2.25 1.75–2.25 4.0–5.0 3.0–4.5 ... ... 2.0–3.0 2.0–3.0 ... ... 2.0–3.0 ... 2.0–3.0 1.75–2.25 ... 2.0–3.0 1.5 ... 1.5–3.0 ... ... ... ... ... 6.0–7.0 ... 4.5–5.5 6.0–7.0

Mo

Composition, %

1.00 1.00 1.00 1.00 1.00 1.00 0.50 1.00 0.5–1.0 0.70 0.70 1.50 1.20 1.00 1.00 1.00 1.00 1.50 1.00 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 1.50 7.0–9.0 1.50 1.50 1.50 4.0–6.0 1.50 1.50 1.50 1.50 1.50 1.20 2.00 2.00 2.00

Mn

1.50 0.65 1.50 1.50 1.50 1.50 1.00 1.00 1.00 1.00 1.00 1.00 1.10 1.00 1.00 1.00 1.00 1.00 1.50 2.00 2.00 2.00 1.50 2.00 2.00 2.00 2.00 1.50 1.50 3.5–4.5 2.00 2.00 2.00 1.00 1.50 2.00 1.50 2.00 2.00 1.00 2.00 1.00 1.00

Si

0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.17 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04

P

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted. (a) The wrought equivalent composition is not the same as the cast. (b) Mo is not an intentional addition.

CA-15 CA-15M CA-40 CA-40F CB-30 CC-50 CA-6N CA-6NM CA-28MWV CB-7Cu-1 CB-7Cu-2 CD-3MN CD-3MCuN CD-3MWCuN CD-4MCu CD-4MCuN CD-6MN CE-3MN CE-8MN CE-30 CF-3 CF-3M CF-3MN CF-8 CF-8C CF-8M CF-10 CF-10M CF-10MC CF-10SMnN CF-12M CF-16F CF-20 CG-6MMN CG-8M CG-12 CH-8 CH-10 CH-20 CK-3MCuN CK-20 CN-3M CN-3MN

UNS designation

Composition of Alloy Casting Institute (ACI) heat– and corrosion–resisting casting alloys (continued)

Corrosion–resisting alloys

Name

Table A1.6

0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04

S

... ... ... 0.20–0.40 S ... ... ... ... 0.9–1.25 W, 0.2–0.3V 2.5–3.2 Cu, 0.2–0.35Nb 2.5–3.2 Cu, 0.2–0.35Nb Cu 1.0 Cu 1.4–1.9 Cu 0.5–1.0, W 0.5–1.0 Cu 2.75–3.25 Cu 2.75–3.25 Cu 1.75–2.5 ... ... ... ... ... ... ... Nb 8XC min ... ... ... (10xC)–1.2 Nb ... ... Se 0.2–0.35 ... 0.1–0.3 Nb, 0.1–0.3 V ... ... ... ... ... Cu 0.5–1.0 ... ... ...

Other

Appendix 1: Compositions / 277

0.20 0.50 0.50 0.20–0.50 0.20–0.40 0.20–0.50 0.20–0.50 0.20–0.60 0.25–0.35 0.35–0.45 0.20–0.60 0.20–0.50 0.35–0.75 0.45–0.55 0.35–0.75 0.25–0.35 0.35–0.75 0.35–0.75 0.35–0.75

J82090 J92605 J93005 J93403 J92603 J93505 J94003 J94224 J94203 J94204 N08604 J94213 N08705 ... N08605 N08603 N08005 N08006 N06050

504 446 327 312 302B 309 ... 310 ... ... ... ... ... ... 330 ... ... ... ...

... ... ... ... ... 0.2 ... ... ... ... ... ... ... ... ... ... ... ... ...

... ... ...

N

8–10 26–30 26–30 26–30 19–23 24–28 26–30 24–38 23.0–27.0 23.0–27.0 28–32 19–23 24–28 24–28 13–17 13.0–17.0 17–21 10–14 15–19

19.0–22.0 18.0–20.0 19.0–21.0

Cr

... 4 max 4–7 8–11 9–12 11–14 14–18 18–22 19.0–22.0 19.0–22.0 18–22 23–27 33–37 33–37 33–37 33.0–37.0 37–41 58–62 64–68

27.5–30.0 22.0–25.0 31.0–34.0

Ni

1.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.5 2.5 2.5 2.5 2.5 2.5

0.9–1.2 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b) 0.5(b)

2.0–3.0 2.5–3.0 ...

Mo

Composition, %

1.50 1.50 0.15–1.5

Mn Si

0.35–0.65 1.0 1.5 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0

1.50 3.50 0.50–1.5

P

0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.035 0.04 0.04 0.04 0.04 0.04

0.04 0.04 0.04

Notes: All compositions include Fe as balance. Single values are maximum values unless otherwise noted. (a) The wrought equivalent composition is not the same as the cast. (b) Mo is not an intentional addition.

0.07 0.07 0.05–0.15

C

N08007 J94650 N08151

UNS designation

320 ... ...

Wrought equivalent(a)

(continued)

CN-7M CN-7MS CT-15C Heat resisting alloys HA HC HD HE HF HH HI HK HK-30 HK-40 HL HN HP HP-50WZ HT HT-30 HU HW HX

Name

Table A1.6 S

0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.035 0.04 0.04 0.04 0.04 0.04

0.04 0.04 0.04

Other

... ... ... ... ... ... ... ... ... ... ... ... ... W 4.0–6.0, Zr 0.1–1.0 ... ... ... ... ...

Cu 3.0–4.0 Cu 1.5–2.0 Nb 0.5–1.5

278 / Stainless Steels for Design Engineers

Stainless Steels for Design Engineers Michael F. McGuire, p 279-280 DOI: 10.1361/ssde2008p279

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

APPENDIX 2

Physical and Mechanical Properties of Select Alloys Table A2.1

Physical properties of major stainless steel engineering alloys

UNS

201 301 304 304L 305 316L 321 904L AL6-XN© 409 430 439 468 410 2101 2003 2205 2507

S20100 S30100 S30400 S30403 S30400 S31603 S32100 N08904 N08367 S40920 S43000 S43035 S46800 S41000 S32101 S32003 S32205 S32750

Density, kg/dm3

7.86 8.03 7.90 7.90 7.90 8.00 7.92 7.95 8.06 7.76 7.70 7.70 7.76 7.65 7.8 7.72 7.8 7.8

Modulus of elasticity, GPa

207 193 200 200 200 200 193 190 200 200 200 200 200 200 200 210 200 200

Coefficient of thermal exp., 10–6 × K–1

Thermal conductivity, W/M·°K

Specific heat, J/kg·°K

Electrical resistivity, Ω·mm2/m

16.6 16.6 16.6 16.6 16.6 16.5 16.6 15.3 15.3 10.5 10.3 10.2 10.5 10.5 13.5 13.5 14.6 12.5

16.3 16.3 16.3 16.3 16.3 14.6 16.3 13.2 11.8 25.0 23.9 24.2 25.0 24.9 17.0 17.0 16.5 13.5

502 500 500 500 500 480 500 460 474 477 460 460 477 460 500 510 500 500

0.67 0.73 0.72 0.72 0.72 0.74 0.72 0.95 0.89 0.60 0.60 0.63 0.60 0.56 0.80 0.80 0.80 0.80

280 / Stainless Steels for Design Engineers

Table A2.2

Typical minimum mechanical properties of representative stainless steel engineering alloys

Name

Condition

UNS

201

Annealed

S20100

201F

2B

301

Annealed

301 tensile

Tensile strength, MPa

Elongation, %

Hardness

260 min

550 min

40 min

100 Rb max

S20100

330

700

51

89 Rb

S30100

205 min

515 min

40 min

95 Rb max

2D 1 hard /4

S30100

320

850

49

88 Rb

S30100

580

900

32

25 Rb

301

1

S30100

815

1150

23

35 Rc

301

3

S30100

1000

1270

17

40 Rc

301

/4 hard Full hard

S30100

1160

1380

12

42 Rc

301 sink

2D

S30100

270

690

57

82 Rb

304

Annealed

S30400

205 min

515 min

40 min

92 Rb max

304

Hot rolled

S30400

335

640

51

86 Rb

304

2D

S30400

265

625

55

81 Rb

304

2B

S30400

305

635

52

85 Rb

304

#4 polish

S30400

325

650

51

85 Rb

304

2BA

S30400

315

640

53

85 Rb

304

1

S30400

705

890

23

29 Rc

304L

/4 hard Annealed

S30403

170 min

485 min

40 min

92 Rb max

304L

2D

S30403

255

590

53

80 Rb

304LT

2D

S30403

255

600

51

81 Rb

304DD

2D

S30400

270

610

55

82 Rb

304EDD

2D

S30400

260

600

56

78 Rb

305

Annealed

S30500

170 min

485 min

40 min

88 Rb max

305

2D

S30500

245

560

52

73 Rb

316L

Annealed

S31603

170 min

485 min

40 min

95 Rb max

316L

2B

S31603

310

595

51

82 Rb

321

Annealed

S32100

205 min

515 min

40 min

95 Rb max

321

2B

S32100

285

570

49

78 Rb

904L

Annealed

N08904

220 min

490 min

35 min

90 Rb max

904L

2B

N08904

270

605

50

AL6-XN©

Annealed

N08367

310 min

690 min

30 min

AL6-XN©

2B

N08367

365

745

47

88 Rb

409

Annealed

S40920

170 min

380 min

20 min

88 Rb max

409

2D

S40920

260

440

31

60 Rb

430

Annealed

S43000

205 min

450 min

20 min

89 Rb max

430

2B

S43000

345

515

27

67 Rb

439

Annealed

S43035

205 min

415 min

22 min

89 Rb max

439

2D

S43035

315

455

32

76 Rb

468

Annealed

S46800

205 min

415 min

22 min

90 Rb max

468

2D

S46800

205

415

32

76 Rb

29-4C

Annealed

S44735

415 min

550 min

18 min

25 Rc max

29-4C

2D

S44735

550

650

20

20 Rc

410

Annealed

S41000

205 min

450 min

20 min

96 Rb max

410

2B

S41000

320

515

28

81 Rb

2101

Annealed

S32101

530 min

700 min

30 min

301

/2 hard

Yield strength, MPa

79 Rb 100 Rb max

...

(a) Finish conditions: 2D is cold rolled, annealed, and pickled; 2B is 2D with an added temper mill pass (approximately 0.5% reduction); 2BA is cold rolled, bright annealed, and temper passed.

Stainless Steels for Design Engineers Michael F. McGuire, p 281-283 DOI: 10.1361/ssde2008p281

Copyright © 2008 ASM International® All rights reserved. www.asminternational.org

APPENDIX 3

Introduction to Thermo-Calc and Instructions for Accessing Free Demonstration Version WITHIN THE MAIN BODY of this textbook, a number of diagrams have been plotted and attributed to a software package called Thermo-Calc. The purpose of this appendix is to give a brief introduction to Thermo-Calc, explain what it is, and what are its uses. Also provided are instructions for accessing a demonstration version of the software.

