Atomic mechanisms of diffusional nucleation and growth and comparisons with their counterparts in shear transformations

An integrated overview is presented of a viewpoint on the present understanding of nucleation and growth mechanisms in both diffusional and shear (martensitic) transformations. Special emphasis is placed on the roles played by the anisotropy of interphase boundary structure and energy and also upon elastic shear strain energy in both types of transformation. Even though diffusional nucleation is based on random statistical fluctuations, use of the time reversal principle shows that interfacial energy anisotropy leads to accurately reproducible orientation relationships and hence to partially or fully coherent boundaries, even when nucleation at a grain boundary requires an irrational orientation relationship to obtain. Since the fully coherent boundary areas separating most linear misfit compensating defects are wholly immobile during diffusional growth because of the improbability of moving substitutional atoms even temporarily into interstitial sites under conditions normally encountered, partially and fully coherent interphase boundaries should be immovable without the intervention of growth ledges. These ledges, however, must be heavily kinked and usually irregular in both spacing and path if they, too, are not to be similarly trapped. On the other hand, the large shear strain energy usually associated with martensite requires that its formation be initiated through a process which avoids the activation barrier associated with nucleation, perhaps by the Olson-Cohen matrix dislocation rearrangement mechanism. During growth, certain ledges on martensite plates serve as transformation dislocations and perform the crystal structure change (Bain strain). However, the terraces between these ledges in martensite (unlike those present during diffusional growth) are also mobile during non-fcc/hcp transformations; glissile dislocations on these terraces perform the lattice invariant deformation. Growth ledges operative during both diffusional and shear growth probably migrate by means of kink mechanisms. However, diffusional kinks appear to be nonconservative and sessile (and therefore resist immediate transmission of elastic shear strain energy), whereas those associated with martensitic growth must be conservative and glissile (and fully transmit such strain energy). The broad faces of both diffusionally and martensitically formed plates contain an invariant line, as emphasized by Dahmen and Weatherly. However, in the diffusional case, minimization of growth ledge formation kinetics seems to be the main role thereby played, whereas in martensitic growth, the main purpose of such an interface is to minimize elastic shear strain energy. The latter minimization requires that martensite forms as plates (or perhaps as laths) enclosed by a pair of invariant line-containing interfaces. During diffusional transformations, on the other hand, other interfaces at which growth ledge formation kinetics are not too much faster than those at the invariant line interface can also comprise a significant portion of the interfacial area, thereby leading to the formation of other, quite different morphologies, such as intragranular idiomorphs and grain boundary allotriomorphs. Critical problems remaining unsolved in diffusional transformations include calculation of critical nucleus shapes when the crystal structures of the two phases are significantly different, highly accurate calculation of the energies of the interphase boundaries thus formed, and direct observation of atomic scale kinks on the risers of growth ledges by means of a yet-to-be-invented three-dimensional (3-D) atomic-resolution form of transmission electron microscopy. Experimental identification and characterization of transformation dislocations and experimental testing of “nucleation” mechanisms are now of special importance in fundamental studies of martensitic transformations.     

A history of the controversy over the roles of shear and diffusion in plate formation aboveM d and a comparison of the atomic mechanisms of these processes

The evolution of the controversy over the roles played by shear and by diffusion in the formation of plate-shaped transformation products aboveM _d is traced along its principal pathways. Growth by shear is taken to occur through the glide of dislocations, whereas diffusional growth is considered to occur by means of the biased random walk of individual atoms. Both mechanisms are based upon growth ledges, but during shear, risers, and in some cases terraces, are mobile in non-fcc/hcp transformations; however, during diffusional growth, only the risers (and more likely, only kinks on the risers) are mobile when there is a stacking sequence difference across the terrace of a growth ledge. The aura of the powerful and elegant phenomenological theory of martensite crystallography (PTMC) appears to have persuaded many researchers that even above M_d, plates usually form by shear because one or more of the multiple requirements for the applicability of the PTMC appears to be fulfilled. As the preceding article by Wayman in these conference proceedings emphasizes, however,all of these requirements must be fulfilled in a self-consistent manner if a plate is to qualify as the product of a shear transformation mechanism. Successful application of the invariant line (IL) component of the PTMC to precipitate plates and needles or rods by Dahmen and co-workers provides a phenomenological reason why the crystallographies and surface reliefs of ledgewise shear and ledgewise diffusional growth sometimes appear to be identical. While the IL-containing broad faces of a martensite plate minimize the transformation strain energy, the same situation in plates formed by ledgewise diffusional growth tentatively appears to engender a minimum in the kinetics of growth ledge generation and/or the lateral growth kinetics of such ledges. This article is based on a presentation made at the Pacific Rim Conference on the “Roles of Shear and Diffusion in the Formation of Plate-Shaped Transformation Products,” held December 18-22, 1992, in Kona, Hawaii, under the auspices of ASM INTERNATIONAL’S Phase Transformations Committee.     

