Physics and Engineering.

Chair: Robert S. Fritzius, Shade Tree Physics

Vice-chair: Alexander B. Yakovlev, University of Mississippi 

THURSDAY MORNING

Magnolia B

Antenna Research

8:00 ADAPTIVE NULLING IN A PHASED ARRAY ANTENNA WITH A GENETIC ALGORITHM

Andy Harrison, Radiance Technologies, Inc., Huntsville, AL 35805

The objective of this work was to show an approach to adaptive nulling in systems that incorporate a phased array antenna. This approach for adaptive nulling is based on the genetic algorithm. When operating a system in the presence of one or more strong interfering sources, sidelobe levels may not be low enough to ensure adequate reception of the desired signal. A method for overcoming this problem is adaptive nulling. In this method, the amplitude and phase coefficients of each element in the phased array are adjusted in such a fashion as to place a null in the antenna pattern in the direction of the interfering sources. Since the genetic algorithm is a global search method, a solution is found that places very deep nulls in the desired directions, while maintaining the characteristics of the antenna main beam. The sidelobe structure of the antenna pattern is distorted and other optimization parameters may be added to the genetic algorithm as needed.

8:15 PROBLEMS IN DESIGNING BEAM FORMING NETWORKS FOR HIGH FREQUENCY APPLICATIONS

Manish Hiranandani* and Ahmed A. Kishk, University of Mississippi, University, MS 38677

Recently, Beam Forming Networks (BFN) have been attracting more attention as their applications evolve from military to commercial purposes. Today, the most common beam forming networks are the Butler Matrices and Rotman lens. In this project, we analyze the performance of these BFNs operating in the X-band (8-12 GHz). The Butler Matrix and Rotman lens are N X N passive microwave networks, consisting of 'N' input and 'N' output ports, used to feed an array of 'N' antennas. The above N X N networks generates a set of 'N' orthogonal beams, which could scan the whole visible region. Broad banding techniques are introduced in the conventional circuits to obtain appreciable wide band performance. Problems encountered in designing such networks at high frequencies are addressed. The designs are simulated using a full wave commercial code based on the method of moments. The networks are then fabricated and tested using a Hewlett-Packard 10C Network Analyzer to obtain the S-parameters. Good agreement between the computed and measured results is obtained.

8:30 WIDEBAND TWO-DIMENSIONAL ARRAYS OF MICROSTRIP-PROXIMITY FED RECTANGULAR SLOT ANTENNAS WITH LOW SIDELOBE LEVEL

Abdelnasser Eldek*, Atef Z. Elsherbeni, and Charles E. Smith, University of Mississippi, University, MS 38677

In applications where loss, dispersion, size, weight, cost, performance, ease of installation, and aerodynamic profile are constraints, printed slot antennas offer viable alternatives to other types of traditionally used antennas. Moreover, in some applications like wireless communications and radar systems, very wide bandwidth and high gain are among the main design objectives. This paper presents the design of two-dimensional arrays based on microstrip-proximity fed rectangular slot antennas for marine RADAR applications. Two designs are conducted, one consists of 16X4-element array, and the other consists of a 32X4-element array. A wideband corporate (parallel) feed network is designed to feed the proposed arrays, rather than the narrowband feed network used with the traditional patch array for this type of application. The designed feed network for the 16X4-element array provides uniform power distribution, while the feed network of the 32X4 array uses Dolph-Tschebyscheff coefficients to reduce the levels of the sidelobes. The measured return loss and radiation patterns are presented for the proposed array designs.

8:45 ANALYSIS OF RADIATION APERTURE ANTENNA USING THE REGION-BY-REGION FDTD METHOD

Yizhe Zhang*, Ahmed A. Kishk, Alexander B. Yakovlev, and Allen W. Glisson, University of Mississippi, University, MS 38677

In the finite difference time domain method (FDTD) method a non-rectangular computational domain is usually approximated by a rectangular domain to terminate the region by an absorbing boundary. Internal problems can be modeled efficiently using the region-by-region approach in order to discretize only the non-conducting parts of the problem in order to reduce memory requirements and CPU time. Here, the same technique is used to analyze open structure such as aperture and horn antennas. In this work, the region-by-region approach is applied for modeling rectangular horn antennas. The far field radiation patterns are computed based on the Fourier transformed aperture distribution.