What Is Thermo-Calc? Thermo-Calc (Ref 1) is a powerful, flexible software package available from Thermo-Calc Software AB for performing various kinds of thermodynamic and phase diagram calculations for multicomponent systems. The software is based on the so-called CALPHAD (CALculation of PHAse Diagrams) method (Ref 2), which describes mathematically the thermodynamics of a system through a representation of the Gibbs energies of the different crystalline phases relevant to that system and defined by the chemical composition of the system. Thermo-Calc minimizes the total Gibbs energy of the system with respect to various constraints such as temperature, pressure, and chemical composition and thus predicts the most stable energy state (or equilibrium state) that can form. By suspending certain phases (i.e., manually removing certain selected phases from the system and thus restricting the formation of such phases), Thermo-Calc can also be used to investigate meta-stable equilibria-type problems.

Thermo-Calc is used in conjunction with thermodynamic databases containing polynomial functions that describe the Gibbs energies of the different phases according to certain models that take into consideration nonideal chemical interactions in solution phases. These databases are based on the critical evaluation of thermodynamic and phase equilibria data for binary, ternary, and some higher-order systems, which are then assembled into self-consistent databases. Different databases are available for different broad classifications of materials, systems, or applications. For example, there are databases for steels and iron-based alloys; ironbased slags; nickel superalloys; aluminum, magnesium, titanium, and zirconium alloys; cemented carbides; nuclear materials; and more. Further information on the different databases available can be found at the Thermo-Calc Web site: www.thermocalc.com The thermodynamic database for steels (Ref 3), as developed by Thermo-Calc Software AB, was used in conjunction with Thermo-Calc for all the calculations made during the preparation of this book. The version of the database used for these calculations contains data for 20 elements and 85 phases. Although the databases are based primarily on the critical assessments of binary, ternary, and some quaternary systems, the CALPHAD methodology provides a theoretical framework on which extrapolations can be made to predict the phase equilibria for higher-order, multicomponent systems (the higher the order of the system, the weaker the nonideal interaction

282 / Stainless Steels for Design Engineers

parameters become). Thermo-Calc can therefore be used in conjunction with such databases to make predictions for multicomponent systems and alloys of industrial relevance as illustrated by some of the examples given in the main body of this book. These calculations can be validated against real alloy data if this information is available but is not based on (or adjusted to fit) such higher-order alloy data. Higher order in this sense means more than four elements (i.e., larger than a quaternary system). Four specific types of calculation can be performed using Thermo-Calc, although the range of problems to which these can be applied is broader: 1. Single-Point Equilibria: The temperature, pressure, composition/activity of a component (or the amount of a phase) are fixed and the stable or meta-stable equilibrium for the specified conditions is calculated. 2. Step: The amount of one state variable parameter (or condition) can be changed, while the other conditions remain fixed. For example, to see how the different phases and their amounts and compositions would vary with temperature for a given alloy, one would “step in” temperature. Alternatively, one can vary the composition of one of the components/elements and calculate how the phase amounts change for a fixed temperature or predict how the solidus or liquidus would change with varying alloy composition. 3. Map: Two axis variables (such as temperature, pressure, composition, or activity of the components) are changed at the same time. Isoplethal sections are generated by varying temperature and composition of one of the components. Isothermal sections are the result of varying the amounts of two of the components for a fixed temperature. Examples of each of these kinds of diagrams are given in the main body of the text. 4. Scheil: Thermo-Calc includes a ScheilGulliver model for nonequilibrium solidification and a modified Scheil model that considers partial equilibrium for components that are selected by the user.

Applications of Thermo-Calc Thermo-Calc is used around the world within academia, in government research laboratories, and by commercial industry. The software can

be used to perform calculations for most applications involving phase equilibria, meta-stable equilibria, phase transformations, phase diagrams, and various thermodynamic properties, as well as critical assessments and data evaluations for multicomponent systems. While many types of calculations can be made using Thermo-Calc, the software typically is used to predict: • Stable and meta-stable phase equilibria for binary, ternary, and higher-order systems (calculations for alloy compositions with 6, 10, 15 elements are not uncommon, as illustrated by some of the examples in the main body of this book). • Amounts of phases (mass, volume and mole fractions) formed (phase balance) as a function of temperature, pressure, and composition and also the chemical compositions of the phases formed • Phase transformation temperatures such as liquidus, solidus, and solvus temperatures. Phase transformation temperatures can be predicted based on the actual chemistry (not nominal chemistry). • Thermochemical data such as enthalpies, heat capacity, and activities • Driving forces for precipitation • Phase diagrams (isothermal and isoplethal sections for multicomponent, multiphase systems as illustrated in this book) • Molar volume, density, and thermal expansion • Scheil-Gulliver (nonequilibrium) solidification simulations Thermo-Calc is not restricted just to modeling the alloy. Complex systems representing processing, for example, can also be considered. For example, another application is to calculate the carbon potential of multicomponent gas phase systems as a function of composition, temperature, and pressure and then predict what phases an alloy might form at a given temperature when exposed to such a carbon potential. Thermo-Calc can thus be applied to a number of practical problems related to metallurgy, processing, in-service performance, etc. as summarized by: • Alloy Design: Modification of alloy chemistries to improve properties or reduce costs using calculations to guide which compositions may be most suitable before preparing them for testing

Appendix 3: Introduction to Thermo-Calc / 283

• Heat Treatment: Prediction of formation of problematic phases prior to thermal processing • Casting: Calculation of liquidus and solidus temperatures; calculation of thermodynamic properties of the alloy for input into casting modeling codes • Welding and Joining: Prediction of the phases formed at the joining of two dissimilar materials or the interaction with filler material • Quality Control: Investigation of properties and phase balance within designated compositional tolerances More examples are available in the literature (search on key terms Thermo-Calc or CALPHAD). A list of published articles citing Thermo-Calc is available at www.thermocalc. com.

How to Obtain a Free Demonstration Version of Thermo-Calc Thermo-Calc is available in two formats: Thermo-Calc Classic, which has a command line interface and can be run under a number of different operating systems (including Microsoft Windows and Linux/Unix), and Thermo-Calc for Windows, which has an easyto-learn graphical user interface but only operates in the Microsoft Windows environment. Demo versions are available for both of these versions of the software. The demo versions are free to use, subject to the terms outlined in the Thermo-Calc Software End User License Agreement. It should be noted that the demo versions are limited to using just three elements (whereas in the full product the current upper limit is 40 elements)

and are supplied with only certain small databases that are for demonstration purposes and the evaluation of the software only. A link to register and download the demonstration version of the software can be accessed via a link on the Thermo-Calc web site at www.thermocalc.com. All fields in the registration form should be completed before continuing to the download page, where further instructions regarding installation of the software will be provided. On installation of the software, additional documentation, including a Users Guide/Examples manual in the form of PDF files will also be installed. Technical support for the demo versions of the software is limited, but problems related to installation or general inquiries can be addressed by visiting www.thermocalc.com and linking to their support. The demo version will run for approximately 1 month on a single computer, and installation on a network system is not supported. If you wish to run the software after the demo license has expired, it can be downloaded again (i.e., obtaining a new demo license). REFERENCES

1. J.O. Andersson, T. Helander, L. Höglund, P.F. Shi, and B. Sundman, Thermo-Calc and DICTRA, Computational Tools for Materials Science, Calphad, Vol 26, 2002, p 273–312, 2002 2. N. Saunders and A.P. Miodownik, Pergamon Materials Series, CALPHAD (Calculation of Phase Diagrams): A Comprehensive Guide, 1, Elsevier, 1998 3. TCFE5—TCS Steel/Fe-Alloys Database, Version 5.0, 2007, Thermo-Calc Software AB, www.thermocalc.com

Copyright © 2008 ASM International®. All rights reserved. Stainless Steels for Design Engineers (#05231G)

www.asminternational.org

Index A acetic acid corrosion rates for various alloys of, plus formic acid, 103(F) duplex alloys, 102 isocorrosion curves in, 36(F) isocorrosion performances of various alloys, 102(F) acids, corrosion in, and bases, 31–36 adsorption-induced brittleness, 51, 53 adsorption-induced plasticity, 51, 52–53 aeration cells, differential, 38–39 aerobic bacteria, influencing corrosion, 55 aesthetic finishes Chrysler Building in New York City, 196, 197(F), 213(F) considerations, 217–219 flatness, 219 surface, 217–219 aging treatments, precipitation hardening stainless steels, 168, 170 aggressive chemical agents, refinery applications, 255 Alloy Casting Institute (ACI) composition of, heat- and corrosion-resisting casting alloys, 277(T), 278(T) naming system, 147 alloy design avoiding unwanted phases, 94 duplex alloys, 92–94 Thermo-Calc software, 93, 282 alloying, sensitization, 47–48 alloying elements alloys, 1 influence on alloy families, 147, 149 influence on corrosion rate in contaminated sulfuric acid, 33 (F) influence on thermodynamic activity of C, N, S and O, 157(T) influence on uniform corrosion, 29–30 martensitic precipitation-hardening stainless steels, 141(T) martensitic stainless steels, 130, 131(T) alloy oxidation, behaviors, 66 alloys influence of, content on corrosion rate in hydrochloric acid, 35(F)

isocorrosion curves for, in sulfuric acid, 33(F) isocorrosion curves for, sulfuric acid with chlorides, 34(F) alloy systems alloying elements, 1 austenitic alloys, 69, 72 austenitic stainless family, 71(F) body-centered cubic (bcc) phase, 1 families in perspective, 69–72 most widely used, 1 Schaeffler–Delong stainless steels constitution diagram, 5(F), 70(F) alpha prime formation kinetics for duplex alloys, 94, 96(F) influence of, formation on hardness, 116(F) iron-chromium phase, 8 martensite, 7–8, 73 transition temperature change with, formation with aging, 98(F) alteration, surface, 199 aluminum inclusions and pitting, 40, 41 influence on thermodynamic activity of C, N, S and O, 157(T) metal migration into acetic solution from, 234(T) migration into acetic solution from stainless, aluminum or carbon steel, 234(T) oxidation resistance, 79, 226, 228 protective layer formation, 64–65 aluminum alloys, 1 aluminum oxide, 191 aluminum-titanium-nitride (AlTiN), 191 aluminum/titanium precipitates, possible, 138(F) American Iron and Steel Institute (AISI), 240 American Petroleum Institute (API), 135 American Society for Testing and Materials (ASTM), 238–239 American Society of Mechanical Engineers (ASME), 266 anisotropy deep drawing, 174 ferritic stainless steels, 120–121 Lankford ratio, 120, 175 stainless long products, 179 annealing austenitic stainless steels, 162–164 bright, 198

