Multi-scale modeling of transport phenomena and microstructure development in laser keyhole welding

Wenda Tan, Purdue University

Abstract

This study is focused on a multi-scale investigation of laser keyhole welding processes based on numerical modeling techniques. ^ A macro-scale model is developed to simulate the dynamics in keyhole and molten pool. The model accounts for the multi-phase and multi-physic phenomena, and is able to capture the transient motion of and complex boundary conditions on the keyhole wall. The model is applied to the keyhole welding processes with pulsed and continuous wave lasers, and the issues of laser energy distribution, keyhole growth and fluctuation, and multi-phase interaction are studied. ^ Micro-sale models are also developed to simulate the growth of grains and sub-grain dendrites during the solidification process. The grain growth model predicts the competitive grain growth in the 3D molten pool with complex geometry. The dendrite growth model integrates the Cellular Automata (CA) and Phase Field (PF) methods, and is able to efficiently produce quantitative prediction of dendrite morphology in the grains. ^ The multi-scale model is used to assist the investigation on laser keyhole welding of a stack of multiple stainless steel thin plates, which is an important process for fuel cell manufacturing. The processing- microstructure relationship is predicted through numerical modeling and validated against experiments, and the mechanical properties as well as corrosion resistance in thefuel cell environment of the welds are also tested. ^ The multi-scale model in its present form is geared towards laser welding processes of austenitic stainless steel, but it can be modified for other engineering materials and/or other manufacturing processes with free surface motion and multi-phase fluid mechanics and heat transfer.^

Degree

Ph.D.

Advisors

Yung C. Shin, Purdue University.

Subject Area

Engineering, Mechanical|Engineering, Materials Science

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS