GRADUATE RESEARCH
INTERNAL FRICTION STUDY OF Sr Fe1.2 Co0.3 Ox A MIXED CONDUCTOR
Bryant J. Polzin, Dr. Robert P. Foley
Mechanical, Materials, and Aerospace Engineering Department
The oxide Sr Fe1.2 Co0.3 Ox demonstrates both electronic as well as ionic oxygen conductivity. Because of this conductivity mixture, Sr Fe1.2 Co0.3 Ox is finding application in high temperature electrochemical devices such as solid-oxide fuel cells, oxygen pumps, batteries, and oxygen-permeable membrane catalysts. In addition, Sr Fe1.2 Co0.3 Ox demonstrates structural stability in both oxidizing and reducing atmospheres. This stability offers one possible way to explore the different conductivities of Sr Fe1.2 Co0.3 Ox. We used an inverted torsional pendulum apparatus to investigate the mechanical spectra and material damping caused by oxygen rearrangement in the material. Results of damping versus test temperature and damping versus test frequency are used to categorize the energetics and kinetics of such oxygen rearrangements. These new results advance the understanding of the mechanism of ionic conductivity and demonstrate the usefulness of the internal friction method for investigating the microstructure of such materials.
DELINEATION OF ADSORBED-DESORBED ORGANIC SPECIES IN METHANOL ELECTRO-OXIDATION IN PROTOTYPE DIRECT METHANOL FUEL CELLS-AN IR-IR INVESTIGATION
Bhaskar Sompalli, Li Liu, Dr. G. Smotkin
Chemical and Environmental Engineering Department
Methanol oxidation in Direct Methanol fuel cells has an extremely complex mechanism with a network of paths leading to multiple intermediate species and products. Investigation of the process in-situ spectroscopically has enormous potential in terms of understanding the mechanism. But pertaining spectro-electrochemical work do not indulge in real fuel cells. The current work aims at investigation of methanol oxidation in a prototype gas phase Direct Methanol Fuel Cell for in-situ studies of adsorbed species coupled with on-line studies. The interesting results of this work indicate that adsorbed CO poison is more tenaciously bonded to the Pt electrocatalyst compared to a Pt-Ru alloy. The adsorbed CO peaks exhibits a distinct Stark effect and a coverage dependence. There is also a distinct effect of Ru on CO in alloys which shows lifetime inhomogeneous broadening due to increased backbonding between the Pt and adsorbed CO. This work also helps to confirm the bifunctional mechanism proposed for superior performance of the alloys over Pt.
POTENTIAL AND COVERAGE DEPENDANT PM-IRRAS STUDY OF COads ON SINGLE PHASE Pt, PtxRuy, PtxRuyOsz and Ru
Renxuan Liu, Hakim Iddir, Dr. G. Smotkin
Chemical and Environmental Engineering Department
The potential and coverage dependence of the stretching frequency of adsorbed carbon monoxide, on single phase arc-melted Pt, Pt50Ru50, Pt80Ru20, Pt80Os20, Pt80Ru10Os10, Pt65Ru25Os10, and Ru electrodes in 0.5 M H2SO4 was investigated by polarization modulated FTIR reflection absorption spectroscopy. The CO peak position, intensity, and bandwidth vary with substrate composition and potential. The onset of CO oxidation is at more negative potentials on alloy electrodes than on Pt, which results in enhanced fuel cell performance. The CO wavenumber shift is dependent on CO coverage and the content of alloy. The peak positions and bandwidths are discussed with Blyholder and Nørskov's theory.
METHANOL OXIDATION ON ALLOYS COMPRISED OF Pt, Ru, Os
Li Liu, Rameshkrishnan Viswanathan,
Dr. G. Smotkin
Chemical and Environmental Engineering Department
There are two parallel streams of research on the methanol oxidation, the study of model surfaces consisting of single crystal or polished polycrystalline surfaces and the study of highly dispersed catalysts incorporated into fuel cell membrane electrode assemblies (MEAs). In order to correlate the high scan rate data of polycrystalline smooth electrodes with the steady state voltammetry of the high surface-area fuel cell catalysts, the effect of spin-coating Nafion film on the smooth electrode surfaces was studied. Methanol oxidation currents on Nafion coated Pt and Pt modified electrodes (first scans), without the need for cyclic voltammetric activation and prior dissolution of oxophilic modifiers (Ru, Os) at potentials greater than 0.8 V, were greatly enhanced. At 50oC, Nafion-coated Pt-Ru-Os (65:25:10) has the best performance of methanol oxidation.
PROTON SPILLOVER PROMOTED ISOMERIZATION OF n-BUTYLENES ON Pd-BLACK CATHODES/NAFION 117
Maria Salazar, Dr. G. Smotkin
Chemical and Environmental Engineering Department
The non-Faradaic electrochemical modification of catalytic activity (NEMCA) is an effect ascribed to the electrochemically induced and controlled spillover of ions from a solid electrolyte onto a catalyst surface, it takes place at three phase boundary between the electrode catalyst, electrolyte and reactant vapor phase. We now report the first demonstration of NEMCA for non-redox catalytic reactions, specifically the isomerization of alkenes on high surface area Pd/C cathodes interfaced to Nafion with a Pt-black/H2 counter electrode. A Nafion electrolyte fuel cell assembly(FCA) was used to study current-voltage and conversion-voltage relationships in the isomerization of 1-butene to cis- and trans-2-butene.
