Computational Materials Science

Computational modeling is now regarded as a necessity for the advance of scientific knowledge and engineering practice.  We are interested in modeling materials processing, microstructural evolution, and phase transformations.  

Simulation of Microstructural Evolution During Alloy Solidification

Solidification is a crucial step for many manufacturing processes, such as casting, welding, and additive manufacturing. While solidification happens during processing of all types of materials, solidification of metallic alloys has been of importance to scientists and engineers. The importance comes from the fact that the solidification microstructure has a significant influence on properties of the solidified materials. In this section, some of the recent numerical studies on modeling kinetics of solidification and microstructural evolution are presented.

Modeling Columnar to Equiaxed Transition

We are developing a 3D model to simulate columnar to equiaxed transition. Some preliminary results for Inconel 718 in different temperature gradient and cooling rates are shown here. The model is based on Phase Field and Lattice Boltzmann methods

Dendrite Growth Modeling

Dendrites are the tree-like microstructures that form during solidification. They form when molten metal or alloy freezes from liquid state, which may happen during many industrial manufacturing processes. The morphology, size, and spacing between dendritic arms have a significant influence on material properties.

We developed a numerical model based on lattice Boltzmann (LB) and cellular automaton (CA) methods to simulate dendrite growth in 3D.

The low computational cost and great scalability of the LB-CA model enabled us to perform large-scale 3D simulations in macro size domains. The code was parallelized using MPI and was run on Extreme Science and Engineering Discovery Environment (XSEDE) national supercomputer network.

It is known that melt flow can significantly alter the dendrite growth kinetics by affecting solutal gradient around the dendrites. The lattice Boltzmann-cellular automaton model was extended to solve for fluid flow and count in the convection effects.

Marangoni Convection Effects on Directional Dendritic Solidification

 A series of experiments were performed on the International Space Station (ISS) through the Pore Formation and Mobility Investigation (PFMI) for succinonitrile (SCN) -0.24 wt% water binary alloy to investigate the effect of bubble presence on dendritic solidification. We developed a numerical scheme for the PFMI experiment based on Phase Field-based and Lattice Boltzmann methods to simulate the multiphase flow and bubble dynamics while a CA algorithm was employed to track the solid/liquid interface. The results show that the large induced convection can change isotherms and solute distribution which prevents the dendrites to grow in the preferred orientation and cause the dendritic arms to lean away from the original direction. This project was funded by NASA through grant number: NNX16AT75G. 

Modeling Coupled Motion and Growth of Dendrites

We are developing a model to predict the rigid body motion of dendrites under convection. This program is also written in Cuda C. Some preliminary results are shown here. This work is also  funded by NASA. 


Materials Process Modeling

 Modeling Powder Bed Fusion Additive Manufacturing

 The heat transfer and fluid flow phenomena during powder bed AM process are quite important as they determine the thermal gradients and solidification rates, and as result, the grain structure and properties of manufactured products. Due to the highly transient and localized nature of the process, direct experimental measurements of temperature and flow velocity are very challenging. Numerical methods are utilized more and more to simulate the physics of powder bed fusion AM processes. 

A 2D Phase Field-Lattice Boltzmann model was developed to simulate the solidification of the melt pool of a laser melted substrate. The program is written in Cuda C, executed by the GPU. We are currently working to extend the model to 3D.

 Effect of resistance spot welding parameters on weld pool properties
in a DP600 dual-phase steel

 The objective of this research was to quantify the effects of resistance spot welding (RSW) parameters on different weld properties of a dual-phase steel. A finite element based model was used which accounted for the following required physical interactions: the interaction between (1) the electro-kinetics and heat transfer via the Joule effect, (2) the heat transfer and phase transformations through latent heat, and (3) the heat transfer, electro-kinetics, and mechanical behavior via the contact conditions. The effects of the RSW parameters on weld properties were investigated within a design of experiments framework by altering (1) the electrical current intensity, (2) the welding time, (3) the sheet thickness, (3) the electrode face radius, and (5) the squeeze force at multiple levels. The simulation results were analyzed using the analysis of variance (ANOVA) technique to show the effects of these parameters and their potential interactions, along with their significance.

Modeling Phase Transformations During Heat Treatment of Steel Alloys

 A 3D Finite Element model was developed to simulate the hardening heat treatment process and predict the temperature history, volume fractions of various phases and hardness distribution in heat treated components.