arXiv subjects

Branden B. Kappes

Publications and source records attributed to Branden B. Kappes.

4 recordsLinked to original sources

arXiv2020

Physics-informed machine learning for composition-process-property alloy design: shape memory alloy demonstration

Machine learning (ML) is shown to predict new alloys and their performances in a high dimensional, multiple-target-property design space that considers chemistry, multi-step processing routes, and characterization methodology variations. A physics-informed featured engineering approach is shown to enable otherwise poorly performing ML models to perform well with the same data. Specifically, previously engineered elemental features based on alloy chemistries are combined with newly engineered heat treatment process features. The new features result from first transforming the heat treatment parameter data as it was previously recorded using nonlinear mathematical relationships known to describe the thermodynamics and kinetics of phase transformations in alloys. The ability of the ML model to be used for predictive design is validated using blind predictions. Composition - process - property relationships for thermal hysteresis of shape memory alloys (SMAs) with complex microstructures created via multiple melting-homogenization-solutionization-precipitation processing stage variations are captured, in addition to the mean transformation temperatures of the SMAs. The quantitative models of hysteresis exhibited by such highly processed alloys demonstrate the ability for ML models to design for physical complexities that have challenged physics-based modeling approaches for decades.

cond-mat.mtrl-sci↗

arXiv2019

Knowledge of Process-Structure-Property Relationships to Engineer Better Heat Treatments for Laser Powder Bed Fusion Additive Manufactured Inconel 718

Dislocation structures, chemical segregation, {γ^{\prime}, {γ^{\prime \prime}}, δ precipitates and Laves phase were quantified within the microstructures of Inconel 718 (IN718) produced by laser powder bed fusion additive manufacturing (AM) and subjected to standard, direct aging, and modified multi-step heat treatments. Additionally, heat-treated samples still attached to the build plates vs. those removed were also documented for a standard heat treatment. The effects of the different resulting microstructures on room temperature strengths and elongations to failure is revealed. Knowledge derived from these process structure property relationships was used to engineer a super solvus solution anneal at 1020 degC for 15 minutes, followed by aging at 720 degC for 24 hours heat treatment for AM-IN718 that eliminates Laves and δ phases, preserves AM specific dislocation cells that are shown to be stabilized by MC carbide particles, and precipitates dense {γ^{\prime} and {γ^{\prime \prime}} nanoparticle populations. This 'optimized for AM-IN718 heat treatment' results in superior properties relative to wrought/additively manufactured, then industry standard heat treated IN718: relative increases of 7/10 percent in yield strength, 2/7 percent in ultimate strength, and 23/57 percent in elongation to failure are realized, respectively, regardless of as-built vs. machined surface finishes.

physics.app-ph↗

arXiv2011

Reactive force field for lithium-aluminum silicates with applications to eucryptite phases

We have parameterized a reactive force field (ReaxFF) for lithium aluminum silicates using density functional theory (DFT) calculations of structural properties of a number of bulk phase oxides, silicates, and aluminates, as well as of several representative clusters. The force field parameters optimized in this study were found to predict lattice parameters and heats of formation of selected condensed phases in excellent agreement with previous DFT calculations and with experiments. We have used the newly developed force-field to study the eucryptite phases in terms of their thermodynamic stability and their elastic properties. We have found that (a) these ReaxFF parameters predict the correct order of stability of the three crystalline polymorphs of eucryptite, α, β, and γ, and (b) that upon indentation, a new phase appears at applied pressures \geq 7 GPa. The high pressure phase obtained upon indentation is amorphous, as illustrated by the radial distribution functions calculated for different pairs of elements. In terms of elastic properties analysis, we have determined the elements of the stiffness tensor for α- and β- eucryptite at the level of ReaxFF, and discussed the elastic anisotropy of these two polymorphs. Polycrystalline average properties of these eucryptite phases are also reported to serve as ReaxFF predictions of their elastic moduli (in the case of α-eucryptite), or as tests against values known from experiments or DFT calculations (β- eucrypite). The ReaxFF potential reported here can also describe well single-species systems (e.g., Li-metal, Al-metal, and condensed phases of silicon), which makes it suitable for investigating struc- ture and properties of suboxides, atomic-scale mechanisms responsible for phase transformations, as well as oxidation-reduction reactions.

cond-mat.mtrl-sci↗

arXiv2010

Bandgap opening in metallic carbon nanotubes due to silicon impurities

Controlling the bandgap of carbon nanostructures is key to the development and mainstream application of carbon-based nanoelectronic devices. We report density functional theory calculations of the effect of silicon impurities on the electronic properties of carbon nanotubes (CNTs). We have found that Si adatoms open up a bandgap in intrinsically metallic CNTs even when the linear density of Si atoms is low enough that they do not create a bonded adatom chain. The bandgap opened in metallic CNTs can range between 0.10 eV and 0.47 eV, depending on adsorption site, linear density of Si adatoms, and CNT chirality.

cond-mat.mtrl-sci↗