9:00 ANALYSIS AND DESIGN OF PRINTED VIVALDI ANTENNA FOR WIDEBAND APPLICATIONS

Walker Hunsicker*, Atef Z. Elsherbeni, and Charles E. Smith, University of Mississippi, University, MS 38677

The dual use of printed circuit boards in communications devices for circuit mounts and antenna substrates has generated a whole new category of antenna. Printed and slot antennas take advantage of the highly refined and inexpensive dielectric boards by conforming metallization patterns to produce desired radiation patterns and gains. One such design which exhibits wideband characteristics is the Vivaldi antenna. The exponential taper of this class of antenna is the dominant feature of the geometry. The design of the Vivaldi antenna is documented but with few details on how to mathematically construct it. Most of the available designs exhibit very complicated feeding structures to properly excite the antenna. This paper investigates simple feeding configurations for the printed Vivaldi antenna with an emphasis on wideband performance. The structure proposed is simulated with a finite difference time domain code to extract the input parameters and far field patterns. A prototype of a successful design is built and a comparison between measured and simulated results is performed to confirm the validity of the design.

9:15 Break

Transmission Lines/Waveguides

9:30 ANALYSIS OF PRINTED TRANSMISSION LINES

Asem Al-Zoubi* and Ahmed A. Kishk, University of Mississippi, University, MS 38677

Coplanar waveguides (CPWs) are finding extensive applications in microwave integrated circuits. Many structures have been analyzed using the conformal mapping method, the CPW with lower ground planes, asymmetric CPW (ACPW) with conductor backing, and conductor-backing ACPW with one lateral ground plane. Here, closed form expressions for the effective relative permittivity and the characteristic impedance of the asymmetric CPW with finite-extent ground and finite ground-backing are studied using the conformal mapping method (CMM). Closed form expressions for the effective relative permittivity and the characteristic impedance are derived. Various structures can be analyzed from the original structure. Also, the asymmetric CPW line with finite-extent ground and finite ground-backing is analyzed using the finite difference method (FDM). The formulation of the problem is based on the solution of Laplace's equation subject to appropriate boundary conditions, and the use of Taylor's series expansion to approximate the first and second order derivatives in Laplace's equation. Numerical results from the two methods are compared. The effect of various parameters such as: slots asymmetry, finite ground, substrate dielectric material, strip width, slots width, and substrate thickness of asymmetric CPW with finite-extent ground and finite ground-backing and microstrip line with finite ground planes are studied using the expressions obtained.

9:45 DIPOLE AND SLOT PERIODIC STRUCTURES FOR THE REALIZATION OF HARD AND SOFT SURFACES

Manish Hiranandani*, Alexander B. Yakovlev, and Ahmed A. Kishk, University of Mississippi, University, MS 38677

Hard and soft surfaces have been recently introduced as GO and STOP surfaces, respectively, with respect to the direction of propagation along the surface. Ideally, they could be viewed as periodic alternating perfect electric conductor (PEC) and perfect magnetic conductor (PMC) strips with period length approaching zero. In recent years, different methods of realizing artificial hard and soft surfaces have been developed. One of them is based on the use of dielectric-filled corrugated surfaces. Another common method is to introduce PEC strips on a dielectric slab of with the proper thickness, which is close to a quarter wavelength. A new method has been developed to create artificial hard and soft surfaces by mounting dipoles and slots in a periodic form on a conductor-backed dielectric slab [Maci S. and Kildal, P. S., Int. Symp. Electromagnetic Theory, Pisa, Italy, May 2004]. The new structure removed the restrictions on the dielectric slab thickness. Therefore, soft and hard surfaces can now be realized using thin substrates. These structures act as hard and soft surfaces for TE and TM polarizations, depending on the direction of the incident plane wave. In the present paper, the results obtained in the above paper are verified and the analysis of these structures is further extended in order to create new designs of multiband and polarization independent hard and soft surfaces. A full wave commercial software based on the method of moments is used for the analysis of the periodic structures. Such structures are technologically simple and can be implemented using etching techniques.

10:00 DYADIC GREEN'S FUNCTION ANALYSIS OF AN IDEAL HARD SURFACE RECTANGULAR WAVEGUIDE

Wei Huang*, Alexander B. Yakovlev, Ahmed A. Kishk, Allen W. Glisson, and Islam A. Eshrah, University of Mississippi, University, MS 38677

A rectangular waveguide with an ideal hard surface is studied, which is modeled by alternating the longitudinal perfect electric conductor (PEC) and perfect magnetic conductor (PMC) strips with vanishing widths. Compared to PEC and PMC rectangular waveguides, which can support only TM and TE modes, the most important feature of this "ideal hard surface rectangular waveguide" is that it allows a propagation of the TEM mode with a zero-cutoff frequency, which can provide new applications for this type of guided-wave structure. The purpose of the this paper is to develop an electric dyadic Green's function for the modal analysis of an ideal hard surface rectangular waveguide excited by an arbitrarily-oriented electric current source. A procedure for deriving the Green's function in terms of solenoidal and irrotational parts is presented, wherein the solenoidal part of the Green's function is obtained in the eigenmode expansion form as a superposition of three terms associated with TM, TE, and TEM modes of the ideal hard surface waveguide. A term corresponding to the TEM mode is obtained analytically as the solution of a vector Helmholtz equation in the zero-cutoff limit subject to the boundary conditions of electric field on the ideal hard surface. Numerical results of the field distribution are demonstrated for the TEM mode and a few representative TM and TE modes propagating in a rectangular waveguide with ideal hard surface boundary conditions due to an arbitrarily-oriented electric dipole source.