286 / Index

annealing (continued) deep drawing, 177 duplex stainless steels, 170–171 ferritic stainless steels, 165–166 long-term, of welds, 43 martensitic stainless steels, 166 precipitation hardening steels, 139 anode electrochemical reactions, 12 polarization, 20–21, 23 appliances facades, 239–240 kitchen, 237–240 laundry, 241 stainless steels commonly used for, 239(T) architecture and construction aesthetic considerations, 217–219 average chloride concentration in rainwater in United States, 217(F) balancing corrosion resistance, processing and economy, 214–215 balancing service environment, design and maintenance, 215, 217 cleaning methods for uncoated stainless steel, 220(T) concrete reinforcing bar, 222 corrosion resistance, 213–214 design considerations, 216(F) ecological considerations, 222 environment, 216(F) fabrication and joining, 221 fabrication considerations, 221 fire resistance, 221–222 flatness, 219 grades recommended by expert system, 217(F) graphic depicting low release of metal ions from 304 and 316 stainless steels, 222(F) local weather pattern, 216(F) maintenance, 220–221 maintenance schedule, 216(F) roof, 219(F) ranking common stainless steels by pitting resistance equivalent number (PREN), 214(T) repair, 221 rolled-on stainless steel finishes, 218(F) salt exposure, 216(F) special finishes, 218(F) stainless steel selection expert system, 216(F) surface finish aesthetics, 217–219 surface finish and corrosion resistance, 215 argon oxygen decarburization (AOD) adoption, 70 alloy adjustment, 157 automotive industry, 225 cleanliness, 184 control of nitrogen in refining by, 92 ferrite, 4 first commercial use, 109 foundry practice, 154 inclusions in steel, 40 production process, 155

atmospheres, oxidation, 66–67 atmospheric corrosion uniform corrosion, 36–37 atomic rearrangements, 2 attraction, interatomic, 2 austenite alloying elements, 5 carbide precipitation, 9 carbon and nitrogen, 6, 9 diffusion rates, 6 face-centered cubic (fcc), 5 γ-austenite in precipitation hardening alloys, 138–139 interstitial elements, 6 lattice expansions, 6(F) lean alloy of martensite and, 73–74 mechanical properties, 7 metastable state, 6 phase in duplex alloy at room temperature, 91 Schaeffler–Delong constitution diagram, 5(F), 70(F) Schaeffler diagram, 202(F) secondary type, 7 semiaustenitic precipitation-hardenable stainless steel, 143 sulfur and oxygen, 6 austenite conditioning, 169 austenitic-ferritic “C” alloys, 151–152 austenitic “H” alloys high temperature HE–HP, 152–154 precipitation hardening stainless steel, 170 austenitic precipitation-hardenable stainless steels cold work and aging, 146(F) composition, 145(T) corrosion resistance, 145–146 mechanical properties, 145 austenitic stainless steels alloy families, 69–72 annealing, 162–164 automotive structural components, 229 carburization, 82 composition of, 270(T), 271(T) composition of high-temperature, 82(T) compositions of commonly used lean, 72(T) compositions of corrosion-resistant, 86(T) corrosion resistance ratings, 87(T) corrosion-resistant alloys, 84–89 corrosive environments, 88–89 critical pitting temperature (CPT), 43, 44(F) drawability, 176 ductility, 180 face-centered cubic (fcc), 174 family, 71(F) forming limit diagram of carbon steel and, 176(F) halogens, 82 heat exchangers, 246 high-temperature alloys, 79–83 high-temperature mechanical properties, 82–83 impact strength variation with temperature, 75(F) intermetallic phases, 82, 203 isocorrosion curves for, in hydrochloric acid, 34(F) kitchen appliances, 239

Index / 287

lean alloys, 72–78 machinability, 185 machining setup recommendations, 183(T) martensite and austenite, 73–74 mechanical properties, 74–76, 82–83 mechanical properties after cold work and annealing, 163 metal migration into acetic solution from, 234(T) nitriding, 82 oxidation resistance, 79–81 petroleum industry applications, 249(T) pitting resistance equivalent number (PREN), 43, 78, 85 precipitation of carbides and nitrides, 76–78 recommended thermal processing temperatures, 162(T) resilience and toughness of carbon steel vs. for automotive components, 229(T) SCC (stress corrosion cracking), 49–50 secondary phases in, 82(T) sensitization, 46–47 soaking, 161–162 stabilization, 78 stainless steel in shipping, 245 SCC performance, 262 stress-strain curve for single crystals of stable, 53(F) surface finish, 89 tensile properties of carbon steel vs. for automotive components, 229(T) thermal processing, 161–164 water vapor, 81–82 weaknesses, 69 welding characteristics, 201–204 welding parameters, 207(T) austenitizing, martensitic stainless steels, 131, 132(F), 166–167 automotive and transportation alloy selection for exhaust systems, 226 alloys for major elements of automotive exhaust systems, 227(T) automotive emission standards, 225 bus bodies, 231(F) car manufacturers, 230 catalytic converter, 227(T), 228 center pipe, 227(T), 228 decorative to highly engineered applications in automobiles, 225 exhaust manifold and high-temperature, 227 exhaust systems, 225–228 ferritic stainless, 226 flexible pipe, 227(T), 228 front pipe, 227(T), 228 fuel tanks, 231 life-cycle cost calculation for stainless vs. carbon steel for bus, 231(T) microcar frame, 231, 232(F) muffler, 227(T), 228 rail transport, 232 resilience and toughness of carbon and stainless steels for automobiles, 229(T) stress-strain curves for 301 variants vs. duplex steels and transformation steel, 230(F) structural components, 229–231

tailpipe, 227(T), 228 tensile properties of carbon and stainless steels for automobiles, 229(T) trucks, 231

B bacteria influencing corrosion, 55 bacterial retention, food contact materials, 236(F) bases, corrosion in acids and, 31–36 basic oxygen furnace (BOF), 156 Bauschinger effect, 164 biocorrosion, 55–56 biological neutrality, food contact, 235 bleaching, pulp, 266 body-centered cubic (bcc) phase carbon and alloy steels, 1 change to face-centered cubic (fcc), 127(F) ferrite, 4, 110 ferritic material, 174 metals, 2 boron additions to ferritic stainless steels, 121 ferrite, 4 brazing, 211 bright annealing, 198 brightening stainless steels, 196 buffing, 197 built-up edge (BUE) austenitic stainless grades, 185 carbon and, 182 coolants minimizing, 191 copper and, 183 ferritic stainless steels, 185 grain sizes, 184 iron and tool, 182 nickel and, 183 nitrogen and, 184 precipitation hardening stainless steels, 185 bus life-cycle cost calculation for stainless vs. carbon steel, 231(T) stainless steel body, 230, 231(F) Butler–Volmer equation, 20–21

C calcium effect on machinability of 303, 188–189, 189(F) inclusions and pitting, 41 calcium-fluoride based slag electroslag remelting (ESR), 158 “C” alloys austenitic–ferritic alloys, 151–152 corrosion resisting, 147 duplex alloys, 151 mechanical properties of corrosion resisting cast, 150(T) metallurgy of, 149, 151–152 precipitation hardening, 151 carbide, precipitation kinetics, 7(F)

288 / Index

carbides duplex alloys, 94 flatware, 240 precipitation, 76–78 stainless steel, 9 tooling, 191 carbon alloying element, 1 austenite, 5 ferrite, 3–4 influence of alloying elements on thermodynamic activity of, 157(T) influence on thermodynamic activity of C, N, S and O, 157(T) influence on uniform corrosion, 29 interstitial atoms of, in austenite, 6 machinability of stainless steels, 182–183 precipitation of carbides, 76–78 precipitation rates by, content, 76(F) steel content, 156 variation of martensite hardness with, 128(F) welding of austenitic stainless steels, 201 carbon diffusion, 115 carbon dioxide, wet, 248 carbon solubility, austenitic stainless, 76(F) carbon steel activities and activity coefficients of elements in, 40(T) body-centered cubic (bcc) phase, 1 corrosion rates of stainless vs., 135(F) metal migration into acetic solution from, 234(T) resilience and toughness of, vs. stainless steel for automotive components, 229(T) tensile properties of, vs. stainless steel for automotive components, 229(T) carburization, 82 carburizing, 199 casting stainless steel processing, 158–159 Thermo-Calc software, 283 casting alloys Alloy Casting Institute (ACI), 147 austenitic-ferritic alloys, 151–152 austenitic HE–HP alloys, 152–154 chromium alloys, 148(T) chromium-nickel alloys, 148(T) composition of cast heat-resistant stainless and nickel base alloys, 149(T) compositions of cast stainless corrosion resisting alloys, 148(T) duplex alloys, 151 ferritic HA, HC, HD alloys, 152 foundry practice, 154 high-temperature mechanical properties of “H” alloys, 153(T) influence of alloying elements, 147, 149 mechanical properties of heat-resistant stainless, at room temperature, 152(T) metallurgy of “C” alloys, 149, 151–152 metallurgy of “H” alloys, 152–154 molten metal transfer, 153

naming system, 147 precipitation hardening, 151 room temperature mechanical properties of corrosion resisting stainless, 150(T) welding, 154 catalytic converter, 227(T), 228 catastrophic oxidation, 65 cathode effect of, polarization, 24(F) electrochemical reactions, 12 mass transfer limitations, 24 polarization, 20–21, 23 caustic solutions, 50 center pipe of exhaust systems, 227(T), 228 cerium inclusions and pitting, 41 oxidation resistance, 79–80 protective layer formation, 65 Charpy V toughness high-temperature austenitic alloys, 83(F) niobium-stabilized alloy, 119(F) titanium-stabilized alloy, 119(F) chemical agents, 255 chemical and process industry. See also corrosion types corrosion table for fuming sulfuric acid, 261(T) corrosion table for sulfuric acid, 259(T), 260(T), 261(T) corrosion types, 258–262 erosion, 262 forms of corrosion, 258 intergranular corrosion, 261–262 isocorrosion chart for sulfuric acid, 258(F) pitting and crevice corrosion, 258–259, 261 single- and dual-environment systems, 257–258 specific corrosives, 262 stress corrosion cracking (SCC), 257, 262 chemical neutrality, food contact, 233–235 chemistry, machinability of stainless steels, 182–184 chi intermetallic phase, 9 precipitation kinetics, 7(F) chip breaking, 187(F) chloride concentration in rainwater, 217(F) chloride-containing solutions, 50–51 chloride ion, aggressive against stainless steel, 85 chlorinated oils or waxes, 179(T) chromium alloying element, 1 austenitic alloys, 72 austenitic stainless steels, 85–86 chromium-oxygen system volatility vs. temperature and oxygen pressure, 64(F) corrosion resistance, 228 depletion from austenite near grain boundaries, 77(F) ferrite, 3 ferritic stainless alloys with low, medium and high, 110 inclusions and pitting, 42, 43 influence on thermodynamic activity of C, N, S and O, 157(T) influence on uniform corrosion, 29, 30(F) ion release from stainless steel grades, 222(F)

