Development and testing of algorithms for predicting the glass-forming ability in the multicomponent Sc-Gd-Co-Al metallic alloys

Abstract

To reduce material costs, it is vital to develop reliable theoretical methods for predicting glass formation regions in multicomponent metal systems. We have developed a new model for predicting the compositions of quaternary amorphous metallic glasses and applied it to the Gd-Sc-Co-Al system. The proposed model parameters ${\Theta }_{i-j}^{1,2}\left(x_i\right),{\Theta }_{i-j}^{1,2}\left(x_j\right)$ make it possible to predict the location of glass-forming compositions by finding the minimum of the P_HSS parameter multiplication by the geometric coefficient ${\Gamma }_{1,2}$, ${\Gamma }_{1,2}$. The alloy with the composition Sc₃₃Gd₃₂Co₁₉Al₁₆ was successfully predicted and cast as a glass rod with a lateral size of 3 mm. According to the results of X-ray diffraction and thermal analysis, the sample Sc₃₃Gd₃₂Co₁₉Al₁₆ is an amorphous material with the lowest crystallinity index (4.3 %). The proposed approach can be used as a predictive model for determining glass-forming compositions in various quaternary metal systems.