THE ISOTHERMAL DECOMPOSITION OF AUSTENITE IN THE BAINITE REGION by David William Morgan A thesis submitted i n par t ia l fulfilment of the requirements for the degree of MASTER OF APPLIED SCIENCE i n the department of MINING AND METALLURGY The University of B r i t i s h Columbia A p r i l , 1949 ABSTRACT The isothermal decomposition of austenite i n the bainite region has been examined. The progress of the transformation i n several hypoeuteetoid and eutectoid steels was investigated metallographically from a qualitative point of view. A survey: was made of the information available on the i n i t i a t i o n , course, and end product of the transformation. The factors entering into the transformation were examined separately, their temperature-dependence and interactions investigated. A theory of the decom-position of austenite was proposed, and this theory examined i n the l igh t of the phenomena associated with the isothermal and anisothermal progress of the trans-formation. INEEX i I INTRODUCTION. • . .• 1 I I EXPERIMENTAL METHODS 3 I I I DISCUSSION (a) The Isothermal Transformation of Austenite i n the Bainite Region. 5 (b) The Stear Mechanism of Bainite Formation. . . 12 (c) L a t t i c e Coherency 12 d) The E f f e c t s of Residual Stresses 14 e) Behaviour of the Carbon. . . . . . . . . . . 13 (f) The Extension of F e r r i t e Regions Without Shear. 17 (g) The Energy Change i n Bainite Formation . . . 19 (h) Nucleatiom 20 ( i ) Growth * 22 (j) The Process of Reaction i n the Upper Temperature Region 24 (k) The Process of Reaction i n the Lower Temperature Region 2j? (1) E f f e c t of Grain Size 26 (m) Anisothermal Behaviour . . . . . 27 IY SUMMAHr. 32 Y ACKNOWLEDGEMENTS 36 VI BIBLIOGRAPHY 37 THE ISOTHERMAL DECOMPOSITION OF AUSTENITE IN THE BAINITE REGION I ~ INTRODUCTION Although there has been a great deal of information published on the theory of the demomposition of austenite i n the bainite region, and many theories have been advanced to explain d i f f e r e n t phenomena associated with t h i s decomposition, there has been no recent summary of the available knowledge i n t h i s f i e l d . The object of the present i n v e s t i g a t i o n i s to bring together and correlate the e x i s t i n g data, with a view to a s s i s t i n g i n providing a better understanding of the mechanism of the reaction. The nature of the transformation of austenite i n the bainite region renders i t s study p a r t i c u l a r l y d i f f i c u l t . Two interdependent mechanisms of transformation are av a i l a b l e ; (a) p r e c i p i t a t i o n and growth by d i f f u s i o n , and (b) phase change by • martensitic shear. Since both are temperature-dependent., the complications of the transformation are reduced i f the process i s carried out isothermally rather than during continuous cooling. As some of the factors involved,- such as va r i a t i o n i n carbon concentration on a micro scale, and 2 dis t r ibut ion of internal stress, may not be d i rec t ly observed, their importance must be deduced from the observable phenomena and results associated with the reactions occurring during transformation and other reactions of a s imilar nature. This res t r i c t ion prevents the quantitative evaluation of the rate and course of the transformation from basic principles but, from a knowledge of the reaction behaviour at different tempe- • ratures of transformation, the relat ive importance of the different factors may be estimated. In this investigation the different reactions have been examined individual ly from a thermodynamic and kinet ic viewpoint. Their variat ion with temperature has been indicated. The interaction between the individual reactions has been inves-tigated and a theory proposed, for the decomposition process. This theory has been applied to the experimental data available. Where the terms bainite and fer r i te are used i n many places interchangeably throughout this report, bainite generally i s taken to refer to the l ab i l e aggregate of carbide plus fer r i te (possibly supersaturated with carbom), and fe r r i t e to refer to the body-centered cubic form of i ron , whether supersat-urated with carbon or not, and whether formed by shear or by a diffusive growth. 3 II «. TJiYPgRiMENTAIi METHODS The structures of a number of low-alloy hypo:-eutectoid and eutectold steels have been examined microscopically a f t e r having been p a r t i a l l y transformed isothermally at various tempe-ratures i n the ba i n i t e region. The i n i t i a l stages of isothermal transformation have been investigated i n a series of steels with 0.55% Carbon, with and without 0.3,5% Molybdenum and with varying N i c k e l content. The preparation of these a l l o y s has been, described elsewhere 1)* A s i m i l a r series with higher carbon analysis (up to 0.80 weight percent), was investigated i n the region of ba i n i t e formation. Various low-alloy commercial steels have been examined to varying extents. The specimens were prepared as f l a t discs approximately 0.0.5 inches thick. A wire was attached to each to f a c i l i t a t e handling during heat treatment. The specimens were austenitized f o r 15 minutes at 1600° P. i n a neutral s a l t bath, quenched t o , and held f o r a measured time at, the isothermal transformation temperature i n a s a l t bath, and immediately brine-quenched to room temperature. The treated specimens were ground on emery to remove any possible surface effects and to prepare f o r p o l i s h i n g The specimens were polished e l e c t r o l y t i e a l l y using a mixture of perchloric and acetic aoids, 18^ ml. perchloric aoid, s p e c i f i c gravity 1 . 6 l gm./e.c, 165 ml. acetic a c i d , 10 ml. water, with some aluminum introduced into the s o l u t i o n 2 ) . The p r i n c i p a l ©tenant used was 2% n i t al<. containing 1% Zephiran Chloride.* This preparation of the specimen to be examined was found to produce a highly-detailed undisturbed surface. Structures shown i n the accompagnyimg micrographs were obtained by tre a t i n g commercial SAB 1 0 8 0 s t e e l containing 0 . 7 5 % Carbon* The micrographs were taken using an oil-immersion objective of N.A. 1 . J 5 2 . In i n t e r p r e t i n g these micrographs i t should be noted that the magnifications are s u f f i c i e n t l y large that the observed structure i n the f i e l d shown may not be t r u l y representative of the degree of transformation throughout the specimen. In the following discussion, the r e s u l t s of t h i s metallographic i n v e s t i g a t i o n are given together with a survey of the published r e s u l t s of other investigations. Leading references have been given f o r r e s u l t s drawn from the l i t e r a t u r e . In some oases references have been c i t e d f o r evidence supporting the results of t h i s i n v e s t i g a t i o n . 8 $ aqueous solution, d i s t r i b u t e d by Winthrop Chemical Company, Inc., New York, N.Y. 5 I H - DISCUSSION (a) The Isothermal Transformation of Austenite i n the Bainite Region. . In the early stages of formation, bainite oceurs as lamellae nucleated on the grain boundaries^) ( F i g . I ) . These lamellae consist of f e r r i t e , possibly supersaturated with]: carbon at lower temperatures^ -), and carbide p a r t i c l e s , p r e c i p i -tated, on the f e r r i t e - a u s t e n i t e interface i n the early stages^). Svidence has been given that the habit plane of the bainite i n r e l a t i o n to the parent austenite changes with temperature of transformation^), and that the orientation of the f e r r i t e i n bainite i s independent of temperature of formation^), being the same as proeutectoid f e r r i t e . The cementite i s i n a f i n e state of dispersion, X-ray l i n e i n t e n s i t i e s being considered comparable with those of tempered martensite^). The r e s u l t s of magneto-metric investigations have been interpreted to indicate that i n a l l o y s t e e l s , the carbides tend to be closer to the simpler Fe^G composition as the temperature of transformation i s lower!9). During transformation i n the upper temperature regions the lamellae when i n i t i a l l y formed are i r r e g u l a r , often occuring i n groups of lamellae with s i m i l a r orientation ( F i g . X). These lamellae grow as bloeky formations (Fig. IX) or, sometimes, i n lens shapes