10:15 PARAMETRIC STUDY OF THE DRA-WAVEGUIDE-BASED SPATIAL POWER COMBINING SYSTEM USING FDTD

Yizhe Zhang*, Ahmed A. Kishk, Alexander B. Yakovlev, and Allen W. Glisson, University of Mississippi, University, MS 38677

A generalized scattering matrix (GSM) approach that utilizes the finite-difference time-domain (FDTD) method is used to analyze waveguide-based coaxial probe-fed dielectric resonator antenna (DRA) arrays for use in spatial power combining system. First, a rectangular waveguide with hard walls (dielectric loading along narrow sides of the waveguide) is analyzed to achieve a uniform field distribution in the waveguide cross-section. This is crucial for the waveguide-based DRA array in order to provide a uniform (with respect to magnitude and phase) excitation of antenna elements. Next, a single DRA is studied for operation in the waveguide environment. This includes a parametric analysis of DRA geometrical and material parameters and coaxial probe feed position and dimensions. The analysis is further extended to the case of the DRA array in order to minimize mutual coupling between antenna elements and provide a uniform coupling of the power to individual DRAs. The numerical results obtained using our FDTD code that is based on the region-by-region approach are compared with those obtained using commercial three dimensional (3-D) software and exhibit very good agreement.

10:30 MODELING PERIODIC STRUCTURES USING FINITE-DIFFERENCE TIME-DOMAIN METHOD 

Guiping Zheng*, Ahmed A. Kishk, Allen W. Glisson, and Alexander B. Yakovlev, University of Mississippi, University, MS 38677

In this work, numerical analysis of periodic structures based on the finite difference time domain (FDTD) method is presented. In many electromagnetic applications, structures have properties of periodicity in one or two dimensions, such as for frequency selective surfaces (FSS), photonic bandgap (PBG) structures, and infinite antenna arrays. The FSS structures have been widely used to construct a high impedance ground plane. By using a high impedance ground plane, the performance of antennas can be greatly improved. Low profile, low sidelobe levels, and high efficiency for antennas can be achieved in this way. The FDTD technique has been successfully used to obtain the numerical solution without resorting to the complex frequency domain analysis required in the Method of Moments (MoM). To implement the FDTD method in these periodic structures, Maxwell's equations are transformed and modified so that the complex periodic structures can be simply modeled by one unit cell with proper periodic boundary conditions. Several validation cases including FSS and antenna arrays are presented.

THURSDAY AFTERNOON

Magnolia B

Graphical Processing Simulations and Analyses

1:30 A GRAPHICAL USER INTERFACE (GUI) FOR PLANE WAVE SCATTERING FROM CHIRAL SPHERES

Veysel Demir (1*), Atef Z. Elsherbeni (1), Denchai Worasawate (2), and Ercument Arvas (3), (1) University of Mississippi, University, MS 38677; (2) Kasetsart University, Bangkok, Thailand; and (3) Syracuse University, Syracuse, NY 13244

Various numerical techniques have been developed which model the electromagnetic field propagation in various novel complex media. The validity of these techniques is usually verified by comparison to the exact solutions of canonical problems. Recently, the chiral medium has gained focus in research, and electromagnetic wave propagation in chiral media has been modeled by various numerical techniques in various studies. In most of these studies, the validity of the developed techniques is verified by comparing the numerical results to the results of one-dimensional and two-dimensional problems that have known exact solutions. For the techniques solving three-dimensional problems, plane wave scattering from a chiral sphere is the benchmark. In this contribution, a software package is developed and presented to calculate plane wave scattering from a chiral sphere. The package involves a user-friendly GUI, which enables the user to enter the scattering parameters and observe the results, in near real time, and save the calculated data and displayed figures. Due to the nature of the chiral constitutive relations, the developed program can be used to calculate scattering from a dielectric or a perfectly conducting sphere as well.