Interphase boundary structures associated with diffusional phase transformations in Ti-base alloys

Interphase boundary structures generated during diffusional transformations in Ti-base alloys, especially the proeutectoid α and eutectoid reactions in a β-phase matrix, are reviewed. Partially coherent boundaries are shown to be present whether the orientation relationship between precipitate and matrix phases is rational or irrational. Usually, these structures include both misfit dislocations and growth ledges. However, grain boundary α allotriomorphs (GBA’s) do not appear to develop misfit dislocations at partially coherent boundaries. Evidently, these dislocations can be replaced by ledges which provide a strain vector in the plane of the interphase boundary. The bainite reaction in Ti-X alloys produces a mixture of eutectoid α and eutectoid intermetallic compound. Both eutectoid phases are partially coherent with theβ matrix, and both grow by means of the ledge mechanism, though unlike pearlite the ledge systems of the two phases are structurally independent. Even after deformation and recrystallization, the boundaries between the eutectoid phases and theβ matrix, as well as between these phases, are partially coherent. Titanium and zirconium hydrides have partially coherent interphase boundaries with respect to theirβ matrix. The recent observation of ledgewise growth of γ TiH within situ high-resolution transmission electron microscopy (HRTEM) suggests that, repeated suggestions to the contrary, these hydrides do not grow by means of shear transport of Ti atoms at rates paced by hydrogen diffusion. This paper is based on a presentation made in the symposium “Interfaces and Surfaces of Titanium Materials” presented at the 1988 TMS/AIME fall meeting in Chicago, IL, September 25–29, 1988, under the auspices of the TMS Titanium Committee.     

Intrinsic ledges at interphase boundaries and the crystallography of precipitate plates

The structure of intrinsic ledges at interphase boundaries has been interpreted with extended O-lattice/DSC-lattice approaches. The distribution of structural ledges can be predicted if the spacing difference between parallel matrix and product planes is treated as a measure of the relaxed coincidence condition. A small rotation away from the low-index planar parallelism introduces a series of interfacial dislocations that cancels the spacing difference, resulting in a lattice invariant line. Misfit-compensating ledges at bcc: hcp interfaces are produced as a ledged interface intersects additional O-points that are recognized with the incorporation of previously omitted bcc atom positions into the O-lattice construction. Energetic consideration suggests that structural interfacial energy may decrease when a flat interface becomes ledged with misfit-compensating ledges. Burgers vectors associated with structural ledges and misfit-compensating ledges are displacement shift complete (DSC) lattice vectors. Precipitate and martensite crystallography may both include a lattice invariant line, but they are involved in different interphase boundary characteristics. Assumptions and implications in precipitate and martensite crystallography are discussed in the framework of the O-lattice theory and phenomenological theory of martensite crystallography. This article is based on a presentation made at the Pacific Rim Conference on the “Roles of Shear and Diffusion in the Formation of Plate-Shaped Transformation Products,” held December 18-22, 1992, in Kona, Hawaii, under the auspices of ASM INTERNATIONAL’S Phase Transformations Committee.     

The role of ledges in the proeutectoid ferrite and proeutectoid cementite reactions in steel

The influence of interphase boundary ledges on the growth and morphology of proeutectoid ferrite and proeutectoid cementite precipitates in steel is examined. After reviewing current theoretical treatments of growth by the ledge mechanism, investigations that clearly document the presence and motion of ledges with thermionic emission electron microscopy (THEEM) and transmission electron microscopy (TEM) are reviewed. A fundamental distinction is made between two types of ledges: (1) mobile growth ledges whose lateral migration displaces the inter-phase boundary and (2) misfit-compensating structural ledges. Both types of ledges strongly affect the apparent habit plane and aspect ratio of precipitate plates. Agreement between measured growth rates of proeutectoid ferrite and cementite (plates and allotriomorphs) and predicted growth kinetics assuming volume diffusion-controlled migration of ledge-free disordered boundaries is shown to be consistently poor. Physically realistic growth models should incorporate the ledge mechanism. More accurate comparisons of the growth models with experimental data will need to account for observed ledge heights, interledge spacings, and ledge velocities. In this vein, the sluggish growth kinetics of cementite allotriomorphs observed in an Fe-C alloy are shown to be quantitatively consistent with a strong increase in interledge spacing with time. This paper is based on a presentation made in the symposium “The Role of Ledges in Phase Transformations” presented as part of the 1989 Fall Meeting of TMS-MSD, October 1–5, 1989, in Indianapolis, IN, under the auspices of the Phase Transformations Committee of the Materials Science Division, ASM INTERNATIONAL.     