which are more concave at higher temperatures (Fig. I I I ) . Nucleation apparently stops soon a f t e r the i n i t i a l period, and the lamellae agglomerate by side-growth. The transformation at high temperatures goes to completion by the agglomeration of e x i s t i n g plates followed by the extension of regions so formed, r e s u l t i n g i n an aggregate of f e r r i t e and cementite. Bainite formed i n the lower temperature range i s f i n e r i n structure, l e s s i r r e g u l a r i n cross-section, and more uniform (Tig. 17). As the temperature of transformation i s lowered the tendency of the lamellae of the same orienta t i o n to group together becomes less apparent^) ( F i g . 17). The decompo-s i t i o n of the austenite i n the lower range goes to completion by the formation of new lamellae, apparently nucleated by e x i s t i n g plates. Analysis of the o v e r - a l l transformation r a t e s 7 ) has shown a progress from three-dimensional towards two-dimensional growth as the temperature of isothermal transformation i s lowered. There i s ev idence3»8) that carbon enrichment of the untransformed austenite occurs with the formation of b a i n i t e . This has "been shown to be thermodynamically l i k e l y ? ) . The enrichment of the austenite by carbon from the ba i n i t e i s oountered by carbon depletion during carbide formation 1^). Figure I SAE 1080, pa r t i a l ly transformed isothermally at 700° F. 2000x. Electropolished. Etched i n 2% n i t a l with Zephiran Chloride. The two micrographs above show the appearance of the i n i t i a l high-temperature bainite formation. The bainite i s i n groups of irregular similarly-oriented lamBllae. The grain-boundary or igin of the bainite may heisbe observed. Figure I I SAE 1080, pa r t i a l ly transformed isothermally at 800° F. 2000 x . Electropolished. Etched i n 2% n i t a l with Zephiran Chloride. This micrograph shows an extreme form of high-temperature bainite. The growth i s acicular , but the rate of agglomeration of the lamellae by sidewise growth i s rapid, therefore only the advancing edges of the plates i n a group are separate. This specimen was cooled to the isothermal trans-formation temperature slowly enough to permit the formation of some nodular pear l i te . Figure I I I SAE 1080, pa r t i a l ly transformed isothermally at 600° F . 2000 x . Electropolished. Etched im 2% n i t a l with Zephiran Chloride. These micrographs i l l u s t r a t e the intermediate-temperature bainite growth. Pa ra l l e l lamellae occur i n groups i n the early stages. The lamellae thicken as they grow, often becoming lens-ehaped as i l l u s t r a t ed . Few new bainite plates appear i n the la ter stages of growth, the transformation going to completion by agglomeration. (a) 2,000 cal/mol f o r carbon i n a u s t e n i t e ! 7 ) , and the average time between basic acts of d i f f u s i o n as being given s t r e s s - r e l i e f by agglomeration i n a l l o y s having a highly-Any carbon atoms entrapped i n bainite w i l l have a approximately by h N e' H/TR h - Planck 1s constant H - a c t i v a t i o n energy t -fo r a d i f f u s i o n movement N - Avogadro*s number R - gas constant T - temperature 16 we may oalculate that at a temperature of 400° C. a carbon atom i n f e r r i t e i s l i k e l y to have about 10,000 times as many ohanges of p o s i t i o n i n any given time as a carbon atom i n austenite. This unbalance of d i f f u s i o n rates, coupled with the large free energy change of carbon between austenite and f e r r i t e w i l l r e s u l t i n a rapid increase i n carbon concentration i n the austenite adjacent to a newly-formed block of b a i n i t e . There i s considerable s t r a i n i n the l a t t i e e near the i n t e r f a c e , and i f the b a i n i t e has l o s t coherency with the austenite we may expect a high m o b i l i t y of d i f f u s i n g atoms at the interface 1**). The heat of reaction w i l l a s s i s t a higher l o c a l m o b i l i t y . Such conditions promote rapid nucleation of carbide, and hence w i l l tend to p r e c i p i t a t e the carbide i n a very f i n e form. The f i n e size of the carbide w i l l r e s