UNDERGRADUATE RESEARCH
INCREASING MOLLUSCICIDAL EFFICACY OF ZEBRA MUSSEL CONTROL PROTOCOLS BY PHENOTYPIC ANALYSIS
Bonnie C. Welcker, Dr. Peter Johnson,
Dr. Ali Cinar
Chemical and Environmental Engineering Department
Zebra Mussels are extremely hardy, prolific, and adaptable aquatic invaders first reported in the Great Lakes in 1986. Municipal water treatment workers were the first to discover how difficult these mussels were to control. This may be due to the fact that European populations have formed distinct ecotypes which differ in their tolerances to various environmental parameters. Tracing European origins and North American dispersion patterns using Zhivotovsky's Index of Similarity illustrates a dual Great Lakes settlement from Lake Kara-Teren and Lake Kamshlybash, sub-regions of the Aral Sea, for lakes Erie and Ontario, respectively. Settlements in Lake Michigan are similar to a theoretically deduced population resulting from the mixing of the Erie and Ontario groups at the 0.995 level, indicating the successful hybridizing of previously separated ecotypes. Analyzing Ohio River settlements illustrates no hybridizing and thus strongly suggests human mediated transport as the primary method of new population initiation in this area. This is in conflict with the commonly accepted "down the Mississippi and up the Ohio" hypothesis. Statistical analysis of physical attributes and biomechanical characteristics indicate genetic influences, but confirm environmental factors as most significant in determining when to use chemical, physical, or mechanical control methods. Mechanical and physical control methods are suited to areas with no exposed limestone since comparisons document a statistically significant decrease in shell thickness to length ratios as well as crush strength numbers in these areas. However, the chemical control method best suited to all phenotypes is alkylbenzyldimethylammonium chloride at 4.5 PPM, which results in 100% mortality in 24 hours.
SYNTHESIS AND REACTIVITY OF [Mn(H2O)2V2O6]Thomas P. Hope, M. Ishaque Khan*
Department of Biological, Chemical and Physical Sciences
Following an established procedure, the mixed metal oxide compound Mn(H2O)2V2O6 (1) has been prepared by the reaction of potassium vanadate with manganese chloride in aqueous reaction medium. The Mn/O/V/O based framework structure of 1 consists of corner sharing {VO4} tetrahedra and {MnO6} octahedra. Since Mn(II) centers have removable aqua ligands pointing toward the interior of the cavities present in the structure, reaction of 1 with a series of monodentate- and bidentate ligands have been carried out in water and common organic solvents to explore the possibility of ligand exchange at Mn(II) center. While 1 did not show any appreciable reactivity toward methylamine or thiocyanate, it reacted with chelating ethylenediamine incorporating it into the framework. Reaction of 1 with acetylacetonate and tris(hydroxymethyl)ethane are also promising.
*Corresponding Author
ISOLATION OF EMODIN FROM CORTEX FRANGULAE FOR USE IN A SEMISYNTHETIC SYNTHESIS OF HYPERICIN - THE ACTIVE PRINCIPLE IN ST. JOHN'S WORT
Shakeila Thomas, Dr.Kenneth W. Stagliano
Department of Biological, Chemical and Physical Sciences
St. John's wort is a natural herb that comes from a yellow-flowered plant with the Latin name Hypericum perforatum. It has been used for centuries in the treatment of wounds, kidney and lung ailments, bed wetting in children, and anxiety. However, in more recent years, it has been used as an antiviral and antidepressant for mild to moderate depression. In order to study the chemistry of hypericin in more detail we required multigram quantities of the material. The compound is extremely expensive costing $80/1mg. Furthermore, the isolation of hypericin from Hypericum perforatum is low yielding with yields varying with the climate in which the plant is grown. A total synthesis of hypericin has been reported starting from emodin, but it is also very expensive ($60/250 mg). In this presentation we describe the isolation of emodin from common buckthorn bark (Cortex frangulae). Through our isolation work, we have been able to obtain almost 10 g of emodin from 1 lb of Cortex frangulae ($30/lb).
A NEW APPROACH FOR DETECTING SMALL PARTIALLY OCCLUDED ELLIPSES
Deepesh Jain, Dr. Nikolas P. Galatsanos
Electrical and Computer Engineering Department
Many biological images contain bacteria and other micro-cells that are mostly elliptical in shape. Apart form being elliptical, they are also small and partially occluded. Biologist and other health scientists have to manually count these cells for quantification purposes, which is a very tedious and error prone process. Automating the process of counting the number of ellipses in an image is a problem of great practical importance for many applications. Such applications can be found in the food industry where automatic inspection of samples requires counting of bacteria.
In the past a number algorithms have been proposed for ellipse detection. The Hough Transform (HT) and the Randomized Hough Transform (RHT) are the best known methods for this purpose. Both these methods are based on locating the centers of ellipses using the tangents on the ellipse boundaries. Thus, their success depends on the number of points available and the size of the ellipse. If the size of ellipses is small or in the presence of occlusion, the HT and the RHT do not provide accurate counts of the ellipses. To solve this problem, we propose a new method named the Increasing Decreasing Slope Transform (IDST). As the name suggests, this method relies on tracing the elliptical shape rather than locating the center of the ellipse using the tangents. We provide experimental results with both synthetic and real life biological images that demonstrate that the proposed approach works better than both the HT and RHT algorithms.
ON-LINE PLACEMENT EXAMINATION
Jessica Anderson, Dr. Jim Karagiannes,
Cheryl Caplan, Kee Choi
Department of Biological, Chemical, and Physical Sciences, Educational Technology Center
The rapid development of the Internet has lead to an increased capacity to distribute educational media and offer instruction electronically to distant locations. If courses are going to be taught over the Internet there must be a way to test the students at the remote sights. On-line testing is the logical way to accomplish this. Through the use of FileMaker Pro, Web Star, and Web FM an online testing environment was created.
FileMaker Pro served as a small database by allowing the creation of tables and relationships between those tables. Each quiz is one FileMaker Pro file (table). Code, scripts, and HTML are contained/imbedded in each of the FileMaker Pro files. These tables which include a question table, student information table, temporary and permanent answer tables, and test name table, were designed to follow this particular path of implementation.
