Advance Materials & Additive Manufacturing

Dr. Trevor Fleck

Addititve Manufacturing Process Enhancement

In order for additively manufactured components to be viable for functional use, improvements need to be made in terms of process quality and reliability. This project seeks to identify process enhancements that will increase the reliability of performance through porosity reduction and increased mechanical properties. Projects focus on material extrusion additive manufacturing processes with an emphasis on Fused Filament Fabrication and Direct Write techniques.

Novel Functional Materials for Additive Manufacturing

As additive manufacturing moves beyond rapid prototyping, there is a need to tune specific material systems for specific applications. This work seeks to integrate non-traditional AM materials with material extrusion additive manufacturing processes. Material systems of interest include ceramics, energetic materials, and composite polymers.

Dr. David Jack

Ultrasonic Characterization to Identify Ply-Stack, Orientation and Type for Laminated Composites

Parts, such as an airplane wing, cannot be analyzed using destructive testing methods that cause permanent damage. Thus, a nondestructive testing method, in this case ultrasound, must be employed to ascertain the material properties. The current study utilizes ultrasonic A-scan and C-scan techniques to identify the lamina orientation in each ply of the fiber reinforced laminate. We are currently able to identify both the ply orientation and ply type (unidirectional, 2x2, twill, etc.) for 8-12 layer composite structures. We have received three different FAA 8100-9 airworthiness standards approvals for our manufacturing of coupons, ultrasonic testing of carbon fiber laminates, and the analysis of the generated data to characterize the ply orientation and stack sequence for us in engineering analysis and design. 

Planar Deposition Flow Modeling of Fiber Filled Composites in Large Area Additive Manufacturing
with Dr. Douglas Smith

Large Area Additive Manufacturing (LAAM) is a rapidly expanding 3D printing technology which is made viable by the addition of discrete carbon fibers to the polymer feedstock. The discrete carbon fibers will change the thermo-mechanical properties of printed parts. Baylor researchers have designed and fabricated a large area additive manufacturing print table with the ability to print up to 20 kg of high-strength carbon fiber reinforced material per hour. Researchers use the finite element method to predict the deposition flow along with the internal orientation state of discrete fibers from the flow field and the resulting thermo-mechanical properties of the fiber reinforced composite.

Non-Destructive Inspection Methods for Bondline Characterization for Laminated Composites for Aerospace and Automotive Applications
with Dr. Douglas Smith

Many manufactured composite laminates in the aerospace, automotive and petro-chemical industries are bonded to each other or to a structure using an adhesive.  Knowledge of the domain of the adhesive, thickness, and any defects (i.e., weak bonds or kissing bonds) is essential for confidence in the final part performance. The current study utilizes custom high-frequency, high-resolution, ultrasonic A-scan and C-scan techniques with unique in-house algorithms of the collected data to identify various features on the bondline. Baylor researchers have developed methods to quantify the spatial varying thickness of the bondline of a manufactured laminate and to detect weak bonds using non-destructive techniques with access to only a single side of the laminated structure.

Fiber Orientation Prediction Models for Fiber-Filled Thermoplastic Composites
with Dr. Douglas Smith

The purpose of this study is to use fiber orientation prediction models to optimize injection molding processing and final part dimensions to achieve the desired part mechanical properties. Our goal is to compare the predicted composite part mechanical properties derived from two separate fiber orientation models with experimental results to determine the more accurate fiber orientation model. This work to date has resulted in over 10 journal publications within the top quartile of their respective impact factor category and has caused an industry wide change in the way in which fiber orientation is analyzed. 

Fiber Reinforced Composite Structural Members Made from Recycled Polymers

This research seeks to better understand the use of recycled post-consumer/post-industrial waste composed of HDPE and GFPP in the formation of structural members for use in the rail and boating industries and the mechanisms contributing to the spatially varying material properties. Baylor researchers have modeled shell and foamed core regions using constitutive modeling for recycled composite structures and validated the four-point bend test using Finite Element Analysis. Researchers are investigating the correlation between the underlying viscoelastic/viscoplastic behavior of the constitutive materials and linking that with the microstructure variations to predict the final part performance.

Non-Destructive Evaluation and Inspection to Quantify the Internal Material Phase and Temperature Identification

Researchers have developed techniques through ultrasonic signal analysis to identify the phase and temperature of a known material on a 2D plane or in 3D space as it heats or cools and changes phase. Baylor students have created techniques to evaluate cylindrical geometries (i.e., pipelines) and approximate the internal phase and temperature of the known material as it melts. Continued work seeks to scale up to industrial simulations and analysis for complex material systems for the 3D visualization of the internal phase and temperature from exclusively obtained external measurements.