Comparison of interfacial structure-related mechanisms in diffusional and martensitic transformations

Some central problems in understanding the similarities of and the differences between ledgewise martensitic and ledgewise diffusional growth are examined. Martensitic growth can be described in terms of a lattic correspondence and a plane undistorted by the shear transformation. Diffusional growth can be similarly described in some cases but not in others. On the basis of the Sutton-Balluffi definitions of glissile and sessile boundaries, only misfit dislocations (on terraces or risers) or orthogonal sets of disconnections provide a truly sessile interface. When closely spaced structural ledges (now termed “structural disconnections”) are present during diffusional growth, they must have been glissile in the formation of a local equilibrium structure during the initial stages of growth. Once they are in local equilibrium and evenly spaced, however, they can only move synchronously because of their local strain interaction. Under these circumstances, extrinsic sources of growth ledges are required to move such interfaces in a diffusional manner. During martensitic growth, however, disconnections in the form of transformation dislocations can move freely in a synchronous manner. Also, on this basis, the apparent (undistorted) habit plane is generally useful in deducing the transformation mechanism during martensite formation, but is only occasionally so during diffusional growth, where only the terrace plane is generally useful. This article is based on a presentation in the symposium “Interfacial Dislocations: Symposium in Honor of J.H. van der Merwe on the 50th Anniversary of His Discovery,” as part of the 2000 TMS Fall Meeting, October 11–12, 2000, in St. Louis, Missouri, sponsored under the auspices of ASM International, Materials Science Critical Technology Sector, Structures.     

Computer simulation of ledge migration under elastic interaction

The diffusional growth of a phase by the motion of disconnections (ledges which contain transformation or misfit dislocations) was studied by a finite difference computer model. The elastic stress of these dislocations is considered to alter the (local equilibrium) solute concentration at the riser of ledges and cause a complex diffusion field interaction among ledges as they migrate. In some cases, however, the ledges forming a train can migrate all at the same speed in the presence of elastic interaction. The condition under which ledges overcome the elastic barrier and form a multipleheight ledge was determined. The model was applied to the migration of ledges/Shockley partial dislocations at γ′-plate interfaces in Al-Ag alloys. This article is based on a presentation made during TMS/ASM Materials Week in the symposium entitled “Atomistic Mechanisms of Nucleation and Growth in Solids,” organized in honor of H.I. Aaronson’s 70th Anniversary and given October 3–5, 1994, in Rosemont, Illinois.     

The elastic strain energy of growth ledges on coherent and partially coherent precipitates

The formation rate of growth ledges on a faceted precipitate strongly affects the growth kinetics and the shape of the precipitate. An Eshelby-type model is used to compare the strain energy associated with the nucleation of a ledge on different facet planes of a body-centered cubic (bcc) precipitate in face-centered cubic (fcc) matrix. Ledge nucleation is only likely at facet areas where the interaction energy between the ledge and the precipitate is negative. The strain energy for ledge formation is not symmetric on any of the facet planes, but it is symmetric about the center of the precipitate. For coherent precipitates comparable to those observed in the Ni-Cr system, ledges form with the lowest strain energy on the broad facet of the precipitate implying that precipitate thickening should occur faster than lengthening and widening. A procedure for modifying the Eshelby model is suggested in order to allow strain-energy calculations of partially coherent precipitates. The strain energy for ledge formation on at least one type of partially coherent lath is lowest for a ledge located on the facet perpendicular to the crystallographic invariant line (IL). This situation favors precipitate lengthening in the invariant line direction.     