1. A student logs into the student information table using his/her name and social security.
2. If the student exists, he/she may choose from the test options that are returned.
3. After choosing an exam, a row of empty answers is created in the temporary answer table containing his/her name and social security number.
4. After the student has completed the exam, he/she can then select the Done button at the end of the test. The answers are then scored.
5. The results are then posted along with a score to a permanent answer table. The test field is flagged for the completed exam.
THIN-FILM HETEROJUNCTION BIPOLAR TRANSISTOR CHARACTERIZATION
Mekita F. Davis, Dr. Joy Laskar, Carl Chun, Nicole Krishnamurthy, Edward Gebara
Electrical and Computer Engineering Department, Georgia Institute of Technology
Parasitic capacitances are extracted using a newly proposed small-signal equivalent circuit for the Heterojunction Bipolar Transistor(HBT). In order to monitor the increase of extrinsic parasitic capacitances, the parasitic capacitances are extracted from the HBT before the integration of the INP-based HBT on Silicon. Later these values will be compared to the parasitic capacitances after the HBT is integrated with Silicon. Using, MATLAB, a program is written using the s-parameters and other related information to execute the extraction method and find the values of the InP-based HBT parasitic capacitances before integration.
COMPUTATIONAL METHODS FOR NEAR EARTH ASTEROID MODELING
Jonathan B. David, Bryan J. Field,
Dr. Harold N. Spector
Department of Biological, Chemical, and Physical Sciences
A collision with even a small Near Earth Object (NEO) poses a direct threat to human civilization. There is strong evidence that the Earth has suffered several collisions with asteroids and comets in the past. In fact, it is believed that the extinction of the dinosaurs as the dominant life form on Earth was caused by such a collision. The existence of Meteor Crater in northern Arizona and the famous Tungusta event in Siberia show that collisions with smaller extra-terrestrial objects have taken place in our recent past. It is imperative to accurately model NEOs. To accurately model such systems, we had to consider several computational problems. These included solving the N-body problem, deriving a missile equation, and computer modeling different deflection schemes. By taking these considerations into account, a fairly good model can be constructed for modeling near Earth asteroids (NEAs). We find that of the one hundred and eight potentially hazardous asteroids (PHA) that have been discovered, there is a great danger for a collision within our lifetime. Moreover, after extensive modeling of the 1997XF11 asteroid, we find that it would only take a minor perturbation in the orbit of several different NEOs to create collision conditions.
CHARACTERIZATION OF THE FREQUENCY RESPONCE OF A MEMS ACOUSTIC/PRESSURE SENSOR
E. Soupos, A. Naguib, H. Nagib
Mechanical, Materials and Aerospace Engineering Department
A Micro Electro-Mechanical System (MEMS) based sensor has been developed for implementation as part of an array to measure the acoustic field generated at the lip of an axisymmetric jet during supersonic screech conditions. The sensor consists of a stress-compensated PECVD silicon nitride/oxide, 0.4(m-thick diaphragm in combination with four mono-crystalline ion-implanted p++ silicon piezoresistors.
Currently, the dynamic response of the sensor is tested. The sensor was subjected to periodical pressure fluctuations, generated by a slotted wheel rotating in front of a jet. The angular velocity of the wheel is controlled by means of a variable speed DC motor. Preliminary measurements have yielded frequencies of about 1500 Hz and a pressure amplitude of about 200 Pa. The sensor output was compared to the output of a larger silicon-based piezoresistive pressure transducer from Kulite (model XCS-062.) The sensitivity so far has been evaluated at about 0.6 (V/VPa, between the range of 0-300 Hz. Current experimentation involves reaching frequencies of 4-5 kHz.
VIRTUAL REALITY MODELING LANGUAGE MODEL OF THE ILLINOIS INSTITUTE OF TECHNOLOGY, MAIN CAMPUS
Alessandro Donati, Jonah Kadish, and Shawn Schiffner, Dr. James Roberge
Computer Science and Applied Mathematics Department
The Virtual Reality Modeling Language, or VRML, is a relatively new language that brings three-dimensional content to the World Wide Web. With the use of a VRML browser one can navigate through a virtual world from their own computer. The use of this new medium has provided a new way in which people can visit a location instantaneously, and without any physical transportation. Keeping with IIT's use of cutting edge technology, this has become an excellent way to demonstrate how innovative and technologically advanced IIT is, both to potential students and our peers. The automated tour is also a more effective and innovative alternative to show people the campus than the usual two-dimensional map, or slide-show, that is most implemented today. Thus by creating a VRML "world" of IIT, we are able to give prospective students, and other interested parties, an effective simulation of ones own presence on IIT's Main Campus. The IIT VRML "world" features all of the academic buildings, as well as a few of the well known landmarks. The model is linked to Hyper Text Markup Language (HTML) documents to provide a more in-depth presentation, complete with the building's architecture and it's current usage.
TYPHLITIS IN A TERTIARY TEACHING INSTITUTION
Amol Deshpande, Scott Otto, and
Dr. T. Saclarides, M.D.
General Surgery Department, Rush Medical Center
Purpose: To identify patients with typhlitis and determine incidence and outcome. Methods: A retrospective review of hospital ICD 9 codes for leukemia, lymphoma, necrotizing colitis, appendicits (NOS) (typhlitis and cecitis), HIV infection, aplastic anemia, and neutropenia was performed. Results: From 808 charts, we identified 7 cases of typhlitis in 6 patients (4 males and 2 females). The underlying diagnoses included osteosarcoma (3, anecdotal reports), leukemia (2), and lung cancer (1). Presenting symptoms (pain, nausea, vomiting, diarrhea, fever) developed 2-12 days after the previous dose of chemotherapy (mean 7.5 days). White blood cells (WBC) count was .084-2.4 mm3 (mean of 1.07/mm3), hemoglobin 7.3-13.8 mg/dl (mean=10.6 mg/dl), and the platelet count 1,715-203,000 (median= 21,000, mean=63,387). Management consisted of bowel rest, intravenous hydration, nasogastric suctioning, intravenous antibiotics, transfusions, and granulocyte colony stimulating factor. One patient underwent surgery for intraperitoneal air. Overall mortality was 0%. Literature search demonstrates an incidence of 10-24%, however, our results show major discrepancies: Leukemia (1.9%), lymphoma (0%), necrotizing colitis (0%), appendicitis (NOS) (3.8%), HIV infection (0%), aplastic anemia (0%), and neutropenia (0%). Conclusions: We conclude that typhlitis is a rare condition and surgery is unnecessary except in select instances.