u l t i n i t having a lower a l l o y content than the equilibrium conditions at that temperature require, as has been observed and f u l l y discussed 1?). That the carbide of the simple Fe^C structure should p r e c i p i t a t e i n preference to a more complex a l l o y carbide may also be deduced from consideration of the r e l a t i v e p r o b a b i l i t i e s of forming . c r i t i c a l size n u c l e i of the a l t e r n a t i v e carbides. The problem of carbide p r e c i p i t a t i o n i s rendered more complex by the i n e q u a l i t i e s of concentration, that i s , i f by d i f f u s i o n from a region of r a p i d l y formed b a i n i t e p a high-carbon region i s formed, under the conditions of rapid nucleation outlined above most of the n u c l e i w i l l be of c r i t i c a l or near-e r i t i c a l s i z e f o r that concentration of carbon. The concentration of carbon i n the region where the carbides are precipitated i s 17 reduced by d i f f u s i o n towards the unreacted austenite, which i s of lower carbon content, and by p r e c i p i t a t i o n onto the carbides* Wow, as the carbon concentration i s reduced, the c r i t i c a l s i z e of the carbide nucleus, below which the nucleus i s unstable, increases. At temperatures of rapid d i f f u s i o n , i t i s qiuite conceivable that the c r i t i c a l s i z e of nucleus could increase at a greater rate than the increase in. size by growth of the carbids p a r t i c l e s present, and so could exceed the size of many of these, rendering them unstable. The unstable carbides would then redissolve. The r e s u l t would be an increased carbon con-centration i n the unreacted austenite. The carbide p a r t i c l e s remaining a f t e r the carbon con-centration has become more uniform continue to grow i n a normal fashion. The growth of carbides implies a reduction i n the carbon concentration of the surrounding austenite. This process acts simultaneously with the carbon-enrichment by bainite f o r -mation. At any one temperature of reaction, whether or not the unreacted austenite i s enriched or depleted with regard to carbon w i l l depend upon the r e l a t i v e rates of the bainite and the carbide reactions.i One may expect that carbon-enrichment w i l l occur when the bainite reaction i s more rapid, depletion when slower. This i s indicated experimentally8,10). (f) The Extension of f e r r i t e Regions Without Shear. I f an a u s t e n i t e - f e r r i t e interface i s considered a f t e r i t has l o s t coherency, i t may be seen that i n d i v i d u a l i r o n atoms could jump from positions i n the austenite to more stable positions 18 i n the f e r r i t e without increasing the shear stresses associated with the shear mechanism of ba i n i t e formation. Such a method of phase growth has been shown to be dependent upon the d i f f u s i o n rate of carbon away from the i n t e r f a c e 9 ) . As a r e s u l t of the concentration gradient caused by bainite formation, d i f f u s i o n may increase the carbon concentration i n the unreacted austenite. The amount of the enrichment w i l l be influenced by the degree of carbide p r e c i p i t a t i o n . Therefore three factors w i l l l a r g e l y control the rate of transformation of austenite to f e r r i t e i n t h i s manner: the carbon concentration i n the austenite remote from the interface, the a c t i v a t i o n energy of i r o n transfering across the inte r f a c e , and the d i f f u s i o n rate of carbon. The carbon concentration i s dependent upon the degree of transform.