Nondestructive Testing and Evaluation for Identifying the Extent of Barely Visible Damage (BVD)

High performance composites are sensitive to low energy (tool impact, hail damage, foreign object debris, etc.) impacts significantly reducing the overall part performance. A particular challenge is the lack of certifiable methods for inspection to quantify this damage in service. Student researchers have developed a non-invasive, single side access, technique to scan carbon fiber laminates with BVD. Based on scan analysis, significant internal damage is observed within the laminate even when there is barley visible damage on the surface. Our analysis technique allows us to, without any manufacturing information, specify between which lamina the damage exists. Researchers are establishing techniques to automate the inspection process to characterized the damaged region as well as an improved analysis technique that incorporates Fourier signal analysis.

Laminate Dimensionality: Residual Stresses and Processed Part Curvature

Carbon fiber laminates are extensively used within the automotive and aerospace industries due to their high strength to weight ratios. During manufacturing residual strains are introduced due to a combination of the curing kinetics of the thermoset and the induced thermal strains due to a coefficient of thermal expansion mismatch for the resin and carbon fiber. The objective of this research is to use micromechanical theories to predict the stiffness and the coefficient of thermal expansion of an individual lamina from the constitutive properties for the fiber and the matrix, and couple the lamina results with a finite element structural and coupled thermal-structural analysis to predict the observed stiffness and the observed strain of a processed laminate. 

Field Portable Methods to Identify Foreign Objects within Laminated Composites

In the course of manufacturing laminated composites, foreign objects may be inadvertently (peel-ply, glove shards, bagging material, etc.) or purposefully (in-situ sensors, sensor arrays, secondary components, etc.) placed within a structure.  The knowledge of the dimensions and placement of these foreign objects will be of help for quality control and final part performance quantification. Researchers are seeking to determine both spatial location and depth of foreign objects within carbon fiber laminates using pulse-echo ultrasound, while studying the impact of transducer frequency on the ability to identify these foreign objects. Students have Successfully determined spatial and depth location of foreign objects as small as 1.9 mm in radius. An averaging technique has also been developed to aid a technician in capturing the foreign objects and opens the door for automation.

Modeling and Simulation the Electrical and Thermal Behavior of Carbon Nanotube Network

There does not exist a physics-based model to couple the electrical and thermal conductivity of a macro-scale network of neat single-walled carbon nanotubes (CNT), with an emphasis given to applications with large current loads. The objective of this work is to form a fundamental link between the stochastic nature of the nanostructure and the bulk response of the network, and how this coupling affects the damage mechanisms under large current loading. We have demonstrated success in: (1) Full 3-D network, (2) Coupled thermal and electrical effects, (3) Steady-state and transient responses for a variety of nanostructure configurations, and (4) Provided the first reasonable explanation for the premature observed failure mechanism. 

Growth of CNT Forests for Impact Resistant Composites

Carbon nanotube (CNT) systems when grown vertically have demonstrated unusual dynamic and degree of strain behavior that may make them applicable for impact loadings. The objective of this study is to fabricate CNT reinforced composite structures and identify their resistance to impact loadings. To date we have: (1) Fabricated CNT forests using the chemical vapor deposition process (CVD). (2) incorporated the CNT forests into classical laminated composites, and (3) have performed preliminary low velocity and ballistic impact tests with promising results for mitigating the magnitude of the impact load. 

Development of Finite Element Analysis Tool for Prediction of Cement in Casing Collapse

The tie-back region within an oil and gas well is often used to allow a continuous casing string. This additional string provides enhanced zone isolation and increased structural strength. This region is not well understood. The objective of this work is to develop an in-house thermo-structural FEA program capable of calculating the probabilistic stresses applied to a tie-back in an oilfield well to identify how an imperfect cement job affects the structural stability of the tie-back. 

Dr. Abhendra Singh

Dr. Douglas Smith

Fiber Orientation Prediction Models for Fiber-Filled Thermoplastic Composites
with Dr. David Jack

The purpose of this study is to use fiber orientation prediction models to optimize injection molding processing and final part dimensions to achieve the desired part mechanical properties. Our goal is to compare the predicted composite part mechanical properties derived from two separate fiber orientation models with experimental results to determine the more accurate fiber orientation model. This work to date has resulted in over 10 journal publications within the top quartile of their respective impact factor category and has caused an industry wide change in the way in which fiber orientation is analyzed. 