ANALYSIS OF CARDIAC FUNCTION IN PATIENTS WITH HYPERTROPHIC CARDIOMYOTPATHY USING CINE MRI
Chethan Gangireddy, Dr. Andrew Arai,
Dr. Steven Wolf
National Heart Lung and Blood Institute, National Institutes of Health
Symptoms of congestive heart failure can be attributed to systolic abnormalities in the great majority of patients. A smaller percent of patients have congestive heart failure in the absence of systolic abnormalities. These patients are classified as having abnormal diastolic function. Patients with hypertrophic cardiomyopathy frequently have congestive heart failure but have normal or hyperdynamic left ventricular systolic function. A number of indices of diastolic dysfunction have been described in these patients. The purpose of this study was to develop indices of systolic and diastolic function obtainable in a cardiac MRI study. Patients with HCM were studied as they were presumed likely to have abnormal diastolic function. Primary attention was placed on the time-volume relationships across the cardiac cycle. In particular, a characteristic pattern of delayed relaxation extending from end systole into the early diastole was present in patients but not in normal volunteers. The project also describes mechanism that can distort left ventricular end diastolic volume estimates by 20-35% using breathhold segmented k-space GRE cine sequences and techniques to minimize this distortion.
NEUROBLASTOMA INVASION IS DECREASED BY TUMOR CELL GANGLIOSIDE DEPLETION
Amol Deshpande, Chethan Gangireddy,
Dr. Leonard Valentino
RUSH Children's Hospital, RUSH University
Neuroblastoma in children over one year of age typically presents as disseminated metastatic disease. The molecular and cellular determinants underlying this biological behavior remain obscure. More rapid disease progression and a lower survival rate are associated with high circulating tumor GD2 ganglioside concentration. We speculate that neuroblastoma tumor gangliosides play a role in accelerating tumor progression by promoting tumor cell invasion. To test this hypothesis, invasive tumor cells were cultured for 6 days in 10 ?M 1-phenyl-2-decanoylamino-3-morpholino-1-propanol (PDMP), an inhibitor of glucosylceramide synthase, to inhibit ganglioside biosynthesis. Invasion through PET membranes with 8 micron pores coated with 62.5 ug/cm2 Matrigel( was measured at baseline and following 3 or 6 days of exposure to PDMP. Following six days of culture in PDMP, total cellular ganglioside content of LA1-5n cells decreased to less than 2% of control cells: 0.76 nmol lipid-bound sialic acid (LBSA)/108 cells compared to 43.69 nmol LBSA/108 in control cells without PDMP. LA1-5n cells cultured with PDMP for 6 days were less invasive (38%) compared to control cells not exposed to PDMP. Despite a 75% reduction in cellular gangliosides after 3 days of culture with PDMP, no difference in the invasive phenotype of LA1-5n cells was observed. These results demonstrate that: 1) PDMP is an effective pharmacological agent to rapidly decrease cellular ganglioside content; 2) Reduction of the cellular ganglioside content to less than 2% of baseline reduces the invasive phenotype of tumor cells. Our data supports the notion that gangliosides play a role in the invasive and metastatic phenotype of neuroblastoma.
GROWTH CONDITION OPTIMIZATION FOR GENETICALLY ENGINEERED E.COLI CONTAINING pUC 8:16
Roberto Alvarez, Dr. Benjamin Stark
Department of Biological, Chemical, And Physical Sciences
In the last 20 yrs. many breakthroughs in genetic engineering have given us the capabilities to develop plants, animals, and microorganisms that are not found in nature. These developments have helped in many ways, some examples of which include productive plant growth in harsh conditions, rapid protein and vitamin production to keep up with demand (insulin for diabetics), and detoxification of toxic compounds found in places such as landfills.
Vitreoscilla, a Gram-negative, obligately aerobic bacterium, is a microorganism that has been useful in this field of study. Although this species needs oxygen in order to survive, it is found in oxygen lacking habitats, such as cow dung, rotten vegetables, etc. However, it is able to survive in such conditions because it possesses a gene that encodes for hemoglobin. Bacterial (or Vitroescilla) hemoglobin (VtHb) acts as an oxygen carrier, binding oxygen at low extracellular concentrations and buffering it at high intracellular concentrations. Scientists have been able to isolate this gene and insert it into other useful organisms in order to improve their growth and productivity.
The study of VtHb at the molecular level is important for the overall understanding of VtHb effects. This requires the purification of large quantities of pure VtHb. Genetically engineered E. Coli containing the VtHb gene produces these large quantities. E. Coli is used for two reasons, first, they make a lot of VtHb, and second, large quantities of this protein are found in the periplasm, this in turn aids in the purification process. The VtHb in the periplasm can be extracted by lysis of the out cell wall using a hypotonic solution, without rupturing the inner wall, keeping the cell in tact (this means that there will be no contamination of intracellular proteins in solution). The amount of VtHb protein can be determined by a simple ratio OD410nm/OD280nm, the 410 reading measures the VtHb while the 280 reading measures the total protein concentration in solution.
This project aims to identify growth conditions that will optimize this procedure, i.e., that produces the highest 410/280 ratio in the periplasmic lysate. This can be accomplished by varying growth conditions such as, aeration (shaking speed), incubation (growth period), and temperature. The over all procedure consists of lysing the cells with a hypotonic buffer, spinning down the cells, saving the lysate, and assaying it by taking OD410nm and OD280nm readings.
DETERMINATION OF THE REST MASS OF THE K ( MESON VIA CHERENKOV RADIATION RING IMAGING
Michael S Hreczkosij III, Dr. Kaplan
Department of Biological, Chemical, and Physical Sciences
The K ( meson is the bound state of the strange and up quarks. Measurements of the decay modes for the K ( have proved invaluable in the development of the Standard Model as well as illuminating new physics. Previous measurements of the mass of the K ( meson utilized the method of spectrum analysis of Kaon atomic decays. The six best measurements have resulted in a bipolar mass distribution, inconsistent with the Standard model. I am attempting to measure the mass more accurately by using a Ring Imaging Cherenkov (RICH) detector. The RICH detector measures mass by reflecting the Cherenkov radiation given off by subatomic particles with a large spherical mirror. The radiation is focused as a ring onto a photodetector. The radius of the ring is a function of the mass and momentum of the particle. By measuring the momentum separately, via bending magnets, it is possible to obtain a determination of the mass. The analysis is ongoing presently so any prediction of success or failure would be mere speculation.
DATA COMPRESSION FOR MULTIMEDIA APPLICATIONS
Tayo Ihimoyan, Dr. Yongyi Yang,
Dr. Nikolas Galatsanos
Electrical and Computer Engineering Department
With the explosive growth in the volume of data storage and interchange today, there is a growing need to compress data in order to reduce their volume for both cost-effective storage and transmission. Data compression seeks to reduce the number of bits used to represent information. Our focus is in developing better data compression techniques for multimedia application such as video conferencing, audio transmission over the internet or intranets, etc. Presently, we are focussing our efforts in developing the necessary background for data compression of text streams. These techniques include Huffman coding, arithmetic coding, Lempel-Ziv algorithms which form the foundation of many modern lossless coding algorithms. Our program attains an average compression ratio of 35% on compression of text streams, as compared to the normal ASCII code format for text files. We intend to use the study of the lossless compression methods for text streams as a stepping stone to understanding the compression methods involved when dealing with waveform data such as voice, moving images (video) and their transmission over local and wide area network.
TEST OF A HOLLOW BEAM
German Avila, Dr. S.A. Guralnick
Civil and Architectural Engineering Department
The goal of this project is to investigate and analyze the load-deformation behavior of an actual hollow steel beam. The investigation was achieved by a real-time measurement of the variation of strain and deflection with increasing the load on the beam. State-of-the-art strain gage technology along with a computer-aided data acquisition and handling system were used for this purpose. The experiment used a 4"x4"x3/8" steel section. Electric resistance strain gages were placed on the centerline every half-inch along the side web and every one inch on the top and bottom flanges. A deflection transducer and a load transducer were also attached. A manual hydraulic pump was used to apply the loading. A "Vishay Measurement Group" System 5000 strain gage channel scanner and data acquisition system was used to collect and analyze the data. Agreement between experiment and theory was found to be satisfactory. Moreover, performance of the computer-aided data acquisition system was also found to be satisfactory.
REMOTE LABORATORY OPERATION
Orman C. Greaves, Jr, Dr. Michael Hites
Mechanical, Materials and Aerospace Engineering Department
The conventional method of performing laboratory experiments requires the individual(s) performing the experiment to be local. If the facilities for the experiment are not readily available, the experiment would have to be performed at another location increasing the amount of time, energy, and money spent. Academia and industry have historically been bound by this method to perform data acquisition and experiment control. The National Diagnostic facility has considered a new unconventional method of performing data acquisition and experiment control by using the Internet. This unconventional method can be described as Remote Laboratory Operation. It will allow one to control a laboratory experiment and retrieve the results via the Internet. The current working demonstration models were written using the Internet Toolkit for LabVIEW( from National Instruments. The working demonstration are the Online Can-Crusher, which demonstrates remote experiment control, and the Oscillating Fan Demonstration, which demonstrates data acquisition. The working models incorporate the following: Image acquisition from a CCD camera, Analog signal and processing, Analog signal generation for control and Multiple digital port activation for logic can control circuits. With remote capabilities, laboratories are not limited to local experiments; students and faculty can from any computer connected to the Internet perform experiments. Industry and academia have already used these techniques to expedite collaborations on research.
THE SYNTHESIS OF HETEROAROMATIC DIKETONES AND THEIR USE IN THE SYNTHESIS OF HETEROAROMATIC SUBSTITUTED
TETRACYCLONES (HETEROCYCLONES)
Jackson M. Ndiho, Cathy Castro,
Christopher J. Walsh, Dr. Braja K. Mandal
Department of Biological, Chemical, and Physical Sciences
The synthesis of several novel heteroaromatic compounds containing 2,4-cyclopentadiene-1-ones (heterocyclones) has been studied. The heterocyclones were formed in a Knoevenagel-type condensation reaction between the appropriate diketones and commercially available 1,3-diphenylacetone. The diketones were synthesized in a separate step between heteroaryl organometallic reagents and the electrophilic 1,2-dicarbonyl synthon N,N'-dimethyl-piperizinedione (DMPD). The organometallic intermediates were formed via metal-halogen exchange between n-butyllithium (n-BuLi) and aryl bromides or iodides. The DMPD was synthesized in a concerted cyclization reaction between diethyloxylate and N,N'-dimethylethylene-1,2-diamine in 95% yield. Starting from 2-bromothiophene, 3-bromothiophene or N-hexyl-3-iodocarbazole in tetrahydrofuran, metal-halogen exchange with n-BuLi, followed by the reaction with DMPD gave 2,2'-thiophyl, 3,3'-thiophyl, or 3,3'-carbazyl in 55%, 70%, and 50% yields respectively. Condensation between the diketones and 1,3-diphenyacetone in triethylene glycol at 100oC using catalytic amounts of the basic phase transfer catalyst, benzyltrimethylammonium hydroxide, produced 3,4-di(3-thiophene) I and 3,4-di(2-thiophene) II, substituted heterocyclones in 40% and 35% yields respectively. Unfortunately, the 3,3'-carbazyl failed to condense under the employed conditions to form III. The heterocyclones were purified using flash chromatography with silica-gel as the solid phase and mixtures of methylene chloride-hexanes as the mobile phase. The purified compounds were characterized using FT-IR and UV-VIS spectroscopies and elemental analyses. The analyses confirm the structures given below. These heterocyclones can act as dienes and undergo efficient Diels-Alder cycloaddition reactions to produce, when reacted with derivatives of acetylene, all-aromatic materials that are of interest in the study of conducting polymers and nonlinear optics.
