Stacking Fault Energies in Al-Based Alloys

Ffom a theoretical point of view, stacking fault energies in metals have been reliably calculated from first-principles with different electronic structure methods [4, 5, 6]. For random alloys, the Layer Korringa Kohn Rostoker method in combination with the coherent potential approximation [7] (LKKR-CPA), was shown to be reliable in the prediction of SFE in fcc-based solid solution [8, 9]. 

In this work, we present calculated SFE using the LKKR-CPA method for Al-Cu and Al-Mg which are of interest from the point of view of superplasticity. We use the SFE to validate the rigid band model which allows a deeper insight into the electronic structure and its implication on the nature of inter-atomic potentials. 

In the Al-Mg system the compositional dependence of the SFE is almost a straightforward interpolation between the fault energies of the pure elements, i.e. the SFE decreases upon addition of Mg to Al, a finding that is confirmed experimentally [11]. 

Rosengaard and Skriver [5] have demonstrated, that in all 3d, 4d, and 5d transition metals, the intrinsic stacking fault energy, 7, can be accurately estimated from the relation, 

As we show in Fig. 2 this relation holds as well for the two Al-based alloys studied here. This finding has consequences on the nature of the inter-atomic interactions. From a fee point of view, the hep structure has a stacking fault every second layer. The fact that relation (1) holds means that these stacking faults weakly interact, and therefore the range of the inter-atomic interactions should not go beyond the second neighbor shell whereas conventional central potentials require at least three atomic shells to differentiate the fee and hep stacking sequences. 

---