Prediction of Phase-transformations of Ti6Al4V additively manufactured during Directed Energy Deposition

Directed Energy Deposition (DED) has been highly used to fabricate and repair
components made from different metallic alloys. The characteristics of this process
enhances high cooling and heating rates, which results in an out-of-equilibrium
microstructures of the build parts. Ti6Al4V additively manufactured alloy is widely
used in a variety of applications such as aerospace, automotive, marine equipment
and biomechanical applications. However, the improvement of its mechanical
properties is still a challenge because they are influenced by the process conditions.
The Mechanics of Solid and Materials (MSM) and Metallic Solid Materials (MMS) of
ULiège have been focused to correlate DED process parameters and the obtained
microstructures. The approach has been based on a strong combination between
experimental investigations and numerical modelling [1, 2, 3]. First investigations on
this topic consisted in the implementation of a numerical model to predict phase
transformations of Ti6Al4V [1]. This approach is based on discretization of the thermal
histories (heating and cooling) which have been obtained by Finite Element
simulations [3] to capture temperature oscillations in the clad parts during
Manufacturing. An advanced approach, consisting in the segmentation of the
temperature history in different Time-phase – Temperature – Blocks (TTB) [2], allows
the correlation between the thermal histories and the final microstructure. In this
context, in a first step, we proceed by summarizing the previous results. Then, a new
achievement consisting in modelling a “meso-clad” of Ti6Al4V with mutli-validations
(multiple thermocouples and microscopic measurements) is presented. Finally, we
present our perspectives by exploiting promising tools in particular machine learning.