The brass-type texture and its deviation from the copper-type texture

Abstract

Our basic aim with the present review is to address the classical problem of the “fcc rolling texture transition” – the fact that fcc materials may, depending on material parameters and rolling conditions, develop two different types of rolling textures, the copper-type texture and the brass-type texture. However, since there is by now reasonable agreement about the description of and the explanation for the development of the copper-type texture (though not about all the details), we have chosen to focus on the brass-type texture for which there is no such general agreement. First we introduce the subject and sketch our approach for dealing with it. We then recapitulate the decisive progress made during the nineteen sixties in the empirical description of the fcc rolling texture transition and in lining up a number of possible explanations. Then follows a section about experimental investigations of the brass-type texture after the nineteen sixties covering texture measurements and microstructural investigations. The main observations are: (1) The brass-type texture deviates from the copper-type texture from an early stage of texture development. (2) Deformation twinning has a decisive effect on the deformation pattern in materials developing a brass-type texture by introducing overshooting/latent hardening, but the volume fraction of twinned material is insufficient to have a significant direct effect on the texture. (3) The development of the brass-type texture may or may not include intermediate development of significant texture components with {1 1 1} approximately parallel to the rolling plane. (4) The deformation pattern during the later stages of the development of the brass-type texture is normally dominated by shear banding. (5) Copper–manganese alloys with more than ∼5% manganese develop a brass-type texture, apparently without significant deformation twinning and shear banding. The next section is about quantitative modelling of the development of the brass-type texture. We first reconsider the explanations for the texture transition suggested in the nineteen sixties and reject a number of them in the light of later investigations. There is by now convincing evidence that the development of the copper-type texture can be explained by Taylor-type models with straight-forward {1 1 1}〈1 1 0〉 slip. The combination of the Taylor model with the formation of a large volume fraction of deformation twins provides reasonable simulations of the brass-type texture, but the volume fraction of twins implied is far greater than that observed experimentally. There are also other deductions from this combination which are contradicted by experiments. Only Sachs-type models seem to work without a substantial volume fraction of deformation twins. The modified Sachs model gives simulated textures which approach quantitative agreement with the experimental brass-type texture at 50% reduction, and it gives reasonable simulated textures even at high reductions. However, experimental observations indicate a composite deformation pattern with slip on one single slip plane in heavily twinned grains and multiple slip in the other grains, a pattern which only in a statistical sense is reflected in Sachs-type models. The models for shear band formation and its effect on texture are rather primitive. For Cu–Mn alloys with high Mn content the transition to a Sachs-type deformation pattern may tentatively, in terms of the GIA model, be explained by their high strength. In the discussion we first look at the effects of deformation twins as observed experimentally. We suggest that deformation twinning is the general explanation for the formation of the brass-type texture via the overshooting/latent hardening introduced by the closely spaced twin lamellae. As to quantitative modelling we discuss the physical basis for the use of Sachs-type models to explain the development of the brass-type texture as opposed to the copper-type texture – even though recent FEM calculations have supported the general use of Sachs-type models at moderate reductions. We suggest that the concept of a composite deformation pattern is introduced in n-point models (models with interaction between specific neighbouring grains), and we quote results from a preliminary attempt to introduce the composite pattern in a complex model. We also suggest that more advanced models (like n-site models) are applied to the shear banding stage. As to the fundamental physical process governing the fcc rolling texture transition we quote recent investigations pointing at cross slip – which is suggested to suppress deformation twinning under copper-type conditions.