The Tenth Annual NC State University
Undergraduate Research Symposium

 Abstracts are listed in alphabetical order by the last name of the corresponding author.

Student Author(s):	Barlow, Gregory J. Edwards, Marc A.
Department(s):	Electrical and Computer Engineering
Research Mentor(s):	Edward Grant/Electrical and Computer Engineering
Title of Presentation:	Improving Evolvable Hardware Processes

In recent years there has been significant interest in the area of evolvable hardware, using evolutionary algorithms to generate circuits in Field Programmable Gate Arrays (FPGA's). Previous research by others showed that asynchronous, analog circuits took prohibitively long to evolve and were highly unstable under variances in ambient temperature. In order to produce robust circuits that will be consistent across FPGA families, methods must utilize the digital aspects of the device while preventing both asynchronous and potentially damaging configurations. A variety of evolutionary algorithms have been developed which are appropriate for evolvable hardware, including genetic algorithms, simulated annealing, and the HereBoy algorithm.  Some of the less popular algorithms have the potential to increase the speed of evolution in problems like evolvable hardware. The goals of this project are to develop highly  reproducible circuits and to decrease the necessary time to evolve reproducible circuits. Two primary strategies were utilized to increase the speed of evolution.  First, since instantiating the evolved circuit in the FPGA is the most time consuming step, several algorithms were compared for use in this project based on the average number of iterations necessary to solve a variety of problems.  The second method of increasing speed was to instantiate the entire genetic algorithm in an FPGA.  All previous research has reloaded the FPGA through an MCU or PC connection every generation; by instantiating the entire process in the device, the time to reload the device is significantly decreased.  The highly parallelized nature of the FPGA also increases the speed of other processes involved in evolution.  The project attempts to develop two applications, pattern recognition and a dynamic control system, implemented using evolvable hardware

Student Author(s):	Ballard, Tameshia S.
Department(s):	Biological and Agricultural Engineering
Research Mentor(s):	Kevin Keener/Food Science
Title of Presentation:	Variable Layer Heat Exchanger

The food industry is always seeking new and more efficient ways to process food products.  One area of particular interest is the way food is heated or cooled by means of heat exchangers.  The heat exchangers in current use do not allow for the efficient control of heat flux between the product and the heating or cooling medium.  The industry has difficulty changing the process conditions for existing heat exchangers, which leads to large amounts of product being lost as waste.  The use of a triple tube heat exchanger with a variable layer was investigated as a possible means of better controlling heat  transfer.  A triple tube heat exchanger was designed such that the product was in the inner most tube, the heating or cooling medium in the outer most tube and the middle tube was designated as the variable layer.  The variable layer was designed such that it was easy to evacuate or fill with any given medium.  The variable layers evaluated in this study were a vacuum, air, pressurized air, and water.  The cooling mediums evaluated in this study were chilled water and liquid nitrogen. The products evaluated were water and orange juice. In the designed experiment, inlet and exit temperatures and mass flow rates of the  product and cooling medium  were measured to determine the effect of the variable layer on the overall heat flux.

Student Author(s):	Bennett, Justin M.
Department(s):	Chemical Engineering
Research Mentor(s):	Gregory Parsons/Chemical Engineering Jason Kelly/Chemical Engineering
Title of Presentation:	Gate/Metal Oxide Interface Stability for Future Generation Transistor Devices

The 2000 International Technology Roadmap for Semiconductors indicates that replacing the silicon dioxide (SiO₂) gate insulator with a high dielectric constant (high-k) material is a critical challenge for future generation transistor structures. It is believed that leakage current due to quantum mechanical tunneling can be reduced with a high-k dielectric, but new issues arise, especially related to the reactivity of polysilicon and metal gates with advanced high-k materials. For this work, we have critically examined elemental metals on the periodic table, and using criterial of the Gibbs’ phase rule, thermodynamic stability at 1000^oC, electrical conductivity and band edge alignment, the list of potential future gate materials is narrowed and sorted for stability on six dielectric materials (SiO₂, HfO₂, ZrO₂, Al₂O₃, La2O₃, and Y₂O₃).  We find that Group III and Group IV metals are generally unstable with these six dielectrics, but several transition metals and metal oxides, including Re, Rh, and Ru are expected to be extremely stable, and will likely be more stable on Group III dielectrics than on Group IV dielectrics.  Some of these predictions have recently been tested, including Hf on HfO₂, and  RuO₂, on ZrO₂, and results will be discussed in relation to predictions.  In this way, we show that thermodyanamic analysis of material interface reactions is an important first step to determine the optimal gate materials for future generation transistor devices.