-*-ation, increasing i n the upper bainite range, showing l i t t l e change at lower temperatures.' The temperature-dependence of the rate of growth at t r i b u t a b l e to the i r o n and carbon a c t i v a t i o n requirements w i l l be of the' form ezp(-A/RT), where A i s propor-t i o n a l to the a c t i v a t i o n energies. Since the carbon d i f f u s i o n rate i s the predominant.factor i n such a r e a c t i o n 9 ) , the rate of growth w i l l decrease approximately exponentially with temperature.. Growth i n t h i s manner w i l l f a c i l i t a t e the segregation of any a l l o y i n g elements, those such as n i c k e l and manganese which have lower free energy when dissolved i n austenite tending to diffuse away from the boundary so as to stay i n the austenite, and those elements such as chromium which have lower free energy when dissolved i n f e r r i t e tending to enter the b a i n i t e . Such segregation has been advanced as an explanation f o r the abnormally 19 long times f o r completion of transformation i n the upper bainite region i n c e r t a i n a l l o y s t e e l s ^ ) . (g) The Energy Change i n Bainite Formation. A phase change i s possible only i f the free energy i s decreased by the reaction. Considering bainite formed by shear, with entrapped carbon, four factors determine the free energy changer, the free energy change of the i r o n i n going from the face-centered structure to the body-centered, dG|,Q; the difference i n free energy between carbon i n a face-centered cubic l a t t i c e , and carbon i n the body-centered cubic l a t t i c e dG^; the change i n entropy of the carbon, dS; and the change in. s t r a i n energy assoc-iated with the transformation, dU. This may he written a s 9 ) : dG = dGpe • CdGc - CTdS * dU. ' C - carbon concentration T - temperature There i s no need to consider surface energy i f the l a t t i c e s are assumed coherent at the time of transformation. In any transformation there i s a c e r t a i n a c t i v a t i o n energy which controls the rate of reaction. Tor martensitic-shear reactions with small movements t h i s i s i n s i g n i f i c a n t , however, and may be neglected. Thus assuming that the transformation w i l l occur when-ever the free energy change i s negative, we w i l l examine the factors involved to account f o r the time-dependence of the bainite reaction. dGj,Q, dG c, and dS are independent of time, but C, the carbon concentration, and dU, the s t r a i n energy change, w i l l 20 fluctuate w i t l i time and with the degree of transformation. The carbon concentration w i l l fluctuate by chance d i f f u s i o n , by the effect .of nearby bainite formation, by p r e c i p i t a t i o n of carbides as described, and by variations i n i n t e r n a l stress conditions. References 1 8 , 2 0 , 2 1 ) . The i n t e r n a l stress w i l l be very high under the i n i t i a l effect of quenching, and w i l l be raised by bainite transformation. I t w i l l r elax at an appreciable rate i n the temperature range of bainite formation. The rel a x a t i o n rate i s temperature dependent exponentially, of the form exp(-B/T), where B i s dependent upon the amount of i n t e r n a l stress. I f we consider a single bainite plate growing edgewise by shear, we may see that the rate w i l l be r e s t r i c t e d by the relaxation rate of the opposing residual stresses. As the tempe-rature of transformation i s lowered the decrease i n free energy by the change of i r o n from the face-centered cubic to the body-centred cubic form increases r a p i d l y , p a r t i a l l y o f f s e t t i n g the decrease i n the relaxation rate of the i n h i b i t i n g stresses. (h) Nucleation. Metallographic examination has shown that bainite tends to nucleate p r e f e r e n t i a l l y on the grai n boundaries?> 2 2). This i s supported by the examination of proeutectoid f e r r i t e , which has a cl o s e l y - r e l a t e d mode of formation, and which, because of i t s l e s s e r tendency towards lamellar growth, shows more c l e a r l y i t s region of nucleation. There i s evidence that nucleation and growth of b a i n i t e are greatly assisted by p l a s t i c f l o w 2 ? ) . Nucleation of bainite i n the grain boundaries i s to be expected. 21 The contribution to the s t r a i n energy by volume change i s smaller at the grain boundaries 1^). Amongst the disordered material there i s a higher p r o b a b i l i t y of f i n d i n g s i t e s p a r t i c -u l a r l y suited to n u c l e a t i o n 2 ^ ) . Regions having a favourable i n t e r n a l stress f o r nucleation may be expected because of the constraining e f f e c t of the grain boundaries, the v a r i a t i o n i n i n t e r n a l stress being p a r t i c u l a r l y great during the t