EVALUATION OF EMBOLIZATION PARTICLES USING ON-LINE LIGHT
SCATTERING PARTICLE SIZE ANALYZER IN NORMAL SALINE AND
NON-IONIC CONTRAST MEDIUM SOLUTION
Michael Gallegos, Dr. P. Rasuli, Dr. A. Oskouie, Dr. N. Aderangi
Chemical and Environmental Engineering Department, Department of Radiology, Ottawa General Hospital, University of Ottawa
Polyvinyl Alcohol (PVA) particles are frequently used by vascular radiologists for transcatheter embolization or internal plugging of blood supply to hypervascular or neoplastic tissues. These plastic polymers are also used for controlling hemorrhage from the uterus and bladder, as well as endovascular management of arteriovenous malformations.
Size distribution of PVA particles usually is performed using sieve analysis, and the size range of each fraction is confirmed with scanning electron microscopy. This off-line method of size classification of particles is labor intensive, and expensive.
This research introduces an on-line particle size measurement procedure based on light scattering properties of the embolization particles. In this study, a particle size analyzer, MICROTRAC SRA-150 was calibrated and used. Since particles are circulated through a loop in this instrument, it is feasible to closely simulate blood circulation in the human body. Different manufacturers provided PVA particles with different size ranges for this experiment. These particles were suspended in 50% non-ionic contrast medium and 50% normal saline (0.9% NaCl) Solution. Based on our studies a noticeable number of particles were larger than the labeled size ranges by the manufacturers. The results of our study clearly demonstrated that more research is necessary to establish the level of accuracy in labeled particle size distribution necessary for these highly sensitive medical applications.
IPRO PROJECTS
AUTOMATED PATIENT MONITORING AND DIAGNOSIS BY STATISTICAL AND ARTIFICIAL INTELLIGENCE TOOLS
Neil Aggarwal, Michelle Chesebro, Jeffrey DeCicco, Jennifer Evans, Raghu Raj,
Erika Restrepo, Miraj Shah-Khan, Eric Tatara, Andrew Zloza, Dr. Ali Cinar
Chemical and Environmental Engineering Department
Patient monitoring by automated data collection has created new challenges for health care professionals in their efforts to extract useful information from raw data. New on-line monitoring devices may generate large amounts of data that must be interpreted quickly and accurately. The team investigated the use of statistical methods and artificial intelligence tools such as knowledge-based systems (KBS) and artificial neural networks (ANN) to summarize and interpret data generated by physiological variables that are measured at high frequency, with the refined information to be subsequently merged with the results of laboratory analysis that are reported periodically. The development of a methodology and its associated tools for real-time patient data monitoring and diagnosis
was accomplished by using the commercial programming environments Matlab and G2 (a real-time KBS development shell). The electrocardiogram (ECG), which measures the electrical activity of the heart, was used to detect cardiovascular anomalies through the use of digital filters and neural networks. A KBS was developed which incorporates the various statistical tools with a rule-based decision system. The performance of both the individual tools and the KBS are compared through a case study. The benefits of the methodology and associated tools that will be demonstrated through this project are to detect abnormal situations, to provide preliminary interpretation and diagnosis, and to report these findings to the health care professional.
CLASS PROJECTS
DESIGNING A SIMPLE COMPUTER FOR CS 350
Willis Redd, Dave Pantzer, Jason Singer,
Dr. Morris Chang
Electrical and Computer Engineering Department
This course project is for CS 350, Computer Organization and Assembly Language Programming. This course introduces the MIPS assembly language and the basic structure of the MIPS architecture.
We have developed a simple, general purpose eight-bit microprocessor, whose design was strongly influenced by the MIPS RISC family of processors. This single chip computer has four registers, 512 bytes of RAM, and an eight-bit wide data path. We have named our architecture IITER, short for IIT Eight bit RISC. IITER was designed with Altera's Max-Plus 2, using schematic capture tools for design, and simulation tools for initial testing. The final project was tested with Altera's prototype board, a reconfigurable programmable logic device (PLD) that connects to a PC via a parallel port.
To facilitate the development of applications for our machine, we have created a simple
"compiler" that translates mathematical expressions into IITER assembly language.
We have also developed an assembler that translates IITER assembly language into IITER machine language. Using these tools in combination allows us to easily demonstrate our machine's ability to process simple mathematical expressions.
At the conference, we will present a poster and web site that detail our project, and give a live demonstration showing the compilation, assembly, and execution of simple programs.