Index / 289

machinability of stainless steels, 182 migration into acetic solution from stainless, aluminum or carbon steel, 234(T) oxidation resistance, 71, 79, 80(F), 228 paralinear oxidation from evaporation of chromium superoxide, 64(F) Pourbaix diagram, 17(F) sulfide formation, 186 thermodynamics of oxidation, 57–59 volatile nature of Cr2O3, 63, 64(F) chromium alloys, cast stainless, 148(T) chromium-nickel alloys, 148(T) Chrysler Building architecture using stainless steel, 213 highly polished surface, 196, 197(F), 213(F) cleaning passivation, 195–196 recommended methods, 195(T) stainless steel, 194–196 cleaning methods stainless steels, 235–236 uncoated stainless steel, 220(T) cleanliness food contact materials, 235–236 machinability of stainless steels, 184 coastal climates, 237–238 coatings cookware, 237 tooling, 191 cold heading, 179–180 cold work, 75 coloring of stainless steels, 196 commercial use applications, 237–242 cookware, 237 flatware and cutlery, 240–241 food contact, 233–237 heating and water heating, 241–242 kitchen appliances, 237–240 laundry appliances, 241 stainless steel, 233 composition Alloy Casting Institute (ACI) heat- and corrosionresisting casting alloys, 277(T), 278(T) austenitic precipitation-hardenable (PH) stainless steel, 145(T) austenitic stainless steels, 270(T), 271(T) duplex alloys commercially available, 97 (T) duplex stainless steels, 276(T) ferrite stainless steels, 111(T), 112(T), 273(T), 274(T) martensitic PH stainless steels, 140(T) martensitic stainless steels, 124(T), 125(T), 275(T) PH stainless steels, 276(T) semiaustenitic PH stainless steel, 143(T) tool and cutlery martensitic stainless steels, 134(T) concrete reinforcing bar, 222 constitution diagram Schaeffler–Delong stainless steels, 5(F), 70(F) Schaeffler diagram, 202(F) Welding Research Council’s 1992, 203(F)

construction. See architecture and construction contamination, 235 continuous slab casting, 158 cookware, 237 coolants, 191 copper acid resistance, 71 machinability of stainless steels, 183 copper sulfate. See also sulfuric acid plus copper sulfate corrosion of stainless steel and titanium in, plus sulfuric acid, 31(F), 32(F) corrosion combating, in alloys for petroleum industry, 248, 250 definition, 11 erosion, 262 intergranular, 261–262 isocorrosion chart for sulfuric acid, 258(F) pitting and crevice, 258–259, 261 pitting resistance equivalent number (PREN), 258 single- and dual-environment systems, 257–258 stress corrosion cracking (SCC), 257, 262 table for sulfuric acid, 259(T), 260(T), 261(T) tendency, 15–16 types, 258–262 corrosion cost, 252–253 corrosion kinetics Butler–Volmer equation, 20–21 introduction, 19–20 mass transfer control, 21 migration and ionic diffusion, 21–22 mixed potential theory and polarization diagrams, 22–23 passivation, 23–25 Tafel regime: electrode-kinetics control, 21 corrosion rate alloys in simulated evaporator liquid, 37(F) influence of alloying element on, in contaminated sulfuric acid, 33(F) stainless oil country tubular goods, 135(F) stainless vs. carbon steel, 135(F) Tafel slope, 23(F) vs. surface roughness, 215(F) corrosion resistance architecture, 213–214 austenitic precipitation-hardenable (PH) stainless steel, 145–146 balancing, processing and economy, 214–215 duplex alloys, 99–106, 204 ferritic stainless steels, 109, 110, 121–122 function of salinity and temperature, 244(F) martensitic PH stainless steels, 141–142 material selection for desalination, 244–245 pulp-and-paper industry, 265–267 rail transport applications, 232 ratings of austenitic stainless steels, 87(T) semiaustenitic PH stainless steel, 144 stainless steel for refinery equipment, 254–255 stainless steel in shipping, 245 sulfur hurting, 188 surface finish and, 215

290 / Index

corrosion resisting alloys austenitic stainless steels, 84–89 “C” alloys, 149, 151–152 composition of, austenitic stainless steels, 86(T) composition of Alloy Casting Institute (ACI), 277(T), 278(T) compositions of cast stainless, 148(T) duplex alloys, 91 mechanical properties of stainless, 150(T) corrosion theory corrosion tendency, 15–16 electrochemical reactions, 11–12 Faraday’s law, 12 galvanic vs. electrochemical cells, 14 Nernst equation, 12–14 Pourbaix diagrams, 16–17 standard half-cell reduction potentials vs. normal hydrogen electrode, 14(T) corrosion types atmospheric, 36–37 biocorrosion and microbiologically induced, 55–56 chromium influence, 29 corrosion fatigue, 55 corrosion in acids and bases, 31–36 corrosion with fatigue or fraction, 48–55 crack initiation, 48–49 crack propagation, 49, 52(F) crevice, 38–39, 45–46 critical current density, 29, 30(F) dissimilar metals and differential aeration cells, 38–39 environmental variables, 50–51 environmental variables influencing uniform corrosion, 28–29 grain boundary, 46–48 hydrochloric acid, 33–34, 35(F) hydrogen embrittlement, 54–55 influence of alloying elements, 29, 30(F) localized, 37–38 material variables, 29–31, 49–50 molybdenum role, 29–30 nickel, 30 nitric acid, 34, 35(F) nitrogen, 30 organic acids, 35, 36(F) phosphoric acid, 34–35, 36(F) pitting, 39–45 pitting resistance, 43–45 preventing crevice, 45–46 SCC (stress corrosion cracking), 48–54 SCC mechanisms, 51–54 sensitization, 46–48 sodium chloride/carbon dioxide environment, 30 strong bases, 35 sulfuric acid, 31–33 sulfuric acid plus copper sulfate, 31(F), 32(F) uniform, 27–37 corrosive environments austenitic stainless steels, 88–89 platforms, 254 refinery equipment, 254–255

creep rupture strength, 83, 84(F) creep strength, 83(F) crevice corrosion austenitic stainless steels, 85, 87 corrosion type, 214, 258–259, 261 critical, temperature with alloy content, 45(F) critical crevice temperature (CCT) and critical pitting temperature (CPT), 105(F) dissimilar metals and differential aeration cells, 38–39 duplex alloys, 103–104 geometry, 45 preventing, 45–46 critical crevice temperature (CCT), 105(F) critical current density, 29, 30(F) critical pitting temperature (CPT) austenitic steels, 43, 44(F) critical crevice temperature (CCT) and CPT, 105(F) duplex alloys, 103, 104(F), 105(F) stainless steels for unwelded and welded material, 44(F) vs. pitting resistance equivalent number (PREN), 85(F), 104(F) cryogenic containers, 245 current density, 23(F) cutlery flatware and, 240–241 martensitic stainless steels, 133–134, 240 stainless steels commonly used for, 241(T) cutting tools, 133–134

D deep drawing anisotropy, 174 forming stainless steel, 173, 174–179 geometry, 174–175 hydroforming, 177–178 intermediate annealing, 177 materials composition, 175(T) schematic, 174(F) strain rate, 177 texture, 174 tooling, 176, 178–179 defects, 160 delignification of pulp, 266 demand for steel, 247 desalination materials selection for, 244–245 multi-stage flash (MSF), 243 reverse osmosis (RO), 243–244 technology, 243–244 design, balance, 215, 217 designers car manufacturers, 230 pitting corrosion, 39 development precipitation–hardening stainless steels, 137–138 welding, 211–212

Index / 291

differential aeration cells active alloys, 38–39 microfouling, 56 schematic, 20(F) differential aeration corrosion cell, 12 diffusion atomic rearrangements, 2 ionic transport, 21–22 diffusion rates, austenite vs. ferrite, 6 digesters first kraft from alloy 2205, 266(F) pulp-and-paper industry, 265 disinfection, 236–237 dissimilar metals, 38–39 dissociated ammonia, 198–199 dissolution equation, 37 term, 27 dryers, laundry appliances, 241 dry film, forming stainless, 179(T) dual-environment system, 257 duplex alloys acetic acid, 102 annealing, 170–171 composition of selected, stainless steels, 276(T) compositions, 97(T) concept, 91–92 corrosion resistance, 99–106 corrosion-resistant “C” alloys, 151 crevice corrosion, 103–104 deep drawing, 178 fastest-growing stainless steel family, 91–92 fatigue, 98 Fe-Cr-Ni phase diagrams, 92(F) formation kinetics, 96(F) formic acid, 103(F) forming and machining, 99 heat exchangers, 246 hot forming, 180 hydrochloric acid, 100–101 impact strength, 97–98 impact strength variation with temperature, 75(F) iron-nickel diagrams, 93(F) machinability, 186 machining setup recommendations, 183(T) mechanical properties, 94–98 nitric acid, 101 organic acids, 102, 103(F) partitioning of elements, 93–94 petroleum industry applications, 248(T) phosphoric acid, 101–102 photomicrographs, 95(F) pitting corrosion, 102–103 pitting resistance equivalent number (PREN), 43 PREN influencing fatigue, 98, 99(F) pulp-and-paper industry, 265, 266–267 recommended annealing and stress–relieving temperatures, 170(T) SCC (stress corrosion cracking), 49, 104–106 sensitization, 47

soaking, 170 sodium hydroxide, 101 stainless steel for line pipe, 253 stainless steel in shipping, 245 strength, 96 stress corrosion cracking (SCC) performance, 262 stress-strain curves for 301 variants vs., 230(F) structure and alloy design, 92–94 sulfuric acid, 100 thermal processing, 170–171 umbilical tubing and risers, 253–254 variations of ferrite, austenite, and duplex with temperature, 98(F) welding characteristics, 204–205 welding parameters, 207(T) wrought 2205 duplex microstructure, 91(F)

E earing deep drawing, 178 measuring tendency, 178 ecological considerations, 222 electrochemical cell closed circuit, 11–12 potential, 38 electrochemical corrosion, 19 electrochemical reactions, 11–12 electrode-kinetics control Tafel regime, 21 electrodes, polarization, 20–21 electrolysis cell, 14(F) electrolyte resistance, 22 electrolytic cells, galvanic vs., 14 electrolytic pickling, cold-rolled stainless, 194 electromotive force, 13 electropolishing, 196 electroslag remelting (ESR), 157–158 embrittlement. See also hydrogen embrittlement (HE) alpha prime, 8 high-temperature, 114 σ phase at higher temperatures, 151 engineering alloys minimum mechanical properties of stainless steel, 280(T) physical properties of major stainless steel, 279(T) environment, stainless steel selection expert system, 216(F) environmental variables stress corrosion cracking, 50–51 uniform corrosion, 28–29 epsilon martensite, 7–8, 73–74 equilibrium, argon oxygen decarburization (AOD), 155–156 equivalent weight (EW), 19 erosion, corrosion, 258, 262 expert system recommended stainless steel grades, 217(F) stainless steel selection, 216(F) exposure to salt, stainless steel selection expert system, 216(F)

