Testing Of Fluid-Structure Interaction For Simulating Convection In Shape Memory Alloy Constitutive Models

REU student: Stephen Oehler

Post Doctoral co-advisor: Darren Hartl

Faculty Mentor: Dimitris C. Lagoudas

The finite element techniques for modeling shape memory alloy (SMA) constitutive behavior involve, to a large degree, heat transfer mechanisms such as conduction and convection. During a typical isobaric SMA test, the alloy is heated through either surface conduction or resistive heating and is cooled by convection. Conduction coefficients are generally easy to determine from experiments and, as a result, are relatively simple to model. On the other hand, convective behaviors are more difficult to determine from experimentation due to a combination of factors. At best, convection coefficients can be estimated from surface temperature-time curves generated from experimental data. These estimates are almost always highly approximated and can- not fully capture the combined effects of influential parameters such as surface roughness and fluid viscosity, velocity, and turbulence. As a result, using convection coefficients can prove to be a difficulty in modeling realistic cooling rates of SMAs. This report discusses the possibility of using alternative methods to calculate heat transfer by convection. A fluid-surface interaction code recently integrated into the ABAQUS FEA Suite is tested for reliability in calculating surface heat flux based on established fluid and solid material properties. The outcome of these current and future tests will determine the feasibility of using multiphysical solutions to eliminate the need for estimating a convection coefficient when modeling the cooling phase of SMAs.

Advanced Modeling of Un-inhabited Aerospace Vehicles

REU student: Kyle Benson

Faculty Mentor: Dr. John Whitcomb

Finite element analysis (FEA) for large structures costs large amounts of computer resources and time when solid elements are used. This project studies the use of 2D shell elements for analyzing 3D structures to reduce the number of elements required for accurate results in order to reduce computation costs. The object of the study is a wing structure from a Un-inhabited Aerospace Vehicle (UAV). Using shell elements greatly reduces the number of elements required to perform the analysis. Reducing the number of elements also reduces the analysis run time by 40%. The inaccurate results predicted by the shell I-beam and wing in SolidWorks indicates that the shell elements are not properly connected or that the formulation used by SolidWorks to join incompatible meshes is nor robust. Regardless, the availability of other errors enabled identification of errors. There were no obvious artifacts in the deformed shapes or stress contours that would indicate a faulty prediction. This serves as a warning to exercise great care in using computational tools.

Multifunctional Solid Rocket Propellant/Supercapacitors

Faculty mentor: Dr. Jim Boyd

REU Student: Mitch Pace, Texas A&M University, Aerospace Engineering

Modeling Plasma Flows in Irregular Geometry and Associated Boundary Conditions

REU student: Travis Sikes

Faculty mentor: Dr. Jacques C. Richard

Computational Fluid Dynamic simulations often call for irregular meshes, and incorporating these meshes into computations is quite a challenge. This research created a method to transform an irregular orthogonal hexahedral mesh into computational space for use with finite difference algorithms. Creating this method of transforming, from an irregular hexahedral mesh into computational space, is a first step to simulating flows over a variety of surfaces. The method involves creating a model and transforming it by a Jacobian matrix. This method works for surfaces that one can model using hexahedral meshes; however, it does have limitations because some surfaces are unable to be modeled using hexahedral meshes. The driving factor to how well the method transforms the mesh into computational space for simulations is the quality of the mesh. Therefore, it is advisable to use the best hexahedral meshing software available.

Modeling Dielectric Barrier Discharges

REU Student: Yawo Semanu Ezunkpe, San Jose State

Faculty mentor: Dr. Jacques C. Richard

Voltage and current variations show the basic principle of plasma enhanced chemical vapor deposition (PECVD) using simple parallel plates in dielectric barrier discharge (DBD). A parallel experiment validates the simulation outcome. The electric and velocity fields are simulated using the Lattice Boltzmann Model (LBM) then visualized. The results reveal that the electric field has the same direction as the current and opposite direction to an electron’s motion, which follows the velocity vector field. The parallelism of the electrode plates affects the way the discharge occurs. It appears that increasing potential results in increasing current therefore accounting for the power augmentation. The gap between the electrode plates has a major impact on the discharges and the charged species are the major elements in the PECVD.

The electric field and transverse velocity component (right) that should be small but any light perturbation or misalignment from plates being exactly parallel induces discharges at a corner.

Utilizing Hydrothermal Reactions to Optimize Crystallinity of Metal Organophosphate Porous Materials; Achieving Regularity in Porosity for Aluminum Complexes

REU student: Nancy Garcia

Graduate student: Tiffany Kinnibrugh

Faculty Mentor: Dr. Abraham Clearfield

The synthesis of metal organophosphonates have a variety of potential applications, including proton conductors, ion exchangers and gas absorption. With the goal of obtaining high surface area and porosity, hydrothermal reactions enabled obtaining aluminum and monovalent metal materials. Microprobe and elemental analysis suggest the two products synthesized had chemical formulas Al4(O3PC12H8PO3)3·8.75H2O and  Al4(O3PC12H8PO3)2.29 (HPO3)1.42·10H2O. Surface area measurements indicate an increase in internal surface area with the addition of phosphorous acid. X- ray powder diffraction and NMR data suggest layer material with two types of Al geometry. Optimization of crystalline properties and utility will require further variations of the reactions.

SEM images of a compound at two different magnifications showing particles of varying sizes and shapes, which does not indicate regular or repeating structural components.

Energy Harvest from Frictional Heating caused by Polymer Wear

REU Student: Alejandro Camou

Faculty mentor: Dr. Christian Schwartz

Self-lubricating polymer bearings offer an advantage to industry by eliminating the need for additional lubricants. High-performance plastics such as polyetheretherketon (PEEK), polyphenylene-sulfide (PPS) and ultra-high-molecular-weight-polyethylene (UHMWPE) are commonly used for polymer bearings. Heat is dissipated within these bearings due to friction and also affects the polymer’s wear mechanism. A method has been proposed to harvest this heat energy by applying thermoelectric theory to generate power. The thermoelectric power generation can potentially be used to power an internal cooling system to maintain the polymer’s tribological performance and film-transfer ability. The polymer-metal sliding process was replicated using a dual-axis wear tester. An epoxy resin was used as an alternative to one of the high-performance plastics in order to easily cast samples in a timely manner. The polymer’s surface temperature was observed to have an initial transient behavior and was modeled to solve the heat transfer boundary conditions within the system. The predicted temperature gradient was used to measure the heat input for thermoelectric power generation. Results show that the temperature model accurately replicates the polymer’s surface temperature during sliding based on material and external parameters such as specific heat, thermal conductance, density, pressure, velocity and time. This temperature model can be used to model the sliding temperature and estimate thermoelectric power. Cooling systems for polymer bearings can be appropriately designed based on the thermoelectric power generated from the frictional heat.