1:45 USE OF GRAPHIC PROCESSING UNITS FOR GENERAL SCIENTIFIC COMPUTATIONS

Matthew Inman*, Atef Z. Elsherbeni, and Charles E. Smith, University of Mississippi, University, MS 38677

Over the past few years, developments in the field of graphic processing units (GPU's) for video cards have vastly outpaced their general central processing unit (CPU) counterparts. As specifically applied to vector mathematic operations, the newest generation GPUs can generally outperform current CPU architecture by a wide margin. With the addition of large onboard memory units with significantly higher memory bandwidth than found in the main system, graphic cards can be utilized as a highly efficient vector mathematic co-processor. Implementing functions in high-level languages that utilize the vector processing power of the video cards, an appreciable increase in the effective speed for vector and matrix computations that are widespread in scientific computing currently can be achieved. By formulating proper procedures to realize general vector computations on GPUs it will be possible to maximize the processing power available to an extent greater than possible without the addition of the video card. This talk will review the state of the art in GPU for scientific programming with emphasis on numerical techniques for electromagnetics and antenna applications.

2:00 MUR'S ABSORBING BOUNDARY CONDITION FOR THE SPLIT-FIELD METHOD

Guiping Zheng*, Ahmed A. Kishk, Allen W. Glisson, and Alexander B. Yakovlev, University of Mississippi, University, MS 38677

In this work, a second order Mur's absorbing boundary condition for the split-field method in the finite-difference time-domain (FDTD) method is presented. In many electromagnetic applications, one needs to model an "open" region or a structure that is situated in free space, such as a scattering or a radiation system that is radiating into unbounded space. Unfortunately, due to the limitations of the computational resources, the field computational domain in FDTD must be limited in size and suitable absorbing boundary conditions (ABC) on the outer perimeter of the domain must be applied to simulate its extension to an infinite domain. Before the Perfectly Matched Layer (PML) ABC was introduced, Mur's second order ABC and its extension played an important role in the FDTD technique. Compared to the PML ABC, Mur's second order ABC is much simpler and more convenient. In many applications, when the PML ABC is not easy and convenient to construct, Mur's second order ABC is a good alternative approach. When modeling periodic structures in the FDTD method, Maxwell's equations are transformed, modified, and discretized by using split fields. In order to properly absorb the outgoing waves, the Mur's procedure used to truncate the FDTD computational domain must also be transformed and modified. In this work, several validation cases including frequency selective surface (FSS) are presented.

2:15 FINITE-DIFFERENCE TIME-DOMAIN SIMULATION OF THE HUMAN HEAD

Veysel Demir* and Atef Z. Elsherbeni, University of Mississippi, University, MS 38677

Development of tools for numerical analysis of electromagnetic scattering and absorption in the human body has been the focus of many researches because these tools can be used for various purposes such as the evaluation of the specific absorption rates (SAR) for determination of the safe use of electromagnetic devices, diagnosis of tumors, etc. In this contribution, a finite-difference frequency-domain (FDFD) procedure is developed in order to simulate electromagnetic wave and human head interactions. The advantages of the FDFD method is its accuracy and stability. In order to integrate the human head model into the simulations, the head has to be discretized to fit in the FDFD computational domain. A human head model is obtained from magnetic resonance imaging (MRI) data. Some of the artifacts from these data are cleaned and then an anatomic head model, which includes the three dimensional distribution of various tissues, is constructed. Every biological tissue is assigned specific permittivity, permeability and conductivity values. The electromagnetic simulation is performed using the head model data and these electrical parameters. After the simulation, electromagnetic field distributions in the head can be viewed in any cross-sectional plane cut for medical diagnosis.

2:30 ITERATIVE MULTI-REGION TECHNIQUE FOR LARGE ELECTROMAGNETIC SCATTERING PROBLEMS

Mohamed Al Sharkawy*, Veysel Demir, and Atef Z. Elsherbeni, University of Mississippi, University, MS 38677

One of the techniques for solving large electromagnetic problems is to divide the computational domain into smaller sub-domains and solve each sub-domain separately. Then the sub-domain solutions are to be combined following a procedure that incorporates the interactions between these subdomains to obtain a solution for the complete domain. In this work, we present a new technique based on the finite difference frequency domain (FDFD) method and an iterative procedure between the sub-domains to calculate the scattering from multiple two dimensional objects. The problem is decomposed into separated sub-domains, in each sub-domain, the scattered electromagnetic near fields are calculated due to the incidence of a time-harmonic wave, using the FDFD method. Then fictitious electric and magnetic currents on imaginary surfaces surrounding the objects in these sub-domains are calculated, using the equivalence principle. Radiated fields by these currents are then considered as incident fields on the opposing sub-domains. The same procedure of calculating the subdomain field components, the fictitious currents and the radiated fields on the opposing domains is repeated iteratively until a convergence criterion is achieved. This technique effectively reduces the size of the required memory, especially for practical and three-dimensional problems. Furthermore, the central processor unit time reduction can be achieved if the separation between the sub-domains is large and/or coarser grids are used in some of the sub-domains.

2:45 NEURAL NETWORKS APPROACH FOR SOLVING INVERSE ELECTROMAGNETIC SCATTERING BY A PARTIALLY BURIED DIELECTRIC SPHERE IN A GROUND PLANE

Chye Hwa Loo* and Michael M. Hamid, University of Mississippi, University, MS 38677 and University of South Alabama, Mobile, AL 36688

An analytic solution of the problem of electromagnetic scattering by a dielectric spherical scatterer resting on, or partially buried in, an infinite perfectly conducting ground plane is formulated using the method of images. The scattered field coefficients are solved exactly so that the scattered field can be calculated as a function of the sphere radius and permittivity as well as the burial distance for any specified angle of incidence. The solution of this problem is relevant to analyze the scattering by complex three-dimensional bodies, plastic mines, icebergs, rough surfaces, etc., in which the flat background can be modeled by the ground plane while the complex body can be simulated by a sphere or a system of spheres partially truncated and resting on the ground plane. In order to solve the inverse scattering problem, we employ a radial basis function network to take the scattered field complex coefficients for the TE and TM polarization case as the network inputs to predict the three outputs of the electrical radius, burial distance, and relative permittivity of the sphere. The trained network is able to retrieve the three aforementioned parameters from new data that is different from the learning data. The neural network approach is simple, straightforward and time saving, compared with analytical and numerical techniques.

4:00 Divisional Poster Session

INVESTIGATION OF THE THRU-REFLECT-LINE CALIBRATION TECHNIQUE FOR PACKAGED TRANSISTORS AND INTEGRATED CIRCUITS

Dalia Elsherbeni*, W. Elliott Hutchcraft, Charles E. Smith, and Darko Kajfez, University of Mississippi, University, MS 38677

The widespread use of vector network analyzers (VNAs) to characterize high frequency networks over wide frequency requires very precise calibration. Traditional, full, two-port calibration methods typically use three impedance standards and one transmission standard to calibrate VNAs. The standards normally used in this method are a short, open, load, and thru (the SOLT calibration). SOLT calibration standards can be difficult--if not impossible--to build in many non-coaxial measurement applications (such as in-fixture, wafer and waveguide measurements) which limits application for some frequency ranges and/or geometrical line configurations. In this paper, a newer calibration technique, called the thru-reflect-line (TRL) for two-port calibration that uses at least three standards to define the calibrated reference plane, is studied for use in testing packaged transistors and ICs having microstrip interface connections. Microstrip TRL standards for the thru, reflection, and line are designed, fabricated and tested primarily for X-band applications (8.2-12.4 GHz). TRL is the most accurate and widely used calibration method, but the frequency range coverage is limited to approximately an 8:1 ratio. The measured parameters of the thru, reflect, and line standards in a TRL calibration kit perform the same function as a SOLT calibration. The Reflect standard can either be an Open or a Short, and the accuracy requirements for the Reflect standards is lower than SOLT standards. The line section of a TRL calibration should be about 1/4 wavelength, or 90[degrees] at the center of the frequency range of interest (the exact length is not critical as long as it is known). Design and fabrication techniques for the TRL standards are presented along with accuracy verification tests of the standards and measurements for packaged microwave networks in the frequency ranges of interest.

DEVELOPMENT OF A WINDOWS BASED AUTOMATED SYSTEM FOR ANTENNA CHARACTERIZATION

Lisa Jordan*, Aik Min Choong, W. Elliott Hutchcraft, and Charles E. Smith, University of Mississippi, University, MS 38677

Because of their widespread use, vector network analyzers (VNA) have become the backbone of microwave and RF measurements. They can be used for the determination of an antenna pattern by rotating a test antenna and looking at the transmission from the test antenna to a known receiving antenna. However, the calibration of the network analyzer in addition to controlling the rotation of the antenna under test can be difficult and quite time consuming. To assist the user in the process, and allow the rapid characterization of an antenna, a Windows-based program with a graphical user interface was developed. This program uses a computer to control both the stepper motor which rotates the antenna and a network analyzer that makes the measurements. With computer control, the setup and calibration of the system can be performed automatically without any user intervention. In addition, the pattern data can be transferred from the VNA to the computer. This will allow the easy manipulation and visualization of the data in MATLAB. Details of the interfacing between the computer, the stepper motor, and network analyzer will be shown and the development of the GUI will be presented.