Student Author(s):	Blankenship, George A.
Department(s):	Chemical Engineering
Research Mentor(s):	Peter K. Kilpatrick/Chemical Engineering
Title of Presentation:	UV-vis Spectroscopic Determination of the Three-Dimensional Solubility Parameter for Polyaromatic Compounds

The ability to predict solute solubility in a binary solvent system has a wide range of practical applications (e.g., solvent selection for paints and coatings, refrigerants, polymers, etc.). In this work, the solubility behavior of 9-anthracenecarboxylic acid in various binary solvent mixtures was studied by UV-vis absorption spectroscopy.  The solute solubility limit was measured over the range of blend compositions from f₁= 0 to f₁=1.  For a particular blend composition, the molar absorptivity of the solute at l_max was calculated from a Beer-Lambert calibration curve of absorbance versus standard solution concentration.  The solubility results were fit to a thermodynamic model based on regular solution theory with corrections for excess entropy of mixing. The three-dimensional solubility parameter concept of Hansen was also incorporated to account for contributions from dispersion, polar, and hydrogen bond interactions.  The experimental binary solvent mixtures were carefully selected so that two of the Hansen parameters were equal (or nearly equal) and the remaining parameter varied over a wide range.  This allowed us to systematically probe the individual contributions of dispersion, polar, and hydrogen bond forces in a minimum number of experiments.  Once the individual solute solubility parameters are known, it is possible to predict solute solubility in several hundred commonly used binary solvents.

Student Author(s):	Bogle, Stephanie Taff, Brian Talib, Idris Cooper, Alisa
Department(s):	Materials Science & Engineering Multidisciplinary Studies  Mechanical & Aerospace Engineering  Electrical & Computer Engineering Chemical Engineering Textile Engineering, Chemistry, & Science
Research Mentor(s):	Dennis M. Maher/Materials Science & Engineering
Title of Presentation:	Electrical Properties of Metal-Oxide-Semiconductor Capacitors with Ultra-thin Gate Silicon Oxides: Part II Current-Voltage Characteristics and Analysis

The semiconductor industry strongly relies on its ability to continuously scale device size and gate-dielectric thickness to increase performance and reduce power consumption as well as cost. In research laboratories, the metal-oxide-semiconductor capacitor (MOSCAP) is used in the electrical characterization of advanced gate dielectrics. The present research addresses the issue of the current-voltage (I-V) characteristics of n+/p and n+/n MOSCAPs with scaled-oxide thickness values (i.e., 1.5 to 2.4 nm in steps of 0.3 nm). Model-based analyses of statistically sampled I-V data sets return the area dependence of the current at a given bias voltage and hence, from a linear least-squares analysis, the current density (J) for both the case of gate injection (i.e., n+/p MOSCAP) and substrate injection (i.e., n+/n MOSCAPs). The results of these analyses demonstrate that both gate and substrate injection generate nearly the same current density relative to flatband voltage (i.e., reciprocity holds for these devices). In addition, it is possible to assess the relative accuracy of the effective oxide thickness (tEOT) values that were extracted from the C-V characteristics (see Part I) from current density (J) and a theoretical rule-of-thumb that says a change in tEOT of 0.2 nm will result in a fractional change of one decade in J. The differences in oxide thickness for both the n+/p and n+/n wafers are both ~ 0.8 nm and the fractional change in current density is about four decades. Hence, it is concluded that the relative accuracy of the effective oxide thickness values is quite high. As stated in  Part I, the second phase of this research addresses the impact of a PMA on the no PMA current-voltage characteristics and extracted metrics for the same device wafers.

Student Author(s):	Carroll, Sarah E.
Department(s):	Computer Science
Research Mentor(s):	Robert D. Rodman/Computer Science
Title of Presentation:	Computer-Assisted Speaker Identification Based on Consonant Pronunciation

Speaker identification is the process of determining who spoke a recorded utterance through computer analysis of the recorded speech. Its most important applications are in computer security and forensics. For example, if the FBI has a recorded bomb threat, it is possible to compare it to the speech of a group of suspects. Or, a person’s speech can be used to augment secure identification for sensitive e-commerce transactions such as banking.
         This project focuses on individuating speakers based on their pronunciation of the English stop closure consonants /p/, /t/, /k/ and /ch/.  Most techniques for distinguishing speakers are based primarily on the pronunciation of vowels.  The goal of this research is to determine whether these consonants provide a new method for individuation. In particular, I am measuring the time elapsed between the release of the stop closure and the resumption of vocal cord vibration. The null hypothesis is that there will be no measurable variation among individual speakers of this parameter.
         Samples of speech collected from ten individuals will be digitally analyzed to determine whether their voiceless stop consonant voice onset times are sufficiently different to be used as criteria for distinguishing individual talkers.