292 / Index

F fabrication, 221 facades of appliances, 239–240 face-centered cubic (fcc) phase aluminum alloys, 1 austenite, 5 austenitic materials, 174 change to body-centered cubic (bcc), 127(F) metals, 2 Faraday’s law, 12, 21 fatigue corrosion, 55 duplex alloys, 98, 99(F) fatty oils and blends, suitability in forming stainless steel, 179(T) ferrite carbon and nitrogen, 3–4 carbon diffusion rate, 115 chromium, 3 diffusion rates in austenite vs., 6 δ-ferrite in precipitation hardening alloys, 138–139 hydrogen and boron, 4 mechanical properties, 4 molybdenum, 4 phase diagram of iron chromium, 3(F) phase in duplex alloy at room temperature, 91 Schaeffler–Delong constitution diagram, 5(F), 70(F) Schaeffler diagram, 202(F) stabilization with titanium, 4 thermal conductivity and thermal expansion, 4–5 ferritic “H” alloys, 152 ferritic stainless steels alpha prime formation, 116, 117(F) annealing, 165–166 automotive exhaust systems, 226 body-centered cubic (bcc), 174 carbon diffusion rate in, 115 composition, 111(T), 112(T), 273 (T), 274 (T) corrosion and oxidation resistance, 109, 121–122 deep drawing, 178 embrittling phenomenon, 116 forming limit diagrams, 176 (F) groups of low, medium and high chromium, 110, 113 heat exchangers, 245–246 high-temperature properties, 121 hot rolling, 159 impact strength variation with temperature, 75(F) intermetallic phases, 116 iron-chromium phase diagrams, 113(F), 114(F) kitchen appliances, 239 lowest cost and simplest stainless, 109–110 machinability, 185 machining setup recommendations, 183 (T) mechanical behavior, 116–117 metallurgy, 113–116 metal migration into acetic solution from, 234(T) petroleum industry applications, 247(T) pitting resistance equivalent number (PREN), 43 recommended annealing temperatures, 165(T)

sensitization, 47 soaking, 165 stabilization, 109, 115, 118–120 stress corrosion cracking (SCC), 49 stress relieving, 166 superferritics, 113 texture and anisotropy, 120–121 time-temperature-transformation (TTT) curve for 430, 115(F) titanium and niobium, 118 titanium for carbide and nitride formation, 115 toughness, 116(F), 117(F), 118–119 welding characteristics, 205–206 welding parameters, 207(T) ferromagnetism, 5 fingerprints, cleaning methods for uncoated stainless, 220(T) fire resistance, stainless steel, 221–222 flatness, surface aesthetic, 219 flatware, 240–241 flexible pipe, 227(T), 228 flow lines, 252–253 flux cored wire (FCW) welding, 210 food contact bacterial retention by material and cleaning time, 236(F) biological neutrality, 235 chemical neutrality, 233–235 cookware, 237 flatware and cutlery, 240–241 heating and water heating, 241–242 kitchen appliances, 237–240 material cleanliness, 235–236 metal migration into acetic solution, 234(T) qualifications, 233–237 stainless steels commonly used for appliances, 239(T) stainless steels commonly used for cutlery, 241(T) surface disinfection, 236–237 formability, ferritic stainless steel, 120 formic acid austenitic stainless steels, 89 corrosion in, 35 corrosion rates for various alloys of acetic plus formic acid, 103(F) duplex alloys, 102 isocorrosion curves in, 36(F) forming limit diagram (FLD), 175–176 forming technology deep drawing, 173, 174–179 deep drawing materials composition, 175(T) deep drawing schematic, 174(F) duplex alloys, 99 duplex stainless steel, 178 ferritics, 120, 178 flat, rolled stainless steel, 173–179 forces for hot working, 180(F) forming limit diagram of carbon steel vs. austenitic stainless steel, 176(F) forming limit diagrams for stainless steel categories, 176(F) hot, of stainless steel, 180

Index / 293

hydroforming, 177–178 hydrogen embrittlement, 177 limiting drawing ratio (LDR) vs. Lankford ratio, 175(F) optimized 409 for forming vs. normal 409, 177(F) orange peel, 178 stainless long products, 179–180 stainless steel, 173 stretch forming, 174 suitability of lubricants for use in, 179(T) surface finish, 178 tooling, 176, 178–179 foundary practice, casting alloys, 154 free energy, phases, 2 friction stir welding, 212 front pipe, alloys in automotive exhaust systems, 227(T), 228 fuel tanks, 231 fuming sulfuric acid, 261(T) furnace, stainless steels in, 241–242

G Gallionella, 55 galvanic cell electrochemical reaction, 38 schematic, 14(F) vs. electrolytic cells, 14 gas metal arc welding (GMAW) joint design, 209(F) process, 210 gas tungsten arc welding (GTAW) joint design, 209(F) process, 208–210 geometry crevice corrosion, 45 deep drawing, 174–175 pitting corrosion, 39 Gibbs free energy electrochemical reactions, 12–13 oxidation, 57, 58(F) grade selection, corrosion resistance, processing and economy, 214–215 graffiti, cleaning methods for uncoated stainless, 220(T) grain boundaries austenite, 6 austenite, of martensite, 8 boron additions to ferritics, 121 carbide precipitation, 9, 76–77, 77(F) corrosion, 46–48 defects in stainless steel, 160 depletion of chromium from austenite near, 77(F) ferrite-austenite, 8 grain size austenitic stainless steel annealing, 163 martensitic stainless steels and toughness, 131, 132(F) material structure, 184 graphite, suitability in forming stainless steel, 179(T) grinding, coarse polishing, 197

grit sizes, 197(T) Guinier–Preston (GP) zones, 138

H half-cell reactions reduction potentials, 13–14 vs. normal hydrogen electrode, 14(T) “H” alloys austenitic HE–HP alloys, 152–154 corrosion resisting, 147 ferritic HA, HC, HD, 152 high-temperature mechanical properties of, 153(T) mechanical properties of heat-resistant, 152(T) metallurgy, 152–154 halogens, 82 hardening austenitic stainless steels, 75 ferritic stainless steels, 116–117 heat-affected zone (HAZ) austenitic stainless steel, 207 chromium carbide formation in, 201 duplex stainless steels, 204 ferritic stainless steels, 109, 205 laser welding, 210 martensitic stainless steels, 206 secondary austenite, 7 heat exchangers, 245–246 heat-resistant alloys composition of Alloy Casting Institute (ACI), 278(T) compositions, 149(T) “H” alloys, 152–154 mechanical properties of cast stainless, 152(T) heat tint, coloring stainless steels, 196 heat tinting, cleaning method, 220(T) heat treatment, Thermo-Calc software, 283 heat treatment and conditioning, 168–170 heavy-duty emulsions, 179(T) heavy metals, elimination, 157 high-frequency induction welding, 211 high-speed tool steels, 190–191 high-temperature alloys austenitic stainless steel, 82(T) intermetallic phases of austenitic stainless steel, 82 martensitic stainless steels, 133, 134(F) mechanical properties of austenitic, 82–83, 84(F) oxidation resistance of austenitic, 79–81 water vapor, 81–82 high-temperature embrittlement, 114 high-temperature properties, ferritic stainless steels, 121 hopper cars, 232 hot ductility defects, 160 hot forming, 180 hot mill defects, 160 hot rolling, 159–160 hot Steckel mills, 159 hot strip tandem mills, 159 hydrochloric acid austenitic stainless steels, 88, 89(F) corrosion in, 33–34

294 / Index

hydrochloric acid (continued) duplex alloys, 100–101 influence of alloy content on corrosion rate in, 35(F) isocorrosion curves for austenitic stainless steels in, 34(F) isocorrosion curves for stainless steels in, 34(F) isocorrosion performance of duplex, 101(F) hydrofluoric acid, 193–194 hydroforming, 177–178 hydrogen/argon atmosphere, bright annealing, 198–199 hydrogen embrittlement (HE) corrosion fatigue, 55 crack growth, 49 ferritic stainless steels, 121–122 mechanisms, 54–55 stress corrosion cracking, 51, 52–54 hydrogen ion reduction, 15(F), 28(F) hypochlorite bleaches, 195

I impact strength duplex alloys, 97–98, 204 variation with temperature for stainless steels, 75(F) inclusion-related defects, 160 inclusions chip breaking at sulfides, 187(F) lead, selenium, tellurium, 186 oxides, 188–190 pitting corrosion, 40–43 role in machining stainless steels, 186–190 stainless steel, 10 sulfur, 186–188 induction welding, high-frequency, 211 ingot method, 158, 159 initiation pitting, 39–40, 43 stress corrosion cracking, 48–49 interatomic attraction, thermodynamics, 2 intermetallic phases austenitic stainless steel, 82, 203 ferritic stainless steels, 116 stainless steel, 8–9 International Nickel Company (INCO) process, 196 interstitial elements, 6 ionic current, 11–12 ionic diffusion, 21–22 iron body- and face-centered cubic transformations, 2 electrochemical corrosion, 19 ferrite, 3 ion release from stainless steel grades, 222(F) machinability of stainless steels, 182 migration into acetic solution from stainless, aluminum or carbon steel, 234(T) penetration rates, 19(T) Pourbaix diagram, 16(F) pseudo-binary-phase diagram for, and sulfur, 41(F) iron-chromium phase diagram, 3(F) phase diagram from Thermocalc, 113(F)

phase diagrams with varying carbon, 114(F), 130(F) phase diagrams with varying chromium, 130(F) iron dissolution, 20(F) iron reduction, 15(F), 28(F)

J JFI Steel, oil country tubular goods and line pipe alloys, 135(T) joining stainless steel, 221 Thermo-Calc software, 283 joint design, 208, 209(F)

K kinetics, alpha prime formation, 94, 96(F) kitchen appliances austenitic stainless steel, 239 coastal conditions, 238 exposure of stainless samples to North Carolina beach, 238(F) facades, 239–240 ferritic stainless grades, 239 food contact, 237–240 interior or working parts, 239 knife-line attack, 48, 202 kraft process paper-making, 265–267 pulp-and-paper industry, 36, 265

L Lankford r, earing tendency, 178 Lankford ratio anisotropy measure, 120, 175 limiting drawing ratio as function of, 175 lanthanum, protective layer, 65 laser welding, 210 lattice expansions, 6(F) laundry appliances, 241 laves, precipitation kinetics, 7(F) laves phase, 9 leaching, elements from stainless to foods, 234 lead, stainless steel machinability, 186 lean alloys austenitic, 72–78 compositions of austenitic, 72(T) martensite and austenite, 73–74 lime content, electroslag remelting, 158 limiting current electrode reaction kinetics, 21 increasing mass transfer, 23, 24(F) limiting drawing ratio, 175 line pipe martensitic stainless, 134–135 stainless steel application, 252–253 liquefied natural gas (LNG) vessels, 254 localized corrosion, 37–38

Index / 295

lubricants oxides, 189–190 suitability in forming stainless steel, 179(T) lubrication, 197–198