FRIDAY MORNING

Magnolia B

Physical Measurements and Analyses

8:00 ULTRASONIC DISTANCE MEASURING DEVICE

Kori McDaniel (1*), Henk De Weerdt (2), and James Conrawd (1), (1) University of North Carolina, Charlotte, NC 28223 and (2) University of Mississippi Medical Center, Jackson, MS 39216

A distance measuring device was designed using the AVR Butterfly Atmega169V and the SRF08 Ultra sonic range finder sonar. Measuring devices have been mainly developed using laser detection. Therefore, we have developed a rather simple ultrasonic device, which has broader application. This device was set up to measure distances between zero cm to 11 meters. The device was programmed using C code to administrate the functions to a micro controller. The process consisted of establishing a TWI line bus (12C bus). The TWI line consists of the SDA and SCL, located in USI port E, pin 4 and pin 5 respectively. These two pins were connected to 1.8k resistors, tied to the +5 VDC source and the SRF08. Once the communications between the two devices were established, the device was tested for measurements. During implementation the LCD displayed a count of 5 seconds delay to allow the Atmega169V to reset the 12C bus. During the delay the SRF08 was triggered by the toggle switch to deliver the +5VDC source to the power connection. The program was set to transmit the returning data from the SRF08's echo to the 12C bus for display on the LCD. Out of 50 measurements taken, only 2% error was observed. After debugging of the C code, subsequent readings presented satisfactory accuracy. With some modification this program will have the capability of automation and multiple measurements from different objects. The main program and all of its source files are available for modification.

8:15 TRANSIENT HEAT-TRANSFER ANALYSIS OF SSME TEST-STAND DIFFUSER 

Sam M. Aceil, Alcorn State University, Alcorn State, MS 39096

The high altitude simulation test diffuser is a large cylindrical structure where combusted rocket fuel passes through with supersonic speed up to 4 to 5 Mach number. The temperature at the wall of the diffuser could be as high as 1800 R. These conditions require special design for the wall of the diffuser. Analysis indicates that the dominant modes of heat transfer are conduction and convection. Considering the highly transient nature of test conditions, the lumped capacitance method was utilized to obtain reasonable first approximations for diffuser temperature variations. The result is a first order non-homogeneous differential equation with the following solution T(t) = b/a + ([T.sub.0] - b/a) [e.sup.-at] where "T" and "t" represent temperature and time, and a, b and [T.sub.0] are constants involving physical properties and initial conditions. The result of this investigation indicates that the average wall temperature varies from the nozzle entrance to the diffuser exit. The maximum temperature rise occurs between 30-50 seconds after the start of the test and it changes from 710 to 940 R. This investigation was part of a summer assignment at NASA Stennis Space Center.

8:30 MECHANICAL TESTING AND STATISTICAL CHARACTERIZATION OF HAND LAY-UP GLASS REINFORCED POLYMERIC (GRP) COMPOSITES

Reid J. Averill, University of Mississippi, University, MS 38677

The objective of this study is to determine mechanical properties of hand lay-up glass reinforced polymeric composites as specified in relevant standards. These mechanical properties will then be statistically characterized. The mechanical properties that are desired include ultimate tensile strength and tensile modulus, ultimate flexural strength and flexural modulus, ultimate compressive strength and compressive modulus, and shear strength. Specimen samples shall be fabricated according to all relevant standards. The ultimate tensile strength and tensile modulus shall be determined using a MTS servo-hydraulic test machine following ASTM D 3039 standard. The ultimate flexural strength and flexural modulus shall be determined using a 3-point flexural test according to the ASTM D790 standard. The ultimate compressive strength and compressive modulus shall be determined using the ASTM D 695 fixture according to the SRM 1R-94 (modified ASTM D 695 standard). The shear strength shall be determined using the losipescu shear test according to the ASTM D 732 standard. Test results are being used for the design of a new application for polymer composites.

8:45 Break

Special Invited Speaker

9:00 RAINWATER ASTRONOMICAL OBSERVATORY SYMPOSIUM

James G. Hill, Director, Rainwater Observartory and Planetarium, NASA/JPL Solar System Ambassador, French Camp, MS 39745

"Visit to a Ringed Plant: Science Results from the Cassini Mission to Saturn"

The program will begin with an introduction to the new research telescope facility to be housed at Rainwater Observatory. A major part of the session will be devoted to discoveries of the Cassini spacecraft and Huygens probe. Time permitting, highlights will be provided regarding "NASA's Origins Program: Searching for Plants and Other Stars."