Student Author(s):	Conlee, Christopher R.
Department(s):	Textile Chemistry
Research Mentor(s):	David Hinks/Textile Chemistry
Title of Presentation:	Azo Coupling in High-Pressure Carbon Dioxide

Azo dyes constitute a commercially important class of organic colorant.  However, the conventional synthesis of azo dyes requires high concentrations of mineral acid which leads to high electrolyte levels in effluent following neutralization. High electrolyte levels in effluent is a major environmental problem.
            This work  investigated two new synthetic strategies that utilized high-pressure carbon dioxide in the azo coupling reaction.  The first method followed the conventional diazotization from p-nitroaniline, followed by coupling with N,N-diethylaniline to produce the desired azo compound, but without addition of mineral acid. A mixture of water and carbon dioxide was used as the acid source for the reaction. Effects of temperature and pressure on reaction yields were examined.
            The second synthetic procedure coupled an aromatic nitroso compound with an aromatic amine to form the azo linkage.  Both reported routes provide an opportunity to reduce significantly the environmental impact associated with conventional aqueous azo coupling reactions.

Supercritical carbon dioxide (scCO2) provides an efficient and environmentally benign solvent for creating and modifying polymer blends in-situ. The resulting materials exhibit modified bulk and surface properties due primarily to plasticization effects, although specific chemical interactions may likewise be exploited for this purpose. While impregnation of one polymer with another in the presence of scCO2 has been previously demonstrated for a couple of examples in the literature, little effort has focused on the impregnation of one polymer into an existing polymer-inorganic composite as a means by which to modify the functionality of either the polymer matrix, the inorganic additive or the existing polymer-additive interface. In this work, a scCO2/poly(dimethyl siloxane) (PDMS) solution has been employed to swell thin films of poly(methyl methacrylate) (PMMA) containing glass beads measuring 40 nm in diameter and initially dispersed throughout the polymer matrix. The low-molecular-weight PDMS is dissolved in the scCO2 and then penetrates throughout the film as the film swells at a given temperature and pressure. The affinity of the PDMS toward the silica particles causes the PDMS to concentrate around the silica particles, thereby producing core-shell dispersions. This modified polymer matrix affects the fracture toughness of the material by creating a rubbery layer around the dispersed silica particles. Results from gravimetric, mechanical property and morphological  tests will be presented and discussed.

Student Author(s):	Drake, Amanda L. Luhr, Nicholas H.  Marley, Tommeka L.
Department(s):	Materials Science and Engineering
Research Mentor(s):	C. Maurice Balik / Materials Science and Engineering
Title of Presentation:	Organoclay Nanocomposites to Enhance Thermoplastic Gel High Temperature Properties

Thermoplastic elastomer gels based on styrene-ethylenebutylene-styrene (SEBS) copolymer (Kraton G1654) have been investigated as a less expensive replacement for silicone gels currently manufactured by Raychem Corporation and used as sealant materials. Thermoplastic gels are prepared by melt blending SEBS with hydrocarbon oils. Typical service environments that will be encountered require this sealing material to be capable of withstanding 90^oC with respect to elasticity, tensile properties and creep properties. Unmodified SEBS gels soften and become unusable well below this temperature. The addition of an organoclay is hypothesized to increase the thermal resistance of the gel material. It has been shown that greases which are stable at 90^oC can be formed by the dispersion of organically modified Bentonite clays in hydrocarbon oils. The addition of these same organoclays to the SEBS gels may enhance their high temperature properties, reducing creep and compression set. A matrix of samples has been prepared with various clay/polymer/oil compositions. Rheometry has been performed to obtain measurements of modulus and creep recovery. Hardness, stress relaxation, and thermal softening point data have also been obtained. Statistical process optimization software was used to find the best composition of the polymer, clay and oil that will produce the desired responses of low elastic modulus and good creep recovery, combined with the necessary high temperature resistance to flow.

Student Author(s):	Eni, Egbe U. Wilson, Bryan L.
Department(s):	Textile Engineering
Research Mentor(s):	Perry L. Grady/Textile Engineering Jeffrey Joines Textile Engineering
Title of Presentation:	Energy Analysis of the Yarn Production Process

Ring spinning, Rotor spinning, and Air Jet spinning are three primary methods of yarn production used in today's industries.  Each specific yarn production method encompasses certain machines and processes the fiber must be subjected to before yarn formation is complete, for example: opening, carding, drawing, lapping etc.  Each process employed consumes various amounts of energy depending on certain parameters such as; the amount of motor horsepower required, the weight and amount of product produced, the speed the product is fed into the machine, as well as the speed of the product exiting the machine etc.  Every additional process adds to the total amount of energy required to produce the final product.  By analyzing each process one can determine the optimal machine settings required to increase its efficiency.  Improving the efficiency of the individual procedures will inevitably decrease the total energy usage as well as the total production cost.  The College of Textiles at North Carolina State University has already created a visual C++ and Excel based model to analyze the energy consumption of individual processes.
            The task of the student was to first transfer and correct any and all codes in C++ to an Excel workbook.  The Excel workbook is capable of performing all previous required task formally executed by the C++ model in addition to maintaining a record of each yarn plant's energy consumption on a weekly basis.  The above student is now working towards creating an Energy web site for the American Textile Manufactures, which will be used to access an online version of the Energy models.  The online version will be based on Html and VB Script coding.  In the future the energy web site will be connected to a database containing the results and progress of participating manufactures which will be accessible by way of password.

Student Author(s):	Guichard, Alex R.
Department(s):	Materials Science and Engineering
Research Mentor(s):	Phillip Russell/Materials Science and Engineering
Title of Presentation:	Electron Beam-Induced Chemistry

Focused Ion Beam Milling (FIBM) is the standard for the micromachining of electronic devices; and recently, chemically-Enhanced FIBM (CE-FIBM) has been used to increase machining rate of many materials.  However, due to sample damage and ion implantation caused by the Ga+ ion beam, some fabrication processes require a micromachining technique that is less damaging to the sample.  Electron-beam induced chemistry is expected to cause less surface damage than FIBM and CE-FIBM, although machining times might be longer. E-beam induced chemistry is a process that involves either etching or deposition of a desired material via a gaseous precursor introduced to the sample surface, and a focused electron beam that provides the localized energy to make the chemical reaction proceed to completion.  One practical application of this technique is Extreme Ultraviolet (EUV) lithography mask repair.  EUV is a next-generation lithography that can eventually pattern lines as small as 13.4 nm.  In order to create an Electron Beam Milling System, a previously scrapped Scanning Electron Microscope (SEM) was refurbished and modified with a gas introduction system that consists of a gas reservoir and a hollow point needle aimed very close to the specimen surface.  We examined the beam-induced etching of three materials: developed photoresist, SiO2 film on a Si substrate, and Cr patterned EUV masks.  The gas used for the reaction was XeF2.  Initial results are encouraging, and well-controlled electron beam-induced etching has been exhibited on both photoresist and SiO2.  SEM and Atomic Force Microscopy (AFM) have been used to quantify the volume of removed material as a function of gas pressures and etch times.
 

Student Author(s):	Hampton, Keith R. Lowe, Billy J. Burwell, Kwanita T.
Department(s):	Materials Science and Engineering
Research Mentor(s):	Ray Benson/Materials Science and Engineering
Title of Presentation:	Determination of the Most Effective Cooling Sequence after Extrusion on the Microstructure and Machining Characteristics of Brass Key Blank Strip

The cooling sequence for brass key blank strip after extrusion has a large effect on the microstructure and machining properties of the material.  Ilco Unican uses two brass alloys: ILCO 353 and ILCO 072-1.  The ILCO 353 alloy is being produced with an elongated grain structure under the current manufacturing process.  This elongated grain structure results in poor machinability for the brass key stock.  The design team will attempt to determine the critical cooling temperatures, T1 (temperature of extrusion) and T2 (temperature of coil right before quench), which are anticipated to be the highest temperatures that will provide an equiaxed grain structure upon quenching.  After the critical temperatures are determined, the team will implement an efficient system for obtaining T1 and T2 and thereby producing a more desirable equiaxed grain structure.

Student Author(s):	Hill, Lentrell Schmid, Debra Yager, Jeremy
Department(s):	Material Science and Engineering Wood and Paper Science Mechanical and Aerospace Engineering
Research Mentor(s):	Jagannadham Kasichainula/Material Science and Engineering John S. Stewart/Wood and Paper Science
Title of Presentation:	Wear of Cemented Tungsten Carbide Tools During the Machining of Medium Density Fiberboard

Fiberboard is an essential building material for a wide variety of products.  It is a good means of recycling both sawdust from milling and scrap from previous operations.  Fiberboard is produced in several grades for a wide variety of purposes from furniture manufacture to building construction.
        Fiberboard inflicts severe wear on cutting tools. We have investigated wear of tungsten carbide cutting tools while machining fiberboard. These cutting tools are a composite of cemented tungsten carbide particles with varying percentages of cobalt binder.   In the Wood Machining and Tooling Research Laboratory, we conducted experimental machining  tests using these cutting tools to machine through the medium density fiberboard (MDF).  These machining test procedures were followed
by characterization of the wear damage imposed on the cutting tools. The study of the wear mechanisms associated with the machining process was carried out by examining the worn tools with an optical microscope and a scanning electron microscope (SEM) to measure the nose width of the worn tools and.  X-ray dispersive analysis and X-ray Mapping were used to determine the wear mechanisms of the cemented tungsten carbide tool.

Student Author(s):	Hunt, Marcus A.
Department(s):	Materials Science & Engineering
Research Mentor(s):	C. Maurice Balik/Materials Science & Engineering
Title of Presentation:	Diffusion and Solubility of Polymerizable Solvents in Ethylene Propylene Diene Monomer Rubber

The objective of this work is to obtain the diffusion coefficients of several polymerizable solvents in ethylene propylene diene monomer (EPDM) rubber over a temperature range of 10–70^oC.  The diffusion coefficient is obtained for the polymer/solvent systems of 5-methylene-2-norbornene (MNB), 5-ethylidene-2-norbornene (ENB), cis-cyclooctene (CCO), dicylopentadiene (DCPD), and methylene chloride (MC) in EPDM rubber using the soak-and-weigh technique.  A rubber sample is placed in each solvent using a water bath to maintain constant temperature.  The sample is removed from the solvent at various times during the test to measure the mass of the absorbed solvent.  Each sample is allowed to absorb solvent until the system reaches equilibrium.  A plot of mass absorbed vs. square root of time is used to calculate the diffusion coefficient of each polymer/solvent system.  Activation energies are obtained from Arrhenius plots of the diffusion coefficient against temperature. This data will be compared with activation energies and diffusion coefficients for the same solvents obtained in natural and nitrile rubbers.

Student Author(s):	Kondrad, Heidi Goodwin, Kim LaBorde, Leslie Smith, Ben
Department(s):	Materials Science & Engineering Chemical Engineering
Research Mentor(s):	Richard Spontak/Chemical Engineering/Materials Science & Engineering
Title of Presentation:	Nanocomposite-Reinforced Nylon 11 as an Improved Filler for NdFeB Magnets: A Multidisciplinary Design Approach

This multidisciplinary senior design project is sponsored by Magnequench Inc., a producer of neodymium iron boride (NdFeB) powders and permanent magnets. An important class of permanent magnets are those that are bonded within a polymer matrix. Bonded magnets are employed in a variety of applications, the most important of which is as small motors in automobiles. For example, windshield wiper motors and power accessory motors all incorporate permanent magnets. Automotive applications introduce the particular demand for materials to withstand high temperatures (up to ca. 180°C). Therefore, these magnets must retain not only their mechanical strength and integrity, but also their magnetic properties, at elevated temperatures. Our primary objective is to improve the temperature resistance of polymeric binders in bonded magnets, while retaining specific mechanical and magnetic properties at elevated temperatures. Furthermore, the design must be economically practical and compatible with existing manufacturing processes for bonded magnets. Our solution to this problem involves the creation of nanocomposites via the incorporation of various nanoscale clays into the polymeric binder of choice, Nylon 11. Literature reports support that nanocomposites often result in marked improvements in various polymer properties, such as heat deflection temperature and modulus. The clays we have chosen to use in this study are Bentone 38, Cloisite 93A and Cloisite 30B, which cover a broad range of chemical functionality. The composition of these clays is varied from 0 to 10 wt% in the polymeric binder. Thermal, mechanical and microscopical analysis will be used to determine the optimum composition of the selected clay for this application. The results obtained from these analyses will be reported, along with a modified product based on our findings.
 

Student Author(s):	Lewis, Amy N.
Department(s):	Biological and Agricultural Engineering
Research Mentor(s):	William F. (Bill) Hunt/Biological and Agricultural Engineering
Title of Presentation:	A Retrofit Stormwater Wetland

As pollution rates increase alternative water quality treatment practices must be implemented. One such treatment is a stormwater wetland. The following paper will discuss the functions and traditional views of wetlands. These functions include, but are not limited to, the nitrogen and phosphorus cycles, as well as biological productivity, water balance, erosion control, and suspended solids reduction. Stormwater wetland design is site specific. A wetland designed for one area may not work even a mile down the road. To size a stormwater wetland drainage area, soil type, soil use, and storage depth of the wetland must be considered. The following wetland will be placed at Smithfield Selma Senior High School (SSSHS) where it would receive runoff from the parking lots and buildings. The wetland will be 0.3 acres in size and use a weir as the outlet device. The size of the weir is a function of the retention time. The weir, which regulates water from large storms, will be thirteen feet wide and contain a two-inch orifice to regulate the retention time of the first one-inch flush.
            Once the wetland has been sized, a general map must be produced to determine the excavation plan. The excavation plan will include the amount of soil and existing vegetation that must be moved. Excavation is generally performed with heavy equipment such as bulldozers or backhoes. Plants must also be chosen for each of the four zones: deep pools and forebays (at least 2.5 feet deep), shallow water (0-1ft deep), shallow land (0-12 inches above normal pool), and non-floodable land (2-4 feet about normal pool). Normal pool is the depth of water after the wetland has drained to the chosen depth of the shallow water area. When choosing plants water depth, temperature variance,and shade/sunlight tolerance must be considered.