Atomic structure of interphase boundary enclosing bcc precipitate formed in fcc matrix in a Ni-Cr alloy

The atomic structure of the interphase boundaries enclosing body-centered cubic (bcc) lath-shape precipitates formed in the face-centered cubic (fcc) matrix of a Ni-45 mass pct Cr alloy was examined by means of conventional and high-resolution transmission electron microscopy (HRTEM). Growth ledges were observed on the broad faces of the laths. The growth ledge terrace (with the macroscopic habit plane ) contains a regular array of structural ledges whose terrace is formed by the (111)_fcc//(110)_bcc planes. A structural ledge has an effective Burgers vector corresponding to an transformation dislocation in the fcc → bcc transformation. The side facet (and presumably the growth ledge riser) of the bcc lath contains two distinct types of lattice dislocation accommodating transformation strains. One type is glissile dislocations, which exist on every six layers of parallel close-packed planes. These perfectly accommodate the shear strain caused by the stacking sequence change from fcc to bcc. The second set is sessile misfit dislocations (∼10 nm apart) whose Burgers vector isa/3[111]_fcc =a/2[110]_bcc. These perfectly accommodate the dilatational strain along the direction normal to the parallel close-packed planes. These results demonstrate that the interphase boundaries enclosing the laths are all semicoherent. Nucleation and migration of growth ledges, which are controlled by diffusion of substitutional solute atoms, result in the virtual displacement of transformation dislocations accompanying the climb of sessile misfit dislocations and the glide of glissile dislocations simultaneously. Such a growth mode assures the retention of atomic site correspondence across the growing interface. formerly Graduate Student, Kyoto University, Kyoto 606-01, Japan This article is based upon a presentation made at the Pacific Rim Conference on the “Roles of Shear and Diffusion in the Formation of Plate-Shaped Transformation Products,” held December 18–22, 1992, in Kona, Hawaii, under the auspices of ASM INTERNATIONAL’S Phase Transformations Committee.     

Mechanism of Al2CuLi (T 1) nucleation and growth

Conventional strain contrast transmission electron microscopy (CTEM) and high-resolution transmission electron microscopy (HRTEM) were performed to establish the nucleation and growth mechanism of Al₂CuLi (T₁) precipitates in an Al-Li-Cu alloy. It is shown that the growth mechanism ofT ₁ precipitate plates occurs by the diffusional glide of growth ledges composed of b = 1/6〈112〉 partial dislocations on 111 matrix planes and that the growth ledges migrate by the ledge-kink mechanism, as previously suggested by Cassadaet al. 1 for this system.T ₁ plate nucleation is modeled as the dissociation of a perfect b = 1/2〈110〉 matrix dislocation in the vicinity of a dislocation jog. The coordinated dissociation of the dislocation line segments on each side of the sessile jog provides the displacement necessary for the formation of a new hexagonal plate or plate ledge. Strain contrast analysis of the Burgers vector of plate edges and the edges of growth ledges indicates the stacking of partial dislocations is of mixed displacement. Formerly Graduate Student, Department of Materials Science, University of Virginia,     

On the role of interphase-boundary structure in plate growth by diffusional mechanisms

The structures of planar phase interfaces and of interfacial defects responsible for their diffusional migration are discussed in terms of extensions of the O-lattice concept, in which the intersection of the two structures is treated analytically. Two cases are considered and illustrated with well-characterized experimental examples: one in which two-dimensional structural matching leads to O-planes, and a second in which linear matching yields an array of O-lines. It is suggested that growth ledges moving normal to the O-lines will often require lateral kink formation and motion for their propagation. The misfit associated with transformation ledges is modeled in terms of real (screened) dislocations, which may coexist with virtual (unscreened) dislocations representing a residual component of misfit. A macroscopic shear can result from the cumulative action of transformation ledges with shear components parallel to the habit plane. This article is based0 on a presentation made at the Pacific Rim Conference on the “Roles of Shear and Diffusion in the Formation of Plate-Shaped Transformation Products,” held December 18-22, 1992, in Kona, Hawaii, under the auspices of ASM INTERNATIONAL’S Phase Transformations Committee.     

The pearlite-austenite growth interface in an Fe-0.8 C-12 Mn alloy

The interphase boundary structure and interface processes at the pearlite-retained austenite growth interface in Fe-0.8 wt% C-12 wt% Mn alloy have been investigated by transmission electron microscopy. Facetting, misfit correcting dislocations, and ledge defects are all observed at the previously assumed disordered boundary. Hot stage electron microscopy revealed that the ledge defects are mobile, indicating the migration of the growth interface occurs by the lateral movement of steps. It is found that the growth ledges are continuous across the growth interfaces of the pearlitic ferrite and cementite. This provides a mechanism by which the interface processes of the two pearlite phases may be coupled.     

