Research
Concurrent atomistic-continuum methods
Material's response at the continuum or bulk scale is intrinsically linked to its structure and behavior at the microscale. The role of defects or microstructural features such as grain boundaries, interfaces, dislocations and voids have been investigated to great lengths in slow moving or quasi-static environments. Our work focuses on exploring the role of microstructural features on bulk response under extreme environments.
At the mechanics of materials laboratory, we are developing a novel concurrent multiscale technique to study steady state shock propagation. This technique uses co-existing atomistic and continuum domains, with the shock wave inside the purely atomistic domain. Using ideas of control volume, we allow the atomistic region to follow the shock wave in the material and tracks changes in microstructure. The formulation is being used to study the role of grain boundaries in the process of void nucleation and spallation.
Machine learning assisted mechanics & material science
Data driven methods are increasingly being used to predict material's performance as well as design innovative microstructures to maximize performance. From mechanics perspective, machine learning can help augment high-fidelity simulations to accelerate their prediction. From manufacturing and materials science perspective, machine learning methods can help predict optimal microstructures, alloy composition, and manufacturing parameters. Our work leverages machine learning techniques such as encoder-decoder networks, graph neural networks and generative adversarial networks to predict material damage and failure, and design failure resistant material microstructures. The eventual goal is to leverage existing experimental and computational data to predict material design and their performance with precision.
Fracture and fatigue failure of AM materials
Fracture and fatigue are ubiquitous phenomenon in most engineering structures, and are often the responsible mechanism for catastrophic failure. Additively manufactured (AM) composites and metallic materials are now finding applications in the aerospace, automotive, and infrastructure industries. Advances in additive manufacturing have enabled the development of lattice structures, metastructures, and bimetallic alloys for high-temperature applications.
Over the past several decades, many approaches have emerged to model and predict crack propagation. Phase field formulation of fracture mechanics allows fast and scalable simulation of crack propagation in brittle, ductile, and anisotropic settings. Our work leverages a novel numerical framework that solves, in strong form, the equations of linear elasticity on a block-structured adaptive mesh refinement (BSAMR) mesh. Using this framework, we are investigating crack propagation in brittle and ductile materials and their interaction with microstructural features such as grain boundaries and pores.
Crack interaction with inclusion resulting in partial penetration and redirection.
High strain-rate experimental mechanics
Materials behave very differently when subjected to fast (high strain-rate) loading than when subjected to quasi-static loading. Under impact or shock loading, the material is subject to strain rates ranging from 1000 to 10,000,000 /s. It is therefore important to characterize materials under these conditions.
Our work focuses on characterizing materials under high strain rate environments using Split Hopkinson Pressure Bar housed in Davis Hall and Auburn's Gas Gun housed within Center of Polymer and Advanced Composites. Using strain gauges and high-speed cameras we investigate materials such as 3D printed composites, metals and ceramics.
Planetary impact and crater formation
Asteroid impacts have been known to influence planetary formation, orbital motion, geomorphology and biological life on the planetary body. Planetary impact is an incredibly complex phenomenon. Typical impacts occur at the speeds of tens of kilometers per second with temperatures high enough to melt the target layers. The target undergoes extreme pressure and temperature changes exhibiting phase transitions, brittle fragmentation, porous compaction, plastic strain localization and melting.
Our work focuses on shallow marine impact craters, where the target was submerged under a shallow water layer. The goal is to understand how the presence of water at the time of impact influences crater shape, sediment transport, temperature profiles and tsunami generation. Two craters of primary interest are Wetumpka (AL) and Flynn Creek (TN) that were formed ~84 and ~382 million years ago respectively.
