A Coupled DEM/SPH Computational Model to Simulate Microstructure Evolution in Ti-6Al-4V Laser Powder Bed Fusion Processes

Sharen Cummins,Chris Davies,Paul W Cleary,Gary Delaney,Dayalan Gunasegaram,Matthew Sinnott,Arden Phua

doi:10.3390/met11060858

Abstract

A new multi-stage three-dimensional transient computational model to simulate powder bed fusion (L-PBF) additive manufacturing (AM) processes is presented. The model uses the discrete element method (DEM) for powder flow simulation, an extended smoothed particle hydrodynamics (SPH) for melt pool dynamics and a semi-empirical microstructure evolution strategy to simulate the evolving temperature and microstructure of non-spherical Ti-6Al-4V powder grains undergoing L-PBF. The highly novel use of both DEM and SPH means that varied physics such as collisions between non-spherical powder grains during the coating process and heat transfer, melting, solidification and microstructure evolution during the laser fusion process can be simulated. The new capability is demonstrated by applying a complex representative laser scan pattern to a single-layer Ti-6Al-4V powder bed. It is found that the fast cooling rate primarily leads to a transition between the β and α martensitic phases. A minimal production of the α Widmanstatten phase at the outer edge of the laser is also noted due to an in situ heat treatment effect of the martensitic grains near the laser. This work demonstrates the potential of the coupled DEM/SPH computational model as a realistic tool to investigate the effect of process parameters such as powder morphology, laser scan speed and power characteristics on the Ti-6Al-4V powder bed microstructure.

Highlights

Additive manufacturing (AM) based on powder bed fusion (L-PBF) has brought impressive advances in the manufacture of bespoke parts with complex geometries
The difficulties arise from the fact that the AM process is a sum of several sub-processes that occur at different length and time scales and are governed by different physics
The complex dynamics of the powder flow and the ability of the discrete element method (DEM) model to capture the detailed interactions with the rake geometry including fine details of the flow of material between the rake teeth can clearly be observed

Summary

Introduction

Additive manufacturing (AM) based on powder bed fusion (L-PBF) has brought impressive advances in the manufacture of bespoke parts with complex geometries It poses many technical barriers due to highly transient and varying physical phenomena which occur on a broad range of length and time scales and are difficult to observe and characterize [1]. One of these key challenges is the ability to predict and control the microstructure, and the component’s mechanical properties during a L-PBF process. In AM, microstructure may be modified during subsequent reheating phases as powder layers above are melted These processes occur at different length and time scales and require different computational techniques. The use of the word ‘particle’ is reserved for the basic computational element of the melt pool model.)

Objectives

Methods

Results

Discussion

Conclusion