M machinability, material’s, 181(F) machining stainless steels austenitic, 185 carbides, 191 carbon, 182–183 chromium, 182 cleanliness, 184 coatings, 191 coolants, 191 copper, 183 cross-section size, 185 duplex, 186 duplex alloys, 99 ferritic, 185 high-speed tool steels, 190–191 introduction, 181–182 iron, 182 lead inclusions, 186 machinability of stainless steel families, 185–186 manganese, 183 martensitic, 185 material’s machinability, 181(F) molybdenum, 183 nickel, 183 niobium, 184 nitrogen, 184 oxide inclusions, 188–190 physical and mechanical properties, 182–185 precipitation hardening, 185–186 process, 184 role of chemistry, 182–184 role of inclusions, 186–190 selenium inclusions, 186 setup recommendations for turning wrought stainless steels, 183(T) structure, 184 sulfur, 183 sulfur inclusions, 186–188 super stainless steels, 186 tellurium inclusions, 186 titanium, 184 tooling and coolants, 190–191 maintenance balancing service, design and, 215, 217 stainless steel, 220–221 stainless steel selection expert system, 216(F) manganese alloying element, 1 austenite, 5 inclusions and pitting, 41 influence on thermodynamic activity of C, N, S and O, 157(T)

machinability of stainless steels, 183 sulfide formation, 186 manganese sulfides inclusions, 41 stress risers, 188 x-ray examination, 189(F) Marangoni effect, 207 marine systems corrosion resistance vs. salinity and temperature, 244(F) desalination, 243–245 heat exchangers, 245–246 materials for desalination, 244–245 shipping, 245 typical analyses and properties of marine alloys, 245(T) martensite carbon and nitrogen, 9 composition range, 7 formation, 126–127 forms, 7–8, 73–74 lattice expansions, 6(F) lean alloy of, and austenite, 73–74 platelets from surface, 126(F) reversion of, formed by cold work, 75(F) Schaeffler–Delong constitution diagram, 5(F), 70(F) Schaeffler diagram, 202(F) tempering, 7 varying hardness with carbon content, 128(F) martensitic alloys “C” alloys, 149, 151 composition of, precipitation hardening (PH) alloys, 140(T) corrosion resistance of, PH alloys, 141–142 mechanical properties of, PH alloys, 139(T) microstructures of, PH alloys, 140(F) precipitation hardening stainless steels, 139–142 martensitic stainless steels annealing, 166 applications, 133–135 austenitizing, 166–167 composition, 124(T), 125(T) composition of, 275(T) compositions of tool and cutlery, 134(T) corrosion rates of stainless oil country tubular goods (OCTG) alloys, 135(F) corrosion resistance, 123 distinction from other alloys, 123, 126 expanding austenite stability range with nickel, 131(F) flatware and cutlery, 240–241 hardness variation with carbon content, 127, 128(F) high-temperature use, 133, 134(F) hot rolling, 159 influence of alloying elements, 130, 131(T) iron-chromium phase diagrams, 130(F) machinability, 185 machining setup recommendations, 183(T) martensite formation, 126–127 OCTG and line pipe, 134–135 passivation, 195 petroleum industry applications, 247(T) phase structure, 127–128, 130–131

296 / Index

martensitic stainless steels (continued) photomicrographs, 129(F) recommended annealing, austenitizing, and tempering temperatures, 166(T) sensitization, 47 smallest stainless steel category, 123 soaking, 166 strain energy, 126(F), 127(F) stress relieving, 167–168 tempering, 167 tempering and toughness, 132, 133(F) thermal processing, 131–133, 166–168 tool and cutlery alloys, 133–134 toughness by austenite grain size and phosphorus, 132(F) welding characteristics, 206 welding parameters, 207(T) mass transfer, 23, 24(F) mass transfer control, 21 mass transport, 24–25 material selection, welding, 206–208 material structure, 184 material variables stress corrosion cracking (SCC), 49–50 uniform corrosion, 29–31 mechanical behavior, ferritic stainless steels, 116–117 mechanical properties austenite, 7 austenitic precipitation-hardenable (PH) stainless steel, 145 corrosion resisting cast stainless alloys, 150(T) deep-drawing stainless steels, 175(T) duplex alloys, 94–98 ferrite, 4 heat-resistant cast stainless alloys, 152(T) high-temperature, of austenitic stainless steels, 82–83, 84(F) high-temperature, of “H” alloys, 153(T) lean austenitic alloys, 74–76 machinability of stainless steels, 182–185 marine alloys, 245(T) martensite, 8 martensitic PH stainless steels, 139(T) minimum, of stainless steel engineering alloys, 280(T) semiaustenitic PH stainless steel, 144(T) stainless steels, 10 mechanisms pitting corrosion, 39–40 precipitation-hardening, 138 stress corrosion cracking (SCC), 51–54 melting production process, 155–157 metal dusting, oxidation, 67 metal flow directions, 208(F) metal ions release, 222(F) metallurgy “C” alloys, 149, 151–152 ferrite stainless steels, 113–116 ferritic stainless, 226 “H” alloys, 152–154 introduction, 1–2

metal oxides parabolic rate constants for growth, 59(F) standard Gibbs free energy of, formation vs. temperature, 57, 58(F) metals oxidation, 57 with oxide scale, 61(F) metastable condition, 2 metastable pitting, 40 microbiologically induced corrosion, 55–56 microcar frame, 231, 232(F) microorganisms, food contact, 235 microstructures, martensitic precipitation-hardening stainless steels, 140(F) migration, ionic transport, 21–22 mineral resin, bacterial retention, 236(F) mischmetal, 41 mixed potential theory, 22–23 mold powder, continuous casting, 158 molybdenum alloying element, 1 austenitic alloys, 72 carbide precipitation, 9 corrosion resistance, 228 disulfide, 179(T) ferrite, 4 influence on resistance to stress corrosion cracking, 50(F) influence on thermodynamic activity of C, N, S and O, 157(T) influence on uniform corrosion, 29–30 influencing critical pitting temperature in welded vs. unwelded austenitic grade, 208(F) machinability of stainless steels, 183 oxidation resistance, 226, 228 protective layer formation, 64–65 muffler, automotive exhaust systems, 227(T), 228 multistage flash (MSF), 243

N naming system, Alloy Casting Institute (ACI), 147 National Association of Corrosion Engineers (NACE) alloy listing, 247 diagram showing alloy suitability, 250, 251(F) regulating high-strength alloys, 142 restrictions in use recommendations for stainless steels, 252(T) natural gas, vessels for liquefied, 254 Nernst equation open circuit potential, 20 thermodynamics of electrochemical reactions, 12–14 neutrality biological, in food contact, 235 chemical, in food contact, 233–235 New York City’s Chrysler Building, 196, 197(F), 213(F) nickel alloying element, 1 austenite, 5

Index / 297

austenitic alloys, 72 carbide precipitation, 9 corrosion rates for stainless steels and, base alloys, 80(F) influence on thermodynamic activity of C, N, S and O, 157(T) influence on uniform corrosion, 30 influencing oxidation of iron-chromium alloys, 79, 80(F) ion release from stainless steel grades, 222(F) machinability of stainless steels, 183 migration into acetic solution from stainless, aluminum or carbon steel, 234(T) resistance to stress corrosion cracking, 50(F) nickel base alloys, 149(T) niobium carbide former, 78 creep resistance, 71–72 high-temperature martensitic stainless, 133, 134(F) machinability of stainless steels, 184 replacing titanium, 205 role in sensitization, 47 stabilization, 226 stabilization of ferritic stainless steels, 118–119 Nippon Steel, oil country tubular goods and line pipe alloys, 135(T) nitric acid austenitic stainless steels, 88 corrosion behavior of high-silicon alloys in concentrated, 35(F) corrosion in, 34 duplex alloys, 101 isocorrosion curve for, 35(F) pickling oxide scale, 193–194 nitrides duplex alloys, 94 stainless steel, 9–10 nitriding austenitic stainless steel and, 82 surface alteration, 199 nitrogen austenite, 5 austenitic alloys, 72 austenitic stainless steels, 71, 86 delay in carbide precipitation by, 78(F) ferrite, 3–4 influence of alloying elements on thermodynamic activity of, 157(T) influence on thermodynamic activity of C, N, S and O, 157(T) influence on uniform corrosion, 29, 30 interstitial atoms of, in austenite, 6 machinability of stainless steels, 184 solubility in austenite, 77–78 stainless steel for line pipe, 253 nondestructive evaluation (NDE), 211 normal hydrogen electrode (NHE) half-cell reduction potential vs., 14(T) standard hydrogen electrode, 14 North Carolina, exposure of stainless steel, 238(F)

O Occupational Safety and Health Administration (OSHA), 211 Ohm’s law, 22 oil and grease marks, cleaning methods, 220(T) “oil-canning,” 219 oil country tubular goods (OCTG) influence of chromium on corrosion rate of steel, 249(F) influence of copper and nickel on corrosion rate of martensitic alloys, 249(F) martensitic stainless, 134–135 stainless steels in petroleum industry, 250–252 oleum, 33 open circuit potential, 13 orange peel, 178 organic acids austenitic stainless steels, 89 corrosion in, 35 duplex alloys, 102 isocorrosion curves in, 36(F) overpotentials, 20 oxidation effect of chromium, 57–59, 60(T) effect of rare earth additions, 65 effect of silicon, aluminum, and molybdenum, 64–65 electrochemical nature of, 60–61 influence of nickel on, of iron-chromium alloys, 79, 80(F) iron-chromium-oxygen phase diagram, 59(F) kinetics and, rates, 61–63 metal dusting, 67 metal with oxide scale, 61(F) oxidation-resisting grades of stainless steel, 60(T) parabolic rate constants for growth of oxides, 59(F) paralinear, from evaporation of chromium superoxide, 64(F) quasi-steady-state approximation of moving boundary problem of internal, 66(F) reaction at anode, 12 schematic predicting thermal stresses, 65(F) spalling and cracking of scale, 63–65 standard Gibbs free energy of formation of metal oxides vs. temperature, 58(F) temperature dependence of metal dusting of iron, 67(F) thermodynamics of, 57–60 transient, 60 under less-oxidizing atmospheres, 66–67 volatile nature of Cr2O3, 63, 64(F) Wagner’s theory, 61–63 oxidation resistance austenitic stainless steels, 79–81 ferritic stainless steels, 109, 121–122 isooxidation curves, 80(F) oxide film coloring stainless steels, 196 parameters for, coloring stainless steel, 196(T)

298 / Index

oxides effect calcium on machinability of 303, 188–189, 189(F) elongated, 190(F) inclusions, 10 inclusions and pitting, 41 machinability of stainless steel, 188–190 metal with, scale, 61(F) pickling to remove, layer, 25, 193–194 removal of oxide scale, 193–194 stabilization of austenitic alloy, 78 Ugima, in 303 matrix, 189(F) un-deformed, 190(F) x-ray examination showing Ugima, and manganese sulfides, 189(F) oxyfuel gas welding (OFW), 210 oxygen austenite impurity, 6 ferrite impurity, 3 impurity, 1 inclusions and pitting, 40, 41–42 influence of alloying elements on thermodynamic activity of, 157(T) influence on thermodynamic activity of C, N, S and O, 157(T) influence on uniform corrosion, 29 steel content, 156 oxygen gas reduction, 15(F), 28(F) oxygen pressure, 64(F)

