Integrated Experimental and Computational Approach for Heterogeneous Microstructures of Martensitic Functional Gradients
Restricted (Penn State Only)
- Author:
- Britt, Cole
- Graduate Program:
- Materials Science and Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 16, 2024
- Committee Members:
- John Mauro, Program Head/Chair
Zi-Kui Liu, Chair & Dissertation Advisor
Jay Keist, Chair & Dissertation Advisor
Hojong Kim, Major Field Member
Allison Beese, Chair & Dissertation Advisor
Todd Palmer, Outside Unit & Field Member - Keywords:
- additive manufacturing
functionally graded materials
CALPHAD
high-strength low-alloy steels
stainless steel - Abstract:
- Functionally graded materials (FGMs) are material systems in which the structure or the chemical composition varies spatially within a single component to achieve a set of desired functions. Changing composition throughout a component requires advanced manufacturing techniques with a high amount of design freedom such as directed energy deposition (DED) additive manufacturing (AM). DED AM is a fabrication process in which metallic powder feedstock is deposited layer-by-layer using a focused energy source such as a laser until a bulk part is achieved. For an FGM, the volume fractions of different feedstock powders are varied as a function of position as the powder is deposited to achieve a desired compositional gradient. Martensitic steels are of specific interest for FGMs to leverage their high strength and overcome their poor corrosion resistance by grading to a more corrosion resistant material such as stainless steel. However, thermal histories can vary greatly within a part, from location to location, as thermal cycles fluctuate based on scan strategy and the time it takes to complete each layer based on component size and geometry, leading to inconsistent carbide precipitation and resulting properties throughout a martensitic steel. FGM processing can further compound on these variations depending on optimization of constituent alloys, such as power, speed, and scan strategy. These complex thermal histories and variable microstructures, further complicated by the impacts of FGM processing, call for the need of computational tools to understand and achieve desired properties. This work presents the use of simulations that established the effect of processing conditions and composition by spatially affecting thermal history and resulting microstructures and properties in DED AM high-strength low-alloy steels (HSLAs) and FGMs. Monolithic and functionally graded martensitic systems were fabricated and characterized with x-ray diffraction, microscopy, electron back-scatter diffraction and hardness testing to observe composition, microstructure, retained austenite, carbide precipitation, and mechanical properties. Hardness and microstructural variation within regions of uniform composition in an FGM suggested varying precipitation due to locally differing thermal histories during fabrication. A schema of temperature history to describe precipitation was proposed for locations throughout the system. An integrated experimental and computational approach was devised and applied beginning with a monolithic system followed by a complete FGM. Temperature simulations were calibrated with in-situ thermocouple data and used as inputs for thermodynamic and kinetic models to describe experimentally obtained heterogeneous microstructures and hardness. Kinetically generated time-temperature-transformation (TTT) diagrams successfully portrayed microstructures with different thermal histories and compositions. Microsegregation was modeled and demonstrated how a difference in cooling rate can affect local martensite start temperatures and shift phase transformations to longer times. Precipitation was simulated to estimate resulting matrix carbon content. Results from the simulations and experiment were then utilized to describe properties using a strength contribution-based hardness model to compare experimentally measured hardness values.