Globular to lamellar transition during anomalous eutectic growth

As a typical solidification microstructure, anomalous eutectic is still not known of its growth mechanism. Mullis et al (2018, Acta Mater. 145:186) experimentally showed that the volume faction of anomalous eutectic was not consisted with the predictions by any model invoking partial remelting of primary solidified microstructure. In the present article, the anomalous eutectic microstructure of Ni-Sn alloy solidified from undercooled melts and laser remelting melt pool has been investigated through experiments and cellular automaton (CA) simulations. Computational and experimental results showed that the nucleated α-Ni particulates grew into globular, lamellar or 'tadpole' morphology. The 'tadpole' morphology, which has a globular 'head' and a lamellar 'tail', is an intermediate pattern between globular and lamellar morphologies, and would be seen as an evidence of the globular to lamellar transition (GLT). The occurrence of the GLT or not determines that the solidified microstructure is either anomalous eutectic or lamellar eutectic. CA simulations showed that the GLT was mainly influenced by the temperature gradient G and pulling velocity V. For positive G and V, representing the directional solidification at the bottom of melt pool, the GLT prefers to occur at high G and low V; for negative G and V, which refers to the solidification in undercooled melts, the GLT prefers to occur at low absolute value of G and V. The GLT was also experimentally observed under the above two circumstances. The GLT growth mechanism abstained from CA simulations well explains the experimental results.