P paint, cleaning method for uncoated stainless, 220(T) paper-making processes bleaching pulp, 266 digestion, 265–266 kraft process, 265 process equipment, 266–267 washing and screening, 266 partitioning elements, 93–94 passenger trains, 232 passivation effect on polarization diagrams, 23–25 removing surface contamination, 195–196 stainless steel, 25 theory, 23 transpassive regime, 24 passive behavior, 27 passivity, 27 penetration equation, 19 penetration rates, 19(T) petroleum industry alloy suitability vs. H2S and CO2 partial pressure, 251(F) austenitic stainless steels for, 249(T) chromium influence on corrosion rate in environments by oil country tubular goods (OCTG), 249(F) combating corrosion in applications, 248, 250 copper and nickel influencing corrosion rate of martensitic stainless alloys for OCTG, 249(F) demand for steel, 247 duplex stainless steels for, 248(T)

ferritic stainless steels for, 247(T) line pipe and flow lines, 252–253 liquefied natural gas (LNG) vessels, 254 martensitic stainless steels for, 247(T) molybdenum influence on SCC susceptibility, 248, 250(F) National Association of Corrosion Engineers (NACE), 247, 248, 250, 252(T) OCTG, 250–252 platforms, 254 precipitation-hardening stainless steels for, 248(T) presence of wet carbon dioxide, 248 refinery equipment, 254–255 restrictions in stainless steel use recommended by NACE, 252(T) stainless steels for refinery processes, 254(T) stress corrosion cracking (SCC), 248 umbilical tubing and risers, 253–254 pH, corrosion tendency, 15–16 phase diagrams computer models, 2 Fe-Cr-Ni, 92(F) iron-chromium, 3(F), 73(F), 113(F), 114(F), 130(F) iron-chromium-oxygen, 58, 59 (F) iron-nickel, 93(F) pseudo-binary-, for iron and sulfur, 41 (F) phases alloy systems, 1 ferrite, 3–5 free energy, 2 intermetallic, of stainless steel, 8–9 structure of martensitic stainless steels, 127–128, 130–131 phosphoric acid austenitic stainless steels, 88 corrosion in, 34–35 duplex alloys, 101–102 electropolishing solution, 196 isocorrosion curves in, 36(F) minimum temperatures for wet, with duplex alloys, 102(F) phosphorus ferrite impurity, 3 impurity, 1, 156–157 martensitic stainless steels and toughness, 131, 132(F) photomicrographs duplex alloys, 94, 95(F) martensitic stainless steels, 127, 129(F) physical properties major stainless steel engineering alloys, 279(T) stainless steels, 10 pickling oxide layer removal, 25, 193–194 uniform corrosion, 28 pigmented pastes, 179(T) Pilling–Bedworth ratio (PBR), 63 pitting activities and activity coefficients in liquid steels, 40(T) austenitic stainless steels, 85, 88 corrosion type, 39–40, 258–259, 261

Index / 299

critical, temperatures, 44(F) geometry, 39 inclusions, 40–43 influence of sulfur level on, resistance, 42(F) metastable, 40 mischmetal, 41 passive anode polarization curve, 40(F) pit initiation, 39–40 pseudo-binary-phase diagram for iron and sulfur, 41(F) resistance, 43–45 “weakest link” phenomenon, 214 pitting corrosion CPT (critical pitting temperature) vs. NaCl concentration, 103, 105(F) CPT vs. pH, 103, 105(F) CPT vs. pitting resistance equivalent number (PREN), 103, 104(F) duplex alloys, 102–103 varying pitting potential with temperature, 103, 104(F) pitting resistance equivalent number (PREN) austenitic alloys, 43, 78, 85 corrosion, 37, 258 critical pitting temperature vs., 85(F) duplex alloys, 43, 92, 102 ferritic alloys, 43 influence on duplex alloy fatigue strength, 98, 99(F) pitting corrosion, 43–45, 214 ranking stainless steels by PREN, 214(T) umbilical tubing and risers, 253 Pittsburgh Convention Center, 219(F) platforms, stainless steel, 254 polarization anode, 23 cathode, 23 influence on uniform corrosion, 29, 30(F) overpotentials, 20 passivating alloys, 39(F) passive anode, curve, 40(F) stainless steel in chloride-containing solution, 40 polarization diagrams effect of cathode polarization, 24(F) effect of mass transport, 24–25, 25(F) mixed potential theory and, 22–23 passivation, 23–25 schematic, 22(F) schematic of passive anode polarization curve, 23, 24(F) polishing grit sizes for target surface roughness, 197(T) surface finishing, 197–198 polycarbonate, bacterial retention, 236(F) polythionate, 50 polythionic acid, 255 porosity, 52 Porsche, auto components, 230 postweld stress relief, 205 potassium hydroxide, 88–89 Pourbaix diagrams chromium, 17(F) construction of, 16–17 iron, 16(F)

powder metallurgy, 159 precipitated phases Guinier–Preston zones, 138 stainless steels, 8(T) precipitation carbides, 9 carbides and nitrides, 76–78 possible aluminum/titanium, 138(F) precipitation-hardening stainless steels advantage over martensitic, 137 annealed condition, 139 austenitic, 144–146 austenitic alloys, 170 cast PH alloys, 151 cold work influence on aging of A–286, 145, 146(F) composition, 276(T) composition of austenitic, 145(T) composition of martensitic, 140(T) compositions of semiaustenitic, 143(T) corrosion resistance of martensitic, 141–142 corrosion resistance of semiaustenitic, 144 development, 137–138 influence of alloying elements, 141(T) machinability, 185–186 machining setup recommendations, 183(T) martensitic, 139–142 martensitic grades, 168–170 mechanical properties of martensitic PH alloys, 139(T) mechanical properties of semiaustenitic, 144(T) mechanism of PH, 138 microstructures, 140(F) passivation, 195 petroleum industry applications, 248(T) phases in stainless steel, 10 possible aluminum/titanium precipitates, 138(F) presence of δ-ferrite and γ-austenite, 138–139 processing routes for S15700, 142, 143(F) properties of A-286 vs. test temperature, 145(F) recommended annealing and stress-relieving temperatures for martensitic grades of, 169(T) semiaustenitic, 142–144 specialized family, 137 stress corrosion cracking (SCC), 141, 142 thermal processing, 168–170 welding characteristics, 206 welding parameters, 207(T) precipitation kinetics, 7(F) prevention, crevice corrosion, 45–46 production processes basic oxygen furnace (BOF), 156 casting, 158–159 defects, 160 electroslag remelting (ESR), 157–158 hot rolling, 159–160 hot Steckel mills, 159 hot strip tandem mills, 159 impurities, 156–157 influence of alloying elements on thermodynamics, 157(T) melting and refining, 155–157

300 / Index

production processes (continued) remelting, 157–158 semiaustenitic precipitation-hardenable stainless steel, 142, 143(F) stainless steel, 155 thermodynamics, 156 vacuum arc remelting (VAR), 157–158 vacuum induction melting (VIM), 157 vacuum oxygen decarburization (VOC), 156 propagation crack, rates of metals vs. current density, 52(F) stress corrosion cracking (SCC), 49 pulp-and-paper industry duplex stainless steels, 265, 266–267 kraft process, 36, 265 paper-making processes, 265–267 pulsed arc transfer, 210

Q quality control, Thermo-Calc software, 283

R rain, cleansing action of, 220 rainwater, average chloride concentration, 217(F) rare earth metals inclusions and pitting, 41 protective layer formation, 65 reduction, 12 reduction potential iron and hydrogen ion reductions vs. pH, 15(F), 28(F) iron and oxygen gas reductions vs. pH, 15(F), 28(F) reference electrode, 13–14 refinery equipment, 254–255 refining production process, 155–157 repair, 221 residential applications cookware, 237 domestic goods, 233 flatware and cutlery, 240–241 heating, 241–242 kitchen appliances, 237–240 laundry appliances, 241 water heaters, 241, 242 resistance Ohm’s law, 22 pitting, 43–45 resistance welding, 210–211 resistivity, 22(T) reverse osmosis, 243–244 ridging, ferritics, 178 rolled finishes applications, 198–199 benefits, 198 roping, ferritics, 178 rust staining, cleaning method for uncoated stainless, 220(T)

S Saab, auto components, 230 safety, welding, 211 salinity, corrosion resistance vs., 244(F) salt exposure, stainless steel selection expert system, 216(F) sanitation, cleaning stainless steels, 195 scale metal with oxide, 61(F) spalling and cracking of, 63–65 Schaeffler–Delong constitution diagram, 5(F), 70(F) Schaeffler diagram, 202–203 seawater. See also marine systems desalination, 243–245 secondary austenite, 7 secondary phases, 82(T) selenium, 186 semiaustenitic precipitation-hardenable stainless steels austenite-stabilizing elements, 142 compositions, 143(T) corrosion resistance, 144 mechanical properties, 144(T) processing by T route, 142, 143(F) sensitization austenitic, 46–47 duplex steels, 47 effect of alloying, 47–48 ferritic, 47 ferritic stainless steels, 115 heat treatment vs. time, 46(F) intergranular corrosion, 9 knife-line attack, 48, 202 martensitic steels, 47 schematic of, due to chromium-rich precipitates, 46(F) welding, 48 service, design, and maintenance, 215, 217 shielded metal arc welding (SMAW) joint design, 209(F) process, 210 shielding gas welding austenitic stainless steel, 203–204 welding parameters for various stainless steels, 207(T) shipping, 245 short-circuiting transfer, 210 sigma intermetallic phase, 8–9 precipitation kinetics, 7(F) silicon alloying element, 1 content in cast alloys, 147 corrosion of high- austenitic steels in nitric acid, 34, 35(F) inclusions and pitting, 41 influence on thermodynamic activity of C, N, S and O, 157(T) oxidation resistance, 71, 79, 226, 228 protective layer formation, 64–65 single-environment system. See also chemical and process industry aggressive chemical species, 258

Index / 301

slabs, 158 slip dissolution, 51 soaking austenitic stainless steels, 161–162 duplex stainless steels, 170 ferritic stainless steels, 165 martensitic stainless steels, 166 soap-fat pastes, 179(T) sodium chloride/carbon dioxide environment, 30 sodium hydroxide austenitic stainless steels, 88–89 corrosion in, 35–36 corrosion rates of duplex alloys, 101(F) corrosion rates of duplex alloys with contaminated environment, 101(F) duplex alloys, 101 sodium hypochlorite cleaning stainless steels, 195 disinfecting stainless, 236–237 software package, Thermo-Calc, 2, 281–283 soldering, 211 solution treatment, precipitation-hardening stainless steels, 168–170 specialization, precipitation-hardening stainless steels, 137 Sphaerotilus, 55 spots, cleaning method for uncoated stainless, 220(T) spray transfer, 210 stability expanding austenite, with nickel, 131(F) lean alloy of martensite and austenite, 73–74 stabilization ferrite, 4 ferritic stainless steel, 109, 115, 118–120 ferritic steels for exhaust systems, 226 lean austenitic alloys, 78 stacking fault, 73–74 stainless long products cold heading, 179–180 hot forming, 180 stainless steel alloys austenite, 5–7 ferrite, 3–5 stainless steels. See also casting alloys activities and activity coefficients of elements in, 40(T) bacterial retention by material and cleaning time, 236(F) carbides, 9 casting, 158–159 casting alloys, 147, 149 categories for oxidation resistance, 59–60 classifications by sulfur content, 187–188 cleaning methods for uncoated, 220(T) composition of austenitic, 270 (T), 271 (T) composition of duplex, 276(T) composition of ferrite, 273(T), 274(T) composition of martensitic, 275(T) composition of precipitation-hardenable (PH), 276(T) concrete reinforcing bar, 222 corrosion rates of, vs. carbon steel, 135(F)

corrosion table for, in sulfuric acid plus copper sulfate, 31(F) deep drawing, 173, 174–179 defects in, hot–rolled bands, 160 flat, rolled, 173–179 hot rolling, 159–160 inclusions, 10, 186–190 isocorrosion curves for, in sulfuric acid plus copper sulfate, 32(F) lubricants for forming, 179 (T) machinability, 185–186 machining setup recommendations, 183 (T) melting and refining, 155–157 minimum mechanical properties of, engineering alloys, 280 (T) nitrides, 9–10 oxidation-resisting grades, 60(T) passivation, 25 penetration rates, 19(T) physical properties of major, engineering alloys, 279 (T) precipitated phases, 8 (T) precipitation-hardening process, 10 precipitation kinetics in 316, 7 (F) properties, 10 ranking by pitting resistance equivalent number (PREN), 214(T) ranking common, by PREN, 214(T) refinery processes, 254(T) remelting, 157–158 resilience and toughness of carbon steel vs. for automotive components, 229(T) Schaeffler–Delong constitution diagram, 5 (F) selection expert system, 216(F) tensile properties of carbon steel vs. for automotive components, 229(T) thermodynamics, 2 welding parameters, 207(T) stains, cleaning method for uncoated stainless, 220(T) standard Gibbs free energy, 57, 58(F) standard hydrogen electrode, 14 Steckel mills, hot, 159 Steel Founder’s Society of America, 147 strain energy, martensitic stainless steels, 126(F), 127(F) strain rate, deep drawing, 177 stress corrosion cracking (SCC) advantages of duplex alloys, 105–106 austenitic stainless steels, 87–88 corrosion form, 258 crack initiation, 48–49 crack propagation, 49 crack propagation rates of metals vs. current density, 52(F) debating mechanisms, 105 dilation of austenite due to hydrogen in solution, 53, 54(F) duplex alloys, 91, 104–106 environmental variables, 50–51 ferritic stainless steels, 121 influence of molybdenum on resistance, 50(F) martensitic precipitation-hardening (PH) stainless steels, 141, 142