Fiber Orientation Modeling in FDM Nozzle Flow
with Dr. David Jack

The effects of nozzle shape and die swell on the orientation of fibers suspended in the polymer melt flow are evaluated for the Fused Deposition Modeling (FDM) Additive Manufacturing process. The COMSOL multiphysics program is used to simulate the flow of polymer melt in the nozzle where die swell just outside the nozzle exit is determined by minimizing free surface stresses.Fiber orientation is predicted along streamlines within the flow using Advani-Tucker orientation tensors and Tucker-Folgar isotropic diffusion with the fast exact closure. Predicted results show fiber orientation is largely dependent on nozzle convergence shape and die swell. 

Non-Isotropic Material Distribution Topology Optimization

Additive Manufacturing processes are known to result in materials with a non-isotropic response, especially when polymer composites are used in the Fused Deposition Modeling process. This project determines optimal topologies assuming that the underlying materials have a preferred, non-isotropic orientation. Results show significant changes in optimal topologies occur when oriented materials are used. 

Large Scale FDM Composite Material Deposition
with Dr. David Jack

Interest in large scale deposition with short fiber composites (i.e., nozzle exit diameters larger than ¼ inch) has resulted in a need to understand nozzle flow, deposition geometries, and fiber orientation in this process. This project focuses on depositing large beads of fiber filled materials on a controllable moving platform to measure various parameters important to 3D printing of these materials. 

Micro-Mechanics Modeling of FDM Composites
with Dr. David Jack

The effect of carbon fiber filler in polymer composites significantly effects the mechanical properties of the final part, particularly when used in the FusedDeposition Modeling process. This project looks at the micro-mechanical behavior using representative volume elements (RVE) with the finite element method (FEM). Of particular interest is computing effective elastic and viscoelastic properties, as well as effective thermal expansion coefficients and residual stress. The inclusion of fiber filler and voids are considered. 

Mechanical Evaluation of Carbon Fiber Filled FDM Components

Adding carbon fiber to ABS or PLA for use in desktop fused deposition modeling is expected to increase the stiffness and strength of the additive manufacture part. This project quantifies the benefit of adding fibers by characterizing the fiber content and measuring material properties for fiber filled FDM parts produced with various deposition path orientations. 

Polymer Flow Processing Design

Design optimization and Design Sensitivity Analysis are developed for molding and extrusion-based polymer processing. Simulations are performed with the finite element method and design derivatives needed for the optimization are computed with the adjoint variable and direct methods. A unique design approach is developed that determines optimal mold cavity and flat die designs where special attention is given to obtaining a design that is optimal over various operating temperatures and with various materials. 

Optimization-Based Inverse Heat Transfer

An optimization-based approach is developed for computing effective surface heat transfer coefficients for components manufactured with the quench heat treatment process. Measured temperature histories serve as input to the process where the difference between these values and those predicted with an ABAQUS-based finite element model are minimized by adjusting the surface heat transfer coefficients. This approach is unique in that phase transformation kinetics are included in the simulation, and the surface heat transfer is parameterized as a function of both time and temperature. 

Computing Eigenvalue and Eigen Vector Design Derivatives

A novel design sensitivity analysis method is developed for computing design derivatives of eigenvalues and eigenvectors where the mode shape normalization condition is used to bring the system of equations to full rank, making it possible to uniquely define the desired derivatives. The effect of rescaling the eigenvectors is also considered, and the popular method by Nelson (1972) is shown to be a subset of this broader approach.

Total Hip Anthroplasty Simulation

A patient specific finite element approach is developed for performing a mechanical simulation of hip cup insertion for THA. Patient specific geometry is obtained from CT scans which is used to generate the complex 3D geometry of the hip. Finite element models are then generated for the acetabulum region of the hip and the hip cup which is inserted as part of the simulation. Stresses and deformations are computed in the FE approach. 

SI Joint Modeling

Patient specific SI joint cartilage models are developed to study stresses and deformations that cause lower back pain. Plaster models are created from cadavers which are scanned to obtain an accurate 3D representation of the SI joint surface. These surface models are then used to create patient specific finite element models of the cartilage. These models are used to predict stresses and contact pressures in the cartilage at various angles of loading.