Summer 2011 Research and Participants

Analyzing Dispersion of Single Walled Carbon Nanotubes in Epoxy by Using Raman Spectroscopy and by Measuring Transmittance

REU Student: Rooservelt Akume, Texas Tech University

Faculty Mentor: Dr. Dimitris Lagoudas

In this experiment, a non-destructive, semi-quantitative method [1] of analyzing the dispersion of Single Walled Carbon Nanotubes (SWCNTs) in SWCNT-epoxy composites is presented. Samples of different weight fractions and pre-curing times are fabricated. Four testing methods are used to examine the samples. Optical microscopy is used to obtain sample images of SWCNT dispersion at macro scale, UV-Vis spectroscopy is used to measure absorbance and transmittance of samples.  Raman spectroscopy is used to obtain a spectrum of the samples. The area under the G-band peak is used to semi-quantify dispersion. Raman mapping technique is used to obtain dispersion images of the composite samples at a resolution of 3μm. The UV-Vis results show that at lower weight fractions, pre-curing the samples leads to increase in absorbance with an accompanying decrease in transmittance. At higher weight fractions, increase in absorbance due to pre-curing is lower. The results from UV-Vis also show increase in absorbance with increasing weight fractions, in accordance with Beer-Lambert Law. Results from Raman and optical microscopy images show improved dispersion with increase in pre-curing time. Results from Raman spectroscopy show that G-band peak area increases with increase in pre-curing time.

Vertical and Rotational Motion of a Wing with Spring and Damper

Faculty Mentor: Dr. James Boyd

REU Student: Keegan Colbert, Texas A&M University

This project was performed to determine an optimal design for an experiment that tests for the lift, drag, and moment coefficients of a wing body and the spring constants and damping coefficients of a linear and rotary spring and dashpot restricting its motion. The wing will be placed in a wind tunnel in which three scenarios will be observed. The first scenario requires manual control of the deflection of the wing’s flap and the measurement of the wing’s angular displacement while vertical displacement is prevented. The second scenario requires the manual control of the pitch angle of the wing and the measurement of the resulting vertical displacement with the flap angle held at zero degrees.  The third scenario requires the manual control of the flap angle and the measurement of the resulting vertical and angular displacement of the wing. Also included in this project, the design and construction of a wind tunnel was necessary to run tests on the wing and obtain the aforementioned measurements. The design of the wind tunnel was constrained by the need for a test flow of approximately 30 mph and the need to mount the wing body inside the test section while providing adequate space for wing rotation and flap deflection. With these constraints accounted for, the test setup could be designed and manufactured.

The Effect of Temperature and Maximum Applied Stress on Residual Strain Accumulated During Pseudoelastic Cycling of NiTi SMA Wires

Faculty Mentor: Dr. Dimitris Lagoudas

REU Student: Kenneth Cundiff, Texas A&M University

The goal of this study is to determine how test temperature and maximum applied stress during pseudoelastic cycling influence the residual strain of a shape memory alloy (SMA). This study will expand the test matrix of a previous study conducted by Celia Caer. Nickel-titanium (NiTi) SMA wires were held at a constant temperature while cyclically loaded up to 1100 MPa and the residual strain after cycling was recorded. The SMAs wires were then flash heated in order to recover any strain due to retained martensite in the wire, and the remaining strain on the wire is due to plastic strains. The conclusion of the previous study was that the strain due to retained martensite is completely determined by the maximum stress level, while the amount of plastic strain is determined completely by the transformation plateau stress level of the SMA, which is determined by the test temperature. Using data gathered from the temperatures and stress levels from the previous study and from this study, this conclusion will either be supported or revised.