Student Author(s):	Martin, R. Keith
Department(s):	Biological and Agricultural Engineering
Research Mentor(s):	Peter L. Mente/Biological and Agricultural Engineering
Title of Presentation:	Lower Extremity Prosthetic with Sensory Feedback Control

Wearers of lower extremity prostheses often have difficulty maintaining balance, negotiating stairs, and detecting obstacles with their feet.  Current lower extremity prostheses offer no provisions to overcome these problems.  This often results in reduced mobility and coordination for the wearer.
            The objective of this project is to provide a solution to these problems through the use of feedback control.  Force sensors placed on the exterior of the prosthetic provide an electrical signal proportional to the amount of force experienced.  This signal actuates electro-mechanical devices built into the socket of the prosthetic and in contact with the skin of the residual limb, providing feedback to the wearer. The wearer can then adjust weight and position in response to the stimulation.  A control circuit monitors and adjusts baseline stimulation and protects against over-stimulation.
            The prosthetic improves proprioception by providing information to the wearer regarding the position of the leg and forces it encounters.

Student Author(s):	McCurry, Justin B.
Department(s):	Civil Engineering
Research Mentor(s):	John Stone/Civil Engineering
Title of Presentation:	Assessing Changes in Transit Customer Satisfaction due to the  Implementation of Real Time Information Technologies

The Winston Salem Transit Authority (WSTA) is a nationally recognized site for testing new bus transit technologies.  Among these technologies are Interactive Voice Response (IVR) voice messaging and menu system, Mobile Data Terminals (MDT), and Automatic Vehicle Location (AVL) devices. The goal of this research project is to determine what impact these new technologies have on the level of customer satisfaction.  This project will be carried out by conducting user surveys to establish a baseline "before" measurement of customer satisfaction, and then conducting follow up surveys to establish the after implementation level of customer satisfaction. A statistical comparison of the before baseline data with the after implementation data will then be conducted to determine to what degree the level of customer satisfaction has been impacted. This research will show how effective the technologies are in increasing levels of customer satisfaction.  In addition this research will demonstrate a practical assessment methodology.  WSTA and other transit agencies can apply the project results for evaluating future decisions to implement new technologies.
 

Student Author(s):	Reynolds, Kelli A.
Department(s):	Biological and Agricultural Engineering
Research Mentor(s):	Roger P. Rohrbach/Biological Engineering Charles J. Tucker/National Institute of Environmental Health Sciences Pierre R. Bushel/National Institute of Environmental Health Sciences Thomas L. Bennett/National Institute of Environmental Health Sciences Jonathan R. Miller/National Institute of Environmental Health Sciences
Title of Presentation:	Automation of Bioinformatics Processes for Microarray Data

The use of DNA microarray technology for scientific experiments has given rise to a massive influx of data to be stored and analyzed. The bioinformatics step of the experimentation process is a bottleneck due in part to the manual handling the data. This project incorporates methods of processing data in a more timely and efficient manner. The methods are three-fold, including a flagging script for biological classification, a clone order database, and a summary report for each experiment. First, a technique for identifying classes of genes has been created using a Perl script. The script compares and flags a verified list of genes with a specific biological classification, such as estrogen responsiveness, cell cycle, etc. against the genes used in the microarray experiments. With this, a scientist can search the database for all of the genes of a specific classification that he/she may be particularly interested in researching. Secondly, a method of ordering DNA clones to be included on future microarray chips must be incorporated that will ensure that duplicates are not printed.  An Access database is used to efficiently monitor the ordering of clones, as well as the requests that come in from scientists. The database distinguishes between clones already ordered and those not yet ordered, those that have proper identifiers and those that need to be researched further, and the species of clone requested. Lastly, at the end of experiment there will be a summary sheet that is accessible via the pre-existing database for each microarray slide used in the experiment. This sheet contains image and statistical information for each experiment.