302 / Index

stress corrosion cracking (continued) material variables, 49–50 mechanisms, 51–54 petroleum industry, 248 resistance, 257, 262(F) stress-strain curve for single crystals of austenitic steel with and without hydrogen, 53(F) susceptibility of martensitic stainless steels, 123 susceptibility to, with oxygen and chloride content for 304 stainless, 51(F) theory, 262 threshold stress for, for various alloys, 88(F) varying resistance to, with nickel content, 50(F) water heaters, 242 zones of susceptibility, 48(F) stress relief annealing (SRA), 48 stress relieving austenitic stainless steels, 164 ferritic stainless steel, 166 martensitic stainless steels, 167–168 stress risers, 188 stress sorption, stress corrosion cracking, 51, 53 stretch forming, 174 strip casters, 158–159 strip tandem mills, 159 strong bases austenitic stainless steels, 88–89 corrosion in, 35–36 structure duplex alloys, 92–94 machinability of stainless steels, 184 phase, of martensitic stainless steels, 127–128, 130–131 submerged arc welding (SAW) joint design, 209(F) process, 210 submerged entry nozzle, 158 sulfate process, pulp-and-paper, 36 sulfides inclusions, 10 inclusions in stainless steel, 186–187 size and shape and machinability, 187 stabilization of austenitic alloy, 78 sulfite process, 265 sulfur austenite impurity, 6 comparing machinability, 190(F) effect on stainless machinability, 187, 188(F) ferrite impurity, 3 impurity, 1, 156 inclusions and pitting, 40, 41 influence of alloying elements on thermodynamic activity of, 157(T) influence on thermodynamic activity of C, N, S and O, 157(T) machinability of stainless steels, 183 metal flow directions in weld pool with and without, 208(F) pitting resistance of unannealed welds, 42(F) pseudo-binary-phase diagram for iron and, 41(F) stainless steel machinability, 186–188

sulfuric acid austenitic stainless steels, 88 corrosion in, 31–33 corrosion table, 259(T), 260(T), 261(T) corrosion table for fuming, 261(T) duplex alloys, 100 electropolishing solution, 196 influence of alloying element on corrosion rate in contaminated, 33(F) isocorrosion, 88(F) isocorrosion chart for, 258(F) isocorrosion curves for various alloys in, 33(F) isocorrosion curves for various alloys in, with chlorides, 34(F) isocorrosion curves of duplex grades, 100(F) isocorrosion rates for various stainless steels, 32(F) oleum, 33 pickling oxide scale, 193–194 sulfuric acid plus copper sulfate corrosion table for stainless steels and titanium in, 31(F) isocorrosion curves for stainless steel and titanium in, 32(F) sulfurized or sulfochlorinated oils, 179(T) superaustenitic stainless steels, 164 superferritics, 113 super stainless steels, 186 surface finishing aesthetics, 217–219 aesthetic surface finishes, 196–199 austenitic stainless steels, 89 bright annealing, 198 brightening, 196 cleaning, 194–196 coloring, 196 and corrosion resistance, 215 deep drawing, 178–179 function of surface treatments, 193–196 introduction, 193 parameters for oxide film coloring of stainless, 196(T) passivation, 195–196 pickling, 193–194 polished finishes, 197–198 recommended cleaning methods, 195(T) removal of oxide scale, 193–194 rolled finishes, 198–199 rolled-on finishes, 218 (F) surface alteration, 199 surface roughness, 215(F) surface treatments brightening, 196 cleaning, 194–196 coloring, 196 passivation, 195–196 removal of oxide scale, 193–194

T Tafel slope, 23(F) tailpipe, automotive exhaust systems, 227(T), 228 tellurium, 186

Index / 303

temperature chromium-oxygen system volatility, 64(F) corrosion resistance vs., 244(F) critical crevice corrosion, with alloy content, 45(F) critical pitting, (CPT), 43, 44(F) impact strength variation with, for stainless steels, 75(F) partitioning ratio varying with, 93(F) standard Gibbs free energy of metal oxide formation vs., 58(F) variation of pitting potential with, for duplex alloys, 104(F) tempering influencing martensitic stainless hardness, 132, 133(F) martensite, 7 martensitic stainless steels, 167 tensile properties austenitic precipitation-hardenable stainless steel, 145(F) austenitic stainless steels, 75 tensile strength equation, 74 texture deep drawing, 174 ferritic stainless steels, 120–121 thermal conductivity duplex alloys, 205 ferrite, 4–5 thermal cutting, 211 thermal expansion austenitic stainless steels, 202 duplex alloys, 205 ferrite, 4–5 thermal processing annealing, 162–164, 165–166, 170–171 austenitic stainless steels, 161–164 austenitizing, 166–167 duplex stainless steels, 170–171 ferritic stainless steels, 165–166 martensitic stainless steels, 131–133, 166–168 precipitation-hardening stainless steels, 168–170 soaking, 161–162, 165, 166, 170 stress relieving, 164, 166, 167–168 tempering, 167 thermal stresses, predicting, 64, 65(F) Thermo-Calc alloy design, 93, 282 applications, 282–283 casting, 283 free demonstration version, 283 heat treatment, 283 iron-chromium phase diagram, 113(F) map, 282 phase determination program, 2 quality control, 283 Scheil–Gulliver model, 282 single-point equilibria, 282 software package, 281–282 step, 282 welding and joining, 283

thermodynamics argon oxygen decarburization (AOD), 156 influence of alloying elements on, activity of C, N, S, and O, 157(T) oxidation, 57–60 stainless steel, 2 thiosulfate, 50 time-temperature-transformation (TTT) diagram high-alloy stainless steel, 94, 96(F) unstabilized 430-type alloy, 115(F) titanium carbide and nitride formation, 115 carbide former, 78 corrosion table for, in sulfuric acid plus copper sulfate, 31(F) deoxidizer in chromium–iron alloys, 156 ferritic alloy stabilization, 4, 205 inclusions and pitting, 41 influence on thermodynamic activity of C, N, S and O, 157(T) isocorrosion curves for, in sulfuric acid plus copper sulfate, 32(F) isocorrosion curves in phosphoric acid, 36(F) machinability of stainless steels, 184 possible aluminum/titanium precipitates, 138(F) role in sensitization, 47 stabilization of ferritic stainless steels, 118–119 stabilization of ferritic steels, 226 titanium-aluminum-nitride (TiAlN), 191 titanium carbonitride (TiCN), 191 titanium nitride (TiN), 191 tooling carbides, 191 coatings, 191 coolants, 191 costs in deep drawing, 176 high-speed tool steels, 190–191 lubricants, 189, 190 materials in deep drawing, 178–179 tools, martensitic stainless steels, 133–134 toughness austenitic stainless steels, 75–76 duplex alloys, 97–98, 204 ferritic stainless steels, 117, 118(F), 118–119 high-temperature austenitic alloys, 83(F) martensitic stainless steels, 131, 132(F) trains, 232 transient oxidation, 60 transpassive dissolution, 27 transpassive regime, 24 transportation. See automotive and transportation trucks, 231 tubular goods, 134–135 tungsten carbides for flatware, 240 influence on thermodynamic activity of C, N, S and O, 157(T) tungsten inert gas (TIG), 208–210 tuyeres, oxygen injection, 156

304 / Index

U Ugima oxide coating cutting tool and lubricant, 189–190 comparing 304L chips with and without, 190(F) machinability by sulfur levels with and without, 190(F) x-ray showing, 189(F) umbilical tubing and risers, 253–254 uniform corrosion. See also corrosion types environmental variables influencing, 28–29 material variables, 29–31 stainless steel, 27–28 United States, chloride concentration in rainwater, 217(F) unmixed zone, 208

V vacuum arc remelting (VAR), 157–158 vacuum induction melting (VIM), 157 vacuum oxygen decarburization (VOD) cleanliness, 184 refining process, 156 vanadium carbides for flatware, 240 high-temperature martensitic stainless, 133, 134(F) Volvo, auto components, 230

W Wagner’s theory, 61–63 washer tubs and drums, 241 water heaters, 241, 242 waterline corrosion, 38 water marking, cleaning method for uncoated stainless, 220(T) water vapor, 81–82 wax-base pastes, 179(T) wax or soap plus borax, 179(T) weather pattern, stainless steel selection expert system, 216(F) weldability, 253 welding austenitic stainless steels, 201–204 cast stainless alloys, 154 characteristics of stainless steels, 201–206 duplex stainless steels, 204–205

ferritic stainless steels, 205–206 flux cored wire (FCW), 210 gas metal arc welding (GMAW), 210 gas tungsten arc welding (GTAW), 208–210 high-frequency induction, 211 joint design, 208, 209(F) laser, 210 martenistic stainless steels, 206 material selection and performance, 206–208 metal flow directions in weld pool, 208(F) new developments, 212 nondestructive evaluation (NDE), 211 oxyfuel gas welding (OFW), 210 parameters for various stainless steels, 207(T) practices, 211–212 precipitation-hardening (PH) stainless steels, 206 processes, 208–211 recent developments, 211–212 resistance, 210–211 safety, 211 Schaeffler diagram, 202(F) sensitization, 48 shielded metal arc welding (SMAW), 210 soldering and brazing, 211 submerged arc welding (SAW), 210 thermal cutting, 211 Thermo-Calc software, 283 tungsten inert gas (TIG), 208–210 Welding Research Council’s 1992 constitution diagram, 203(F) weld shielding gas composition and crevice corrosion resistance, 204(F) Welding Research Council, constitution diagram, 203(F) welds influence of sulfur on pitting resistance of unannealed, 42(F) long-term annealing, 43 wet carbon dioxide, 248

Y yield strength austenitic precipitation-hardenable stainless steel, 145(F) equation, 74 high-temperature austenitic alloys, 83, 84(F) yttrium, 65

ASM International is the society for materials

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

Product code

Stainless Steels for Design Engineers

#05231G

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

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

Email [email protected]

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

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