SPORTS DATA MINING
Peter Radlinski, Tom Grzesik, Anthony Benton, Dawn Cox, Dr. Cindy Hood
Electrical and Computer Engineering Department
Data mining is the process of finding certain patterns in large groups of data which a human would normally not be able to find. These patterns may be used to predict future outcomes. For sports, data mining can be used for a wide variety of things. For example, predicting which team will win, what players to put in the game at certain situations, and the performance of a player against a particular team. This project will be trying to predict the outcome of basketball games. Specific team statistics are needed to allow the data-mining program to find patterns. Since there are so many statistics available, it was required to pick only the most useful data. A few examples of the data that we picked are: overall team shooting percentage, points allowed against each team, and home or away loses and wins. After finding the data, the data is then entered into text or excel files. After all of the data is entered, the data- mining program is then started. The program takes the excel or text files and looks through all of the existing data. The program will eventually find patterns in the data. From these patterns, we determine what can be predicted and if it is useful.
ROBOTIC ARM MOTION DEMONSTRATED BY COMPUTER SIMULATION OF THE KNIGHT'S TOUR
Mohammed Alam, Rahul Kumar, Chris Reiland, Dr. Cindy Hood
Electrical and Computer Engineering Department
The movement that a robot arm must undertake in order to complete any motion is a complicated process dealing with motors and gears. The software that prompts the arm to move is even more difficult to create. Such an arm can be simulated, though, as a way of demonstrating what the physical aspect of robotic action would require. A robotic arm can be used for many things, one of which is entertainment, as in a game. Motion structured around a grid is a useful example, because it allows the movement of the arm to be plainly visualized. A chessboard is an example of a grid-like setting on which the motion can be based. The "Knight's Tour" is a game that involves sixty-three different movements for the arm. It is a game in which a person puts a knight, black or white, on almost any space, and the arm moves that piece around the entire board, touching every space only once. A computer program can be designed using an algorithm that determines where the knight's next position on the board should be, in order to complete the tour. The arm can be implemented into the program, so that its motions are simulated by computer visual graphics or descriptive statements telling the user(s) what action is being taken, and how the robot arm is moving via physical measurements. The "Knight's Tour" allows the robotic arm to demonstrate the difficulties of robotic motion in a three-dimensional setting.
INTELLIGENT STOCK MARKET ANALYSIS FOR INTEL CORPORATION Samir Patel, Yogesh Bheda, Prathaban Thamotharampillai, Dr. Cindy Hood
Electrical and Computer Engineering Department
In the world today, there are millions who invest in the stock market, hoping to earn more than 6 % interest. However the chaotic behavior of the market can be a very dangerous investment. To safely invest means a lot of research with data. An investor has to spend weeks sifting through millions of pages of financial data just to find something useful. Once the investor has the data, more problems arise. Analysis of the data by a human mind can find only so many patterns. Results of an analysis can be a potential bust. The wild nature of the stock market can easily break the chains of the financial analysis.
A better analysis can be made without the waste of time and money. The procedure of data mining can be implemented to quickly get results of a financial analysis. Data mining is perfect for a financial analysis because the amount of financial data that is available. The goal of data mining is to extract previously unknown, comprehensible, actionable information from the increasingly large amounts of data being collected in industry, government and research organizations. The continuous evolution of the field, including future research directions, will be heavily influence by the experiences in applying data mining techniques to real-world problems. Data mining applications vary greatly today and the field can learn important lessons from this variability.
For this financial analysis project, our focus is on predicting the fluctuations of the stock price for the Intel Corp. This was accomplished by collecting a large body of data about the Intel Corporation and processing this information using the Datascope data mining tool. By doing this we are able to detect patterns in the data that can help in the prediction. The data that will be used is as follows:
* Net revenues
* Earnings per share
* Return on average stockholders' equity
* Stock price trading ranges by fiscal year
* Book value per share at year end
* Capital additions to property, plant and equipment
* Geographic breakdown of sales
* Employees at year end
* Research and development
* Common shares outstanding
Actual shares at year-end
Weighted average common & common
Equivalent shares for the year
The results using various data sets and time periods are presented.
IMPROVING CNS
Marvin Molina, Robert Venal, Jennifer Peacock, Drew Johnson, Mo Yuan, Dr. Cindy Hood
Electrical and Computer Engineering Department
CNS has implemented or will implement various changes to the computer and network services at Illinois Institute of Technology. These changes were taken from the Strategic Plan for Computing and Network Services, which outlines CNS's budget, projected expenses, and goals for the years 1996 through 1999. Little student feedback was received by those creating the Strategic Plan for Computing and Network Services, so the plan does not take into account the views and opinions of the students. This was due to a lack of response from the students to CNS's attempts to obtain feedback. Students form one of the largest blocks of computer users on the campus (not only for personal use but also for class). A survey was distributed by students in order to form an accurate opinion of student's views on CNS and what improvements need to be made to both CNS and to computing and network services in general at IIT. From the analysis of the results of the survey, recommendations for improvement have been formulated. These recommendations were designed to fit within the budget and goals of the existing plan. The value of this study is twofold: (i) the compilation and analysis of student input on computing and network services at IIT, and (ii) a realistic set of recommendations to augment the strategic plan currently in place. The hope is that these results will be given serious consideration by those responsible for IIT's computing environment.
TRACKING THE ACCURACY OF WEATHER FORECASTING VIA THE WWW
Steve Bidny, Mauro Jasse, Tony Jirik, Mike Mulligan, Tim Webb, Matthew Bauer
Computer Science and Applied Mathematics
Department
The digitized information available on the World Wide Web (WWW) enables users to gather and analyze data and make informed decisions as never before done. We are going to explore one such opportunity for the general public. Our goal is to develop a WWW site (Weather Tracker), using current advancements in intelligent agents, that is able to track the daily weather forecasts of major weather services. Our WWW site will report the ongoing accuracy of the major weather services so the general public will be able to make informed decisions about the possible weather forecasts.
We will explore all tasks associated with designing, developing, implementing and maintaining the Weather Tracker application. This will include user requirements study, technical research and training, maintenance issues, and project documentation. All of our design documentation will be published on the WWW.
