Characteristics of lath martensite: Part III. Some theoretical considerations
- PDF / 1,050,305 Bytes
- 10 Pages / 603.28 x 788 pts Page_size
- 4 Downloads / 241 Views
I.
INTRODUCTION
A L T H O U G H the phenomenological crystallographic theories of martensitic transformation 1'2 have been very successful in predicting the crystallography of many martensitic transformations, there are still some transformations which remain poorly understood. Two of these "problem" transformations occur in Fe-alloys, i.e., the formation of heavily dislocated lath martensite with a habit plane close to (111)f, and the formation of twinned plate martensite with a habit plane close to (252)~. The application of the crystallographic theories to lath martensite has been restricted due to the lack of accurate crystallographic and substructural data. Thus, attempts to explain the crystallography of lath martensite have been performed by trial and error, i.e., by allowing the lattice invariant shear to take place on different planes and directions. In one solution, originally due to Wechsler et al, 3 but subsequently further developed by Owen et al, 4"5 it is assumed that the lattice invariant shear takes place on {11 l}f(211)f, and the resulting habit plane is about 7.5 deg from {lll}f. The solution predicts an interface dislocation spacing of only about 16A, which is considerably less than the spacings of 26 to 63A, observed for lath martensite as reported in a previous paper. ~4 The {111}(211) shear proposed in the solution can be factored into two shears of the type {111} (110), but for the lath martensite investigated in the previous paper only one Burgers vector of the type a:/2 (110) could be observed. In another solution, due to Crocker and Bilby, 6 it is assumed that the lattice invariant shear takes place on {010}f (101)f, and the resulting habit plane is close to {899}f. The predicted dislocation spacing is about 27A, but the habit plane is in disagreement with the experimental (575)f habit plane observed in several investigations. 789 Thus, B. P.J. SANDVIK and C.M. WAYMAN are, respectively, Research Associate and Professor, Department of Metallurgy and Mining Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801. Manuscript submitted April 30, 1982. METALLURGICALTRANSACTIONSA
none of the previously developed solutions is in satisfactory agreement with experimental observations. Many attempts have also been made to solve the crystallography of (252)f martensite, for which more experimental information is available than for lath martensite. These attempts are based on allowing more than the single shear described in the basic phenomenological theories to take place. Calculations have been carried out ~~using the equivalent double shear theories developed by Ross and Crocker" and Acton and Bevis, ~2 and using the plastic accommodation model proposed by Bowles and Dunne,13 but the predicted results are not completely satisfactory. From the experimental investigation on the crystallography and substructure of lath martensite in an Fe-20Ni5Mn alloy, described in two previous papers, 9:4 it appears that there is now sufficient experimental data available in order to undertake
Data Loading...