Student Author(s):	Saretto, C.J.
Department(s):	Computer Science
Research Mentor(s):	R. Michael Young/Computer Science
Title of Presentation:	Mediation in Mimesis Liquid Narratives

Films and novels effectively convey intriguing stories, powerful emotions, and meaningful messages to their audiences. Telling interactive stories in a virtual environment seems a natural progression for narrative in light of modern technology. Unfortunately, current attempts often fail to engross the user by limiting her abilities to affect critical elements of the story. The Mimesis system attempts to tell liquid narratives that rewrite themselves in response touser activity. In order to maintain the integrity of the story being told, Mimesis employs a mediation system that arbitrates between the will of the user and the ends of the narrative. Speculative planning is utilized to determine actions that the user might perform that would interact harmfully with the current storyline.  The system also determines just how the storyline could be rewritten to accommodate the occurrence of those actions. If the storyline cannot be effectively rewritten or if intervention is preferred for a given action, the system is responsible for determining a realistic way of preventing the action. This is done by substituting one of the action's failure modes when the user attempts the action. An advanced controller determines how to respond to the user's actions. A precaching scheme inside the virtual world environment keeps track of the controller’s decisions and accommodates or modifies user actions in real-time.

Student Author(s):	Taff, Brian Bogle, Stephanie Sallach, Rory Cooper, Alisa
Department(s):	Mechanical & Aerospace Engineering Electrical & Computer Engineering  Materials Science & Engineering  Multidisciplinary Studies Chemical Engineering Textile Engineering, Chemistry, & Science
Research Mentor(s):	Dennis M. Maher/Materials Science & Engineering
Title of Presentation:	Electrical Properties of Metal-Oxide-Semiconductor Capacitors with Ultra-thin Gate Silicon Oxides: Part I Capacitor-Voltage Characteristics and Analysis

The semiconductor industry strongly relies on its ability to continuously scale device size and gate-dielectric thickness to increase performance and reduce power consumption as well as cost. In research laboratories, the metal-oxide-semiconductor capacitor (MOSCAP) is used in the electrical characterization of advanced gate dielectrics. The present research addresses the issue of the capacitance-voltage (C-V) characteristics of n+/p and n+/n MOSCAPs with scaled-oxide thickness values (i.e., 1.5 to 2.4 nm in steps of 0.3 nm). Model-based analyses of statistically sampled C-V data sets return the folowing metrics: effective oxide thickness; flatband voltage (or effective interface charge density); doping density of the polysilicon-gate electrode and the silicon substrate; metal-semiconductor work function difference; and oxide fixed charge. The results of these analyses demonstrate that the extracted values of the effective oxide thickness (tEOT) scale with expectation (i.e., delta tEOT is ~ 0.8 nm). Hence, it is concluded that the trends in the remaining electrical metrics are representative of MOSCAPs with near-atomically scaled ultra-thin gate silicon oxides. Since a so-called post-metallization-anneal (PMA) in forming gas at ~ 400 °C for 30 min was not included in the original process sequence, the second phase of this research addresses the impact of a PMA on the no PMA electrical metrics for the same device wafers. Traditionally, a PMA is included in the process sequence in order to neutralize interface charge.

Student Author(s):	van Maanen, Guillermo J.
Department(s):	Materials Science and Engineering
Research Mentor(s):	Richard J. Spontak/Materials Science and Engineering
Title of Presentation:	Reinforced Thermoplastic Elastomer Gels: Effect of Filler Type and Concentration

Polymer physical gels are ubiquitous in numerous technologies requiring "soft" responsive materials. Thermoplastic elastomer gels (TPEGs) consist of a microphase-separated multiblock copolymer in the presence of a low-volatility selective solvent. These inexpensive materials are currently used in a wide variety of commercial applications ranging from shock-absorbing media for fiber-optic bundles and pressure-sensitive adhesives to ergonomic devices employed in office and recreational activities. From a fundamental standpoint, they are interesting examples of stable nanostructured liquids that exhibit solid-like character due to the formation of a three-dimensional molecular network. One of the methods to modify the properties of TPEGs is through the addition of another (co)polymer or an inorganic filler. The focus of the present work is to discern the effects of filler content and type on the morphological and property characteristics of TPEGs. In this study, a model poly(styrene-b-isoprene-b-styrene) (SIS) triblock copolymer is impregnated with low concentrations of either nanoscale glass spheres, fumed silica or carbon black at different loading levels. The resultant nanocomposite is subsequently gelled through the addition of an aliphatic mineral oil, which serves as a selective solvent for the elastomeric isoprene network (stabilized through the physical formation of glassy styrenic crosslinks). The morphologies of these TPEGs are investigated by transmission electron microscopy, and their viscoelastic properties are measured by dynamic rheology. This study will be used to ascertain if  beneficial property modifications can be induced through the incorporation of nanoscale fillers in polymer physical gels.