FRIDAY AFTERNOON

Magnolia B

Physics of Things that Can Hurt You

1:00 USE OF EMULSIONS FOR CAPTURING AIR-BORNE PARTICULATE RADIOACTIVE CONTAMINANT FROM A DIRTY BOMB

Kaumudi Atapattu (1*), Garey Fox (1), and Victor Medina (2), (1) University of Mississippi, University, MS 38677 and (2) Army Corps of Engineers, Vicksburg, MS 39180

A dirty bomb or a Radioactive Dispersal Device (RDD) combines conventional explosives such as dynamite with radioactive material, which when detonated spreads fine particulates of radioactive material. The main objective of a dirty bomb is to throw an area into a state of fear and panic, disrupting the daily life of millions of people. The fine particulate matter eventually settles on surfaces. The current method of dealing with a dirty bomb requires the removal of all contaminated surfaces: a very costly endeavor. An alternate strategy that has been proposed is the use of emulsions for stabilizing the released contaminant. The emulsions would be sprayed as a thin film on the surfaces in the contaminated area. Previous research has evaluated emulsions for capturing and immobilizing radioactive contamination already deposited on surfaces. A research question not yet investigated is whether such emulsions can immobilize radioactive particulate matter being deposited from the atmosphere. The goal of this experiment was to test the use of the emulsions in capturing and retaining particulate matter deposited from the atmosphere. Emulsions were applied to triplicate samples of wood and concrete. Talcum powder was used to simulate the particulate nature of common radioactive materials of high probability in dirty bombs such as cesium chloride. The samples were allowed to dry and were then shaken. The mass of powder lost and the cumulative mass of the samples were recorded to document capture efficiency. Particulate capture densities (20.4 to 39.2 g [cm.sup.-3]) exceeded those expected after denotation of a RDD.

1:15 CHARACTERIZATION OF MISSISSIPPI LAND-FILL LEACHATES

Pao-Chiang Yuan* and Ana L. Balarezo, Jackson State University, Jackson, MS 39217

In the past three decades, electronic waste has been the fastest growing solid waste problem in the U.S.A. According to a recent Environmental Protection Agency (EPA) study, electronic waste is only 10% of the municipal solid waste (MSW), 50% of American homes have one computer, and there will be approximately 130 million cellular phone subscribers at the end of year 2005. This increasing usage of electronic materials may cause a problem in MSW landfill residues, especially in the leaching process. This paper focuses on possible contaminant changes in recent years, and health risks due to massive electronic waste being dumped into MSW landfills. In early 2004, a survey letter was mailed to the 14 existing MSW landfills in Mississippi to ask them to participate in this study. Six landfills were willing to participate. We sent out two sampling teams to the sites to collect samples, one in the spring and another in early summer. Physical and chemical properties of Resource Conservation and Recovery Act (RCRA) regulated metals were analyzed in the lab; including pH value, COD, BOD, Total Solids, Dissolved Solids, Suspended Solids, Conductivity and Metals. The results are compared with the National Pollutant Discharge Elimination System (NPDES) and RCRA hazardous waste standards. The Atomic Absorption (AA) technique is used to analyze the metals. We found some sites with high levels of suspended solids but most metal concentration did not exceed the RCRA regulated level.

1:30 EARTHQUAKE DAMAGE ASSESSMENT AND LIQUEFACTION POTENTIAL IN NORTH MISSISSIPPI

Chris Mullen*, Khalid-Ahmed B Desai, K.P. George, and Charles Swann, University of Mississippi, University, MS 38677

The State of Mississippi has potential to experience ground shaking from an earthquake within the New Madrid Seismic Zone in Arkansas, and other locations within Mississippi. In this study, earthquake damage assessment for the State of Mississippi is carried out for four different earthquake scenarios using HAZUS-MH software. Peak ground acceleration (PGA) for different counties is computed and damage to building stock, essential facilities, transportation systems and utility systems are compared for different scenarios. Economic impacts of all four scenarios are calculated to rank counties based on dollar loss. Analysis reveals that the New Madrid earthquake scenario causes the maximum damage as compared to other scenarios, and DeSoto County ranks highest in economic loss as compared to all other counties. The damage and economic losses reported in this study do not include liquefaction induced damage. Therefore, liquefaction analysis is performed using a stress-based approach and PGA computed from HAZUS. Standard penetration test (SPT) data on bridges and casino location is obtained from MDOT and other private agency to perform the stress-based approach. The result indicates significant liquefaction potential in DeSoto and Tunica counties. Based on the available SPT data and calculations, it reveals that all casinos in Tunica may experience liquefaction if an earthquake of magnitude 8 occurs in New Madrid seismic zone. Employing standard penetration test data, lateral spread analysis is performed for casino locations indicating lateral spread of the order of 10 to 20 centimeters.