High-resolution transmission electron microscopy investigation of the face-centered cubic/hexagonal close-packed martensite transformation in Co-31.8 wt pct Ni alloy: Part 1. Plate interfaces and growth ledges

The face-centered cubic/hexagonal close-packed (fcc/hcp) martensite transformation in a Co-31.8 wt pct Ni alloy was studied by high-resolution transmission electron microscopy (HRTEM). High-resolution transmission electron microscopy was used to study the structure and properties of growth ledges, the tips of martensite plates, and martensite nucleation sites. The HRTEM image simulations were performed in order to determine the effects of both beam and crystal tilt on the experimental images. In the investigation, it was determined that the fcc/hcp martensite transformation in Co-Ni occurs by the passage of Shockley partial dislocation ledges (b=1/6〈112〉) along every other (111) plane in the fcc matrix. The hcp martensite thickens by the lateral movement of ledges across the fcc/hcp interface. Although superledges were observed, the majority of the ledges were two (0002) planes and this is the basic ledge height. Image simulations show that both beam and crystal tilt can have a marked effect on HRTEM images of fine hcp martensite plates. The effects of tilt must be minimized in order to unambiguously resolve the interfacial structure.     

The selection of precipitate habit planes in Cr-32 Wt Pct Ni

Causes are investigated for the changes in precipitate crystal structure (fcc to 9R) and in morphology (degenerate plate to plate), which are observed to take place in Cr-Ni alloys as the reaction temperature decreases. Transmission electron microscopy (TEM) study is performed to determine the matrix/precipitate orientation relationship, habit plane, and growth ledge spacing. O-lattice modeling is used to show that it is likely that the metastable 9R phase forms as a transition phase at lower reaction temperatures because lattice matching at the bcc/9R habit plane is better than the matching at the bcc/fcc habit plane. The ability of the phenomenological theory of martensite crystallography (PTMC) to predict the habit plane of 9R plates precipitated by a diffusional mechanism is explained by the small lattice invariant deformation required to produce an invariant plane in Cr-Ni. Under this circumstance, the PTMC habit plane nearly coincides with the best-matching interface that presumably appears and is predicted by O-lattice theory. Formerly Graduate Student, Department of Materials Science and Engineering, Virginia Polytechnic Institute and State University This article is based on a presentation made at the Pacific Rim Conference on the “Roles of Shear and Diffusion in the Formation of Plate-Shaped Transformation Products,” held December 18-22, 1992, in Kona, Hawaii, under the auspices of ASM INTERNATIONAL’S Phase Transformations Committee.     

Computer simulation of morphological changes of grain boundary precipitates growing by the ledge mechanism

A previously developed computer model was modified to simulate the growth of grain boundary precipitates which grow by the ledge mechanism. The ledges were assumed to be nucleated in the grain boundary region at constant, parabolically decreasing, and random rates and to grow under the control of volume diffusion of solute to or from the riser of ledges. At lower under coolings at which the motion of individual ledges is slow, late-nucleated ledges soon catch up with first-nucleated ones, and precipitates tend to extend along the grain boundary: the overall precipitate shape is essentially that of a grain boundary allotriomorph. At larger undercoolings, first-nucleated ledges move fast to form a protuberance similar to Widmanstätten sideplates, while late-nucleated ones stay near the grain boundary region. The transition of precipitate shape from one to the other occurs in a very narrow range of supersaturation. The results are compared with various characteristics of the growth of proeutectoid ferrite allotriomorphs and sideplates in Fe-C alloys documented in the literature.     

Interface dislocations and ledges in oxidation and diffusional phase transformations

The origin of ledge concepts in growth from the vapor is reviewed. The ideas are extended to solid-state phase transformations with the added effects of strain and misorientation. Types of ledges and dislocations are classified. The concepts are illustrated for the example of oxidation of a metal. Further extensions to diffusional phase transformations are briefly discussed. This paper is based on a presentation made in the symposium “The Role of Ledges in Phase Transformations” presented as part of the 1989 Fall Meeting of TMS-MSD, October 1–5, 1989, in Indianapolis, IN, under the auspices of the Phase Transformations Committee of the Materials Science Division, ASM INTERNATIONAL.     

