SIM-AM 2023

Metal additive manufacturing melt pool modeling based on a novel FEM-based multi-physics model including melting and evaporation

  • Schreter-Fleischhacker, Magdalena (Technical University of Munich)
  • Much, Nils (Technical University of Munich)
  • Munch, Peter (University of Augsburg)
  • Kronbichler, Martin (University of Augsburg)
  • Wall, Wolfgang A (Technical University of Munich)
  • Meier, Christoph (Technical University of Munich)

Please login to view abstract download link

Under typical process conditions of laser powder bed fusion (LPBF) metal additive manufacturing, the boiling temperature is exceeded in the vicinity of the laser, giving rise to excessive evaporation. The resulting complex melt pool and vapor dynamics can lead to process instabilities, resulting in part defects such as evaporation-induced pores, spatter, denudation, and lack of fusion. In our talk, we present a mathematically consistent multi-physics mesoscale model of LPBF that allows a detailed investigation of the interplay between dynamics of the melt pool and the vapor jet. We consider melt pool formation, melt front propagation (melting/solidification), evaporation and the formation of a vapor jet as well as the interaction of the multiphase system (solid metal, liquid metal and metal vapor). In addition to the evaporation-induced pressure jump, we resolve the evaporation-induced volume expansion and the resulting velocity jump and mass flux across the liquid-vapor interface. For capturing the a priori unknown and complex topology changes of the molten metal-gas interface, such as keyholes, gas pockets, spattering, etc., our model is formulated based on an Eulerian framework combined with a level-set-based diffuse interface tracking scheme. In this context, we address the importance of an accurate formulation of the transport velocity for the level-set in the presence of the strong evaporation-induced velocity change. The latter is mandatory to ensure mass conservation. The finite element method is employed for space discretization with an adaptively refined mesh in the interface region to guarantee a high spatial resolution. Semi-implicit time stepping schemes are used for time discreziation. Together with highly efficient matrix-free solvers based on sum-factorization techniques, our modeling framework is capable of simulating practically relevant time and length scales of the computationally demanding thermal-multiphase flow problem.