Influence of molecular architecture and structure on thermal degradation of PAEK’s in processing temperature range

REU Student: Nangelie Ferrer, Massachusetts - Amherst

Faculty Mentor: Dr. H.J. Sue

Poly(aryl ether ketone)s (PAEKs) are a class of high performance, semi-crystalline engineering thermoplastics of great commercial importance. PAEKs’ excellent mechanical properties, chemical resistance, and thermal stability make them appealing for use in an array of demanding engineering applications. Conversely, Poly(ether ether ketone)’s (PEEK)’s thermal stability is questionable within its processing temperature range, 400 - 420°C, namely, temperatures well above its melting temperature, 340°C. Thus, it is of great importance to identify processing conditions which do not result in PEEKs’ molecular degradation as measured by changes in the physical properties, characteristics of time-dependent thermal degradation at different temperatures, and stress relaxation behavior. Degradation of PEEK samples were introduced by Thermogravimetric Analysis (TGA), then monitored via Differential Scanning Calorimetry (DSC). Identification of the crystalline domains and the onset of thermal degradation was performed using a melt-state rheological method by verifying that all data superimposed when plotted as complex modulus, G*, vs. phase angle, δ (van-Gurp Palmen plot) and storage modulus, G’, vs. loss modulus, G’’ (Cole-Cole plot).  This approach verifies that no transition in relaxation behavior occurs due to either semi-crystalline domains or thermal degradation. All grades of PEEK show significant degradation in air. The lower molecular weight grades of PEEK retain the heat of melting and crystallization, suggesting more resistance to thermal degradation. These observations are associated with the percent of crystallinity in the material. The rheological approach used to represent the relaxation behavior of one of the PEEK grades, VK150g , suggests that the onset of thermal degradation occurs at temperatures much lower than the actual processing temperatures for PEEK. Moreover, it was found that the ideal temperature range for this grade of PEEK is 370 - 380°C.  Knowledge of the thermal degradation of PAEKs within its processing temperature range will be of interest to industry to aid in determining suitable processing conditions for each distinct material to avoid undesirable changes in the structural properties as a result of thermal degradation.  Similarly, it is of interest to academia as this work may set the baseline for future nanocomposite work using these high performing materials.

Computer Simulation of Current Loop Magnetic Field Effects on Plasma Jets used in Plasma Enhanced Chemical Vapor Deposition

REU Student: Nicholas Gawloski, Texas A&M University

Faculty Mentor: Dr. Jacques Richard

The Plasma Enhanced Chemical Vapor Deposition (PECVD) process deposits a thin film onto a substrate using a plasma jet. This process utilizes a jet, between 300° - 350° C, a so-called “cold plasma,” lowering the overall temperature for typical chemical vapor deposition from a range of 650° - 850° C. Accurate computer simulations of the plasma flow are very important to aid in understanding the plasma’s interactions with applied magnetic fields. The Lattice Boltzmann Method (LBM) solves for a distribution of molecular velocities in a magnetohydrodynamic (MHD) flow yielding a visual representation of the plasma flow field. In order to accurately model the geometry of the PECVD process, an existing MHD-LBM code was modified with new boundary conditions. Plasma flow simulations were then run with various input parameters and the resulting flow fields show physically accurate data.

Wash Cycles for Gas Barrier Layer-by-Layer Assemblies

REU Student: Stephen Greenlee, Texas A&M University

Faculty Mentor: Dr. Jaime Grunlan

Gas barrier nanocomposites are grown using a multi-axis positioning robot with an automated wash and dry cycle. While the older version of a wash cycle was able to clean the films effectively, it also washed away the deposited layers on certain substrates. Using ellipsometry it was determined that there could be an error of up to 5% on the thickness of the film. The surface finish was also damaged, causing the film to lose its transparent quality. From this it was determined that a softer, even wash was to be established that also was able to fit into the current design of the robot. In order to create a rinse that would not damage the film a curtain of water would be constructed that held a continuous stream. When a water stream forms separate particles, each particle impacts the film with its own energy which can damage areas of the film and cause an irregular finish. When a continuous stream is used on the wash cycle, the water flows across the film rather than impacting it, causing a more even, cleaner film. The development of this new wash cycle went through multiple different phases, and had a total of four different designs before the final product. The finished product created a curtain by first filling up a cavity in a block of acrylic with water, then forcing the water through a narrow slit in the side. This new washing mechanism will create films that will be more easily measured and researched.