Atomic mechanisms of precipitate plate growth in the AlAg system—II. High-resolution transmission electron microscopy

High-resolution electron microscopy was used to study the interfacial structure of γ′ precipitates in an Al-15 wt% Ag alloy aged at 350°C. The results of these studies show that:

• 1.(1) all ledges are multiples of two {111} planes high, supporting the theory and conventional transmission electron microscopy observations that plate thickening occurs by passage of Shockley partial dislocations on alternate {111} planes
• 2.(2) most ledges are more than just two planes high, indicating a strong tendency toward diffusional and/or elastic interactions
• 3.(3) the terraces between ledges are atomically flat and ledges are uniformly stepped-down from the centers to the edges of isolated precipitates as predicted by the general theory of precipitate morphology
• 4.(4) the {111} planes are continuous across the edges of ledges, indicating that they are largely coherent and not disordered as treated in most kinetic analyses, and
• 5.(5) the edges of precipitate plates appear to be composed of similar two-plane ledges arranged vertically above one another and hence, may grow by the same mechanism of atomic attachment as ledges on the broad faces.

Examination of γ′ plates during early stages of growth indicates that their aspect ratio may deviate from the equilibrium value almost immediately, probably due to the ledge mechanism of growth. Lastly, an atomic model of a γ′ precipitate was used to test the high-resolution images obtained, and illustrate possible atomic growth mechanisms of the ledges.     

Interfacial structure and crystallographic studies of transformations in βt$́and β CuZn alloys—II. martensitic formation of α1′ plates in β′

A TEM study of the interphase boundary structure of 9R orthorhombic α₁' martensite formed in β′ CuZn alloys shows that it consists of a single array of dislocations with Burgers vector parallel to 〈110〉_β′ and spaced about 3.5 nm apart. This Burgers vector lies out of the interface plane; hence the interface dislocations are glissile. Unexpectedly, though, the Burgers vectors of these dislocations are not parallel when referenced to the matrix and the martensite lattices. This finding is rationalized on published hard sphere models as a consequence of relaxation of a resultant of the Bain strain and lattice invariant shear displacements within the matrix phase.     

Interfacial structure and crystallographic studies of transformations in β′ and β CuZn alloys—I. Isothermal formation of α1 plates from β′

A TEM study was made of the structure of the broad faces and edges of α₁ plates formed in β′ CuZn alloys. Thickening kinetics of α₁ plates were also measured. α₁ plates were found to be fully coherent and also free of internal structure during the earliest observed stage of their growth. A single array of misfit dislocations bounding stacking faults, and also growth ledges, develop during the second growth stage and a second array of misfit dislocations appears during the third stage of growth. The inter-dislocation spacings in the two arrays are roughly 2 and 12 nm, respectively. Both interphase boundary structure and thickening kinetics are shown to be inconsistent with various shear mechanisms of growth but to fit comfortably into the framework provided by diffusion-controlled ledge-wise thickening. A ledge structure was also observed on α₁ plate edges, in agreement with deductions made by Simonen and Trivedi from their measurements of the lengthening kinetics of these plates.     

Bainite viewed three different ways

The present status of the three principal definitions of bainite currently in use is reviewed. On the surface relief definition, bainite consists of precipitate plates, producing an invariant plane strain (IPS) surface relief effect, which form by shear,i.e., martensitically, at temperatures usually aboveM _s andM _d . The generalized microstructural definition describes bainite as the product of the diffusional, noncooperative, compctitive ledgewise growth of two precipitate phases formed during eutectoid decomposition, with the minority phase appearing in nonlamellar form. This alternative mode of eutectoid decomposition is thus fundamentally different from the diffusional, cooperative, shared growth ledges mechanism for the formation of pearlite developed by Hackney and Shiflet. The overall reaction kinetics definition of bainite views this transformation as being confined to a temperature range well below that of the eutectoid temperature and being increasingly incomplete as its upper limiting temperature, the kineticB _s , is approached. Recent research has shown, however, that even in steels (the only alloys in which this set of phenomena has been reported), incomplete transformation is not generally operative. Revisions in and additions to the phenomenology of bainite defined in this manner have been recently made. Extensive conflicts among the three definitions are readily demonstrated. Arguments are developed in favor of preference for the generalized microstructural definition, reassessment of the overall reaction kinetics definition, and discarding of the surface relief definition. This paper is based on a presentation made in the symposium “International Conference on Bainite” presented at the 1988 World Materials Congress in Chicago, IL, on September 26 and 27, 1988, under the auspices of the ASM INTERNATIONAL Phase Transformations Committee and the TMS Ferrous Metallurgy Committee.