1:45 THE FIRES OF CANNETO DI CARONIA: VOLCANO PHYSICS IN ACTION?

Robert Fritzius, Shade Tree Physics, Starkville, MS 39759

In January 2004 electrical appliances and pieces of furniture began bursting into flames in Canneto di Caronia, a small town on the northern coast of Sicily. A number of electronic devices in the area were also reported to have malfunctioned. As of the date of this abstract, the nature of these combustion/electronic events, which appeared to be initiated by some kind of strong electrical field discharges, but fueled by some, as yet unidentified, combustible material remains to be completely explained. A description of publicly available information about the fires, the local geographic/environmental setting, and related events will be presented. This will be followed by the author's armchair volcano physics version as to what may have been, and is still happening in the Mt. Etna environs. The volcano physics approach involves Mt. Etna's extended vertically plugged magma plumbing system that may have been reaching out horizontally to touch Canneto di Caronia.

2:00 Break

Manufacturing Physics

2:15 EFFECTS OF MOISTURE, SALTWATER, AND TEMPERATURE ON AN AOC P920 POLYESTER E-GLASS PULTRUDED COMPOSITE

Jarrad Zaiser* and Ellen Lackey, University of Mississippi, University, MS 38677

As engineers work to design bigger, better and more efficient structures, they begin to look for newer and better materials with which to do so. One field that has been getting a lot of attention is the composites field. There are many different composites that have and have not been examined by researchers and some have yet to be discovered. Some composites that have been discovered and experimented with have not yet been tested in every environmental and loading condition. This needs to be done so that the composites can be classified and put to use in the areas where they have performed well. The objective of this study was to experiment, analyze, and to document the results of how AOC P920 polyester/E-glass pultruded composite, a unidirectional fiberglass reinforced composite, will react to various environmental conditions. The conditions examined are elevated relative humidity, increased temperature, saltwater submersion, and preloading. These conditions were evaluated over several periods of time. The properties that were observed are load weight and flexural stress. The results show that the more moisture that is adsorbed the more the flexural stress is affected. Higher temperatures and relative humidity show this affect much sooner than for lower temperatures and relative humidity.

2:30 MERCHANT--THE PHYSICIST OF MANUFACTURING

S. Kant Vajpayee, University of Southern Mississippi, Hattiesburg MS 39406

The term manufacturing denotes here the "discrete" type. A major activity within this type is machining. Since the Industrial Revolution, for almost two hundred years, until the 1940s, machining continued to remain an art. Sixty years ago, M. Eugene Merchant made machining a "respectable" scientific field by investigating the physics of machined metal chip formation. He cleverly based his analysis on the "principle of minimum energy." The forming chips' shearing off from the parent material was illustrated to take place at a certain angle, which could be predicted. The presentation will explain the underpinning of his seminal work and highlight the concept it is based on, along with the concept's ubiquitous nature. The shear angle relationship proved as fundamental to the various machining processes as Newton's laws of motion to the mechanics. Merchant's pioneering work led to hundreds of Ph.Ds. in an area that was all along perceived to be non-academic. Merchant is not only the physicist of discrete manufacture but also its philosopher. He predicted the role of computers in this type of manufacturing at a time when they were being used primarily in academic research and business transactions. He is one of the only 25 celebrities inducted into the Automation Hall of Fame at the Museum of Science and Industry in Chicago. Even at 91 he has not retired, and continues to devote himself to the application of science to manufacturing.

2:45 CHARACTERIZATION OF PULTRUDED COMPOSITES WITH NATURAL FIBER REINFORCEMENTS

Brittany Hancock*, Ellen Lackey, James G. Vaughan, and Kapil Inamdar, University of Mississippi, University, MS 38677

The objectives of this research are to demonstrate the feasibility of using the cost-effective pultrusion* process to manufacture high quality polymeric composites using natural fiber reinforcements (NFR) and to quantify mechanical property variability for these composites. Natural fiber reinforcements offer value added properties such as the beneficial environmental impact of these renewable resources and the less abrasive nature of the fibers for reduced tool wear and less skin irritation for workers. Industrial hemp and jute, both bast fibers with good mechanical properties, was selected for examination in this research project. These initial pultrusion experiments utilized hybrid reinforcement of E-glass with either hemp yarn or woven jute mat. Pultrusion processing experiments and flexural strength testing results indicate that hemp and jute fiber can be successfully used as reinforcements for pultruded polymeric composites.

* A continuous process for manufacturing composites in rods, tubes, and structural shapes having a constant cross-section.

3:00 Divisional Business Meeting and Awards