Characterization of Shape Memory Alloys Using Artificial Neural Networks

REU Student: James Henrickson, Texas A&M University

Faculty Mentor: Dr. John Valasek

Shape-memory alloys have in recent years been shown to be an attractive lightweight alternative to traditional actuators due to their unique ability to recover high levels of plastic strain. Practical implementation of shape-memory alloys, however, is often complicated by their hysteretic, non-linear, thermo-mechanical behavior. Existing constitutive models, although largely accurate in predicting this hysteretic behavior, require thorough characterization of the material. The current characterization procedure requires an extensive experimentation process in which many parameters must be carefully identified. This paper develops a novel method in which an Artificial Neural Network is trained to identify difficult-to-characterize parameters of a given shape-memory alloy specimen using relatively easy-to-identify parameters as inputs. An Artificial Neural Network, a branch of machine learning, is an adaptive computational model that simulates the human brain by training a number of artificial neurons to identify patterns within given data. The result is a non-parametric, black box model of the given training data. This paper outlines a process in which the Hartl-Lagoudas model was implemented to generate temperature-strain plots for a number of theoretical shape-memory alloys. This data was then used to train an artificial neural network, which was in turn used to identify parameters for a number of randomly generated theoretical shame-memory alloys. Results presented in the paper show that a comparison of the values predicted by the artificial neural network with the known target values found that the artificial neural network was able to predict both transformation temperatures and stress influence coefficients within the bounds of typical experimental error.

Thermal and Mechanical Modeling of a Conceptual Shape Memory Alloy (SMA) Aircraft Rotary Actuator

REU Student: William Jinkins, Texas A&M University

Faculty Mentor: Dr. Dimitris Lagoudas

The use of active materials, especially shape memory alloys (SMAs), as aerospace actuators has increased in industry. SMAs subjected to temperature change are able to generate motion while loaded. For example, SMA torque tubes can provide rotational motion. However, relatively long heating times limit the use of SMA actuators to low frequency applications. This is not ideal for aerospace applications, which require faster deployment times than those generally achieved by SMA actuators. The purpose of this study is to examine the effect of the heater array geometry and power configurations on actuation time.

This work presents a computational study of a rotary torque tube actuator and examines the effect of heater size, spacing and power on the actuation time and rotation. The analysis includes a full model including a heater array, an SMA torque tube, an aluminum housing structure, and a thermally conductive layer to promote conduction between structural components. The study involves computing the actuation rotation of a loaded thermally cycled SMA torque tube considering various geometric and power configurations using an accurate constitutive model implemented in a finite element framework. This set of analyses considers the solution to a transient thermomechanically coupled problem. A parametric study is used to find trends related to actuation time. The study shows trends in the sizing of the components that suggests a minimum time to actuation. The research demonstrates that it is possible to increase the actuation rate by changing component geometry to system power inputs.

Optimizing Fire Retardant Clay – Chitosan Coatings for Polyurethane Foam

REU Student: Christopher Kirkland, Texas A&M University

Faculty Mentor: Dr. Jaime Grunlan

Fireproofing normally flammable objects such as fabrics and foams typically requires hazardous chemicals that are now being phased out of production to protect consumers. In their place, new fire-retardant coatings are being developed that are more environmentally friendly and still effective at fire protection. Chitosan, a natural polymer produced from crustacean shells, and montmorillonite clay can be combined into nanoscale coatings through layer-by-layer assembly to reduce the flammability of foams and fabrics. Layer-by-layer assembly works by coating positively and negatively charged chemicals onto a surface to develop a protective shell; this "nanobrick wall" cuts smoldering time, weakens flames, and preserves more of the burned object intact. This in turn gives consumers more time to escape fires and results in less property damage afterwards.