REGISTRATION AT IIT VIA THE WWW: A STUDENT PERSPECTIVE
Victor Hatsushika, Atul Jain, Dacia James, Amrit Sahasranamam, Neelu Sharma, Nikkia Williams, Tracey Williamson, Matthew Bauer
Computer Science and Applied Mathematics Department
Recent announcements by Student Services outlined that the forthcoming upgrade to SIS will eventually enable Student Registration via the WWW. As with any business process, simply replacing a real world, paper and pencil process with a virtual WWW-based one is taking a shortsighted view of the advantages of the WWW. We propose six functions of registration that we hope will be considered for WWW-based registration. We analyze each function for its benefits, challenges, and technical feasibility. The six functions are:
1. Check Against Student's Program of Study
2. Check for Course Prerequisites
3. Check for Closed Classes
4. Check Student Recommendations (see WWW Course Evaluation Application)
5. Check Course Scheduling
6. WWW-based Signatures for Approval
THE USE OF GLOBAL POSITIONING SYSTEM IN THE AIR TRAFFIC CONTROL AND SEARCH AND RESCUE INDUSTRIES: DATA MODEL AND USER INTERFACE
Gary Dollinger, Marty Paz, Matthew Bauer
Computer Science and Applied Mathematics Department
Global Positioning System (GPS) is a tool that can be used to locate the exact latitude, longitude, and altitude of any point in the world. GPS uses twenty-four Earth orbiting satellites that transmit a signal to be used by a receiver for calculating the position. The position is highly accurate within ten meters.
Using the GPS system, an aircraft can determine its exact location. This location can be sent to an air traffic controller with the use of a transponder. The transponder sends the exact latitude, longitude, altitude and airspeed to the controllers' screen where it can be used in the process of conducting traffic.
With this system, radar would no longer be needed and efficiency will be raised to 99.8%. The new system would prevent the constant errors caused by radar and would give the controllers dependable information.
In the event of an air crash, the coordinates can be used to dispatch a ground term to the exact location of the crash site. The existing system of Emergency Locating Transmitters would take up to several hours to locate a site. The new GPS coordinates can be used to pinpoint the crash to the exact point.
This project is a continuation of previous research by Gary Dollinger and details the Data Model and User Interface.
WWW COURSE EVALUATION APPLICATION
Evridiki Efstathiou, Maya Gopalan, Anagh Gupte, Suneeta Julius, Mehruz Kamal, Mark Vennekotter, Michael Lee, Matthew Bauer
Computer Science and Applied Mathematics Department
The World Wide Web (WWW) has long been touted for being the ultimate in information sharing and gathering mediums. One detriment to its use, however, is the fact that the anonymity allowed is often cause for lack of reliability in WWW appended resources. This project, with a careful blend of information gathering and filtering techniques, will embrace this anonymity to allow students to offer up their opinions and suggestions about classes and professors at IIT, effectively producing a self-updating database of course evaluations.
All tasks associated with designing, developing, implementing, and maintaining the WWW Course Evaluation Application will be documented, which will further serve to easily track changes in the steps to gather and process information off the WWW. Issues faced will include the technical matter of combining WWW technologies such as online forms and CGI (Common Gateway Interface) scripting. A detailed user requirements study to allow easy access to data from various department offices will also be conducted.
INTERACTIVE VRML/JAVASCRIPT CHESS
Carl Lebsack, Michael Lee, Matthew Bauer
Computer Science and Applied Mathematics Department
An area in which many computer giants such as Microsoft Corporation and Sun Microsystems have recently invested much time and effort is the research and development of programming interfaces for distributed applications, wherein separate programs working on different operating systems and platforms will be able to communicate through a common interface. Part of this project will be to utilize one such schema so that a client chess program running in a browser will be able to communicate with a chess engine running on a remote server.
The Interactive Chess game will bring together a large number of "hot" internet technologies, including VRML (the Virtual Reality Modeling Language, version 2.0), JavaScript, and CGI (Common Gateway Interface) programming in Perl. The 3D interface will be created in VRML, and will thus allow dynamic zooming and rotating of the chessboard - the chessboard will also be uniquely interactive, through the use of JavaScript, which will determine the moves based on mouse clicks. CGI will be used to submit moves to and transmit countermoves from the server on which the chess engine resides.
GRAPHICS FILTERS- DESIGN AND IMPLEMENTATION
Jason Jasinek, Paul S. Yuen, Tomas Lin, Michael Lee, Matthew Bauer
Computer Science and Applied Mathematics Department
The development of graphics filters which can be found in the majority of graphics applications such as Adobe PhotoShop and Corel Draw is a process which requires a carefully written test base, which allows swift testing and redesigning of the actual filter itself. The importance of this is understood once we take into account the necessary overhead of loading and displaying an image and applying the filter to the individual pixels- a process which is necessary for an actual test-run.
The goal of this project is to come up with a versatile and practical shell program, which provides all the necessary functions that need to exist for a graphics filter testing program. The team will need to provide well suited procedures to serve them.
This project will inherently be an exercise in object reuse, since it is required that the same shell program be used for a multitude of filters. As portability is also key in the development of filters- developers may favor certain platforms- the program itself will be written in the Java programming language, which has much hype surrounding its "write once, run anywhere" property.
JAVA WEB CLIENT/SERVER
Pratik Sonawala, Gauvang Amin, Omran Mansoor, Masato Imafuku, Kazushi Oenoki, Michael Lee, Matthew Bauer
Computer Science and Applied Mathematics Deparment
The Java programming language is a convenient language in which to write applications that follow the "write once, run anywhere" convection, which would be an extremely great asset for any software engineering firm which has a large part of its software life cycles the porting of certain software products to other operating systems on different types of hardware. The way Java accomplishes this is by way of the Java Virtual