Student Author(s):	Wallace, Charles S.
Department(s):	Wood & Paper Science
Research Mentor(s):	John A. Heitmann/Wood and Paper Science
Title of Presentation:	Pretreatment of Wood Pulp Fibers with Enzymes to Reduce Refining Energy Consumption

Mechanical disk refiners are used to develop strength properties of wood pulp fibers.  These refiners often consume 50% of all electrical power used in a typical paper mill.  Prior work has shown that treatments of certain enzymes can loosen the internal cohesion of the cell wall, but that enzymatic treatment alone is not sufficient to create the necessary loosening of the cell wall structure and the fiber swelling suitable for good fiber flexibility and high bonding strength.  This work investigates the use of enzymes as a pretreatment followed by mechanical refining at a reduced rate.
        The study used a bleached kraft softwood market pulp, a commercially available cellulase enzyme and a PFI mill, which is a laboratory refiner.  Enzyme dosage, pulp consistency and the amount of refining were the independent variables examined in the investigation.  Dependent variables monitored were freeness (drainage rate), opacity and strength properties of the paper.  Handsheets were made, and strength properties were tested at comparable freenesses.  Strength properties and energy use at comparable freeness were compared to control experiments without enzyme pretreatment.  The results showed that it would be possible to save 27% of the estimated ten billion killowatt hours of electricity used in refining of wood pulp annually.  This would save an estimated 132 million dollars a year in energy costs.

Student Author(s):	Welf, Erik S.
Department(s):	Wood and Paper Science
Research Mentor(s):	Richard A. Venditti/Wood and Paper Science
Title of Presentation:	The Effect of Heating and Drying on the Properties of Recycled Papermaking Fibers

Environmental and social pressures have accelerated the importance of paper recycling technology to the pulp and paper industry.  Only through innovations in the area of recycled paper processing and utilization can these demands be met.  Papermaking fibers, however, do not possess an infinite lifetime with respect to repeated recycles.  In fact, the quality of the paper products made with recycled fiber in general decreases each time that the fiber is recycled and reused.  This fact prevents the unlimited reuse of paper making fibers and limits the quality of products made with recycled fiber. If the mechanisms of fiber degradation during papermaking and recycling are better determined, then procedures may be outlined that will minimize this fiber degradation.  These improved recycling procedures will enable an increase in the amount of paper that may be recycled each year and allow the usage of recycled wood fibers in higher quality recycled products.  This research focuses on the degradation of cellulose polymers in papermaking fibers during papermaking and the effect that degradation has on fiber properties and product properties.  Because papermaking fibers are subjected to harsh heating conditions during the paper drying process, there is potential for thermal degradation of the cellulose, resulting in lowered fiber strength.  The effects of various heating and drying conditions on the cellulose chain length of the fibers were investigated.  Paper and fiber properties were related to the resulting cellulose chain length.  The experimentation utilized will provide insight into the nature of the causes of cellulose degradation and how paper production parameters affect fiber degradation.  This information should prove valuable in the present and future enhancement of recycled paper production

Student Author(s):	White, Brandon Bogle, Stephanie
Department(s):	Materials Science & Engineering
Research Mentor(s):	Rich Spontak/Materials Science & Engineering Hasaan Jameel/Pulp & Paper Science John Heitman/Pulp & Paper Science
Title of Presentation:	Energy Challenge 2001: Non-Woven Synthetic Composite Sail

The pulp and paper industry consumes nearly 2% of the raw energy produced annually in the U.S.  The U.S. Department of Energy (DOE) has established the Energy Challenge Research Competition to focus the talents of the nation's top universities on reducing the energy consumption associated with current paper production strategies.  This year's Energy Challenge targets the design of a non-woven wood-based composite material to be used in the construction of a sail. In addition to the development of the composite material, methods of energy conservation must be incorporated into the design process. The Kraft chemical pulping process has been used for this purpose, since it yields the strongest wood fiber obtainable after pulping. Anthraquinone is added to reduce the energy consumption associated with pulping.  Because paper is susceptible to water absorption and, thus, mechanical property degradation, our composite material also employs alkyl ketene dimer to reduce water penetration and kymene to increase the wet strength of the composite.  Polyester fiber has also been included in our paper production to increase tear strength. These additives have been systematically varied to identify the optimal combination for the present application. A computer simulation has been used to demonstrate the energy savings associated with the incorporation of Anthraquinone into a full-scale Kraft pulping mill.  Our final production method generates rolls of composite paper with a width of only 12". Upon necessary modification of a standard sail design, we have constructed a 76 sq. ft. sail prototype, which has been used to race against the other participating universities at the Georgia Institute of Technology as the conclusion of this year¹s Energy Challenge Competition.

• Biological Sciences abstracts
• Humanities and Social Sciences abstracts
• Physical and Mathematical Sciences abstracts