Novel Fabrication Techniques for High Performance Composites

REU Student: Jeanne Methel, Purdue University

Faculty Mentor: Dr. John Whitcomb

Conventional fabrication techniques are currently not adequate to fabricate small-scale high performance composites that meet desired geometric and structural requirements. Since these composite structures are often tailored to a specific application, novel fabrication techniques need to be developed to achieve accurate centimeter-scale results. This project concentrates on establishing such fabrication techniques for composite materials intended for micro air vehicles (MAV), in order to: 1) replicate a dragonfly’s wing in shape and scale in the context of biomimetics and 2) manufacture an isogrid cylinder. Different fabrication methods were evaluated qualitatively as well as by checking for even placement of the fibers and of the desired width distribution across the structures. Some experimental methods were refined but all are documented and presented with their advantages and disadvantages. Particular characteristics of these methods are that they can be applied to any structure similar to the ones mentioned above with different customized requirements while using only commercially off the shelves materials and tools.

Finding Symmetry in Quantum Turbulence

REU Student: Cassandra Oduola, Texas Southern University

Faculty Mentor: Dr. Jacques Richard

With the onslaught of natural disasters involving wind and water, the scientific community has tried to refine methods of predicting hurricanes, tornadoes and tsunamis before it is too late. These phenomena are all associated with turbulence. Turbulence is an unpredictable chaotic phenomenon in nature. This type of turbulence has also been observed at the quantum physical scale. This research employs the non-linear Schrödinger coupled with Poisson’s equation for two dimensional quantum turbulence simulations. Research has found evidence of soliton solutions to the non-linear Schrödinger coupled with Poisson’s equation. Solitons are self-reinforcing waves in nature that are also symmetric. Therefore symmetry can be found in quantum turbulence.

Improving Dispersion of Carbon Nanotubes in Epoxy for Increased Electrical Conductivity

REU Student: Vipul Patel, University of Houston

Faculty Mentor: Dr. Dimitris Lagoudas

Carbon nanotubes (CNTs) have van der Waals interaction, and it is difficult to separate and individually disperse them in an epoxy matrix. It is best for the CNTs to be individually dispersed to take advantage of their unique properties (i.e. mechanical, thermal, and electrical properties). My goal is to improve this dispersion in order to get increased electrical conductivity. I have implemented different fabrication techniques and analyzed their samples using optical microscopy and the dielectric spectrometer. In the end, the percolation threshold, which is the concentration of CNTs where there is a significant increase in electrical conductivity, is determined using the optimal fabrication technique. This increase in electrical conductivity is seen due to the formation of a conductive path of CNTs throughout the epoxy matrix, which is non-electrically conductive if examined without CNTs. Similarly, typical percolation concentrations vary anywhere from less than 1% to over 10%; however, a low concentration of CNTs is desired for economy.

Development and Characterization of Multifunctional Structural Epoxies for Aerospace Composites

REU Student: Joseph Pickard, Northwestern College

Faculty Mentor: Dr. H.J. Sue

In order to improve the transport properties of an aerospace-grade epoxy resin without degrading the mechanical properties, we have integrated carbon nanotubes with different chemical and non-chemical dispersion approaches. Initial specimens were produced using well established fabrication techniques which have been shown to enhance fracture toughness and electrical conductivity of epoxy. Solvent-free dispersion methods were also developed. Samples were characterized for thermal conductivity and electrical conductivity. Additionally, dynamic mechanical analysis was also performed to confirm that glass transition temperature was maintained or enhanced. Innovational chemical functionalization methods were pursued in order to enhance dispersion of the CNTs and promote interfacial adhesion with the epoxy matrix.

Cross-sectional Shape Dependence at Nanometer Scale

REU Student: Damaris Valero-Diaz, University of Puerto Rico

Faculty Mentor: Dr. Tahir Cagin

It is expected that for macro scale pillars the shape of their cross-section has no impact on their behavior when referring to a stress that is being applied to it, but this could differ for nano scale pillars. This knowledge will help understand better the behavior of materials and their mechanical properties, such as metal, at nano scale. This research consisted of analyzing the behavior of palladium nano pillars in response to an applied tensile strain. Palladium nano pillar models with the same length, but with different cross-sectional shape were made, maintaining the cross-sectional area constant for an accurate comparison. Molecular dynamic simulations of the same applied tensile strain were carried out for each of the models. Data from the simulations was plotted for better analysis of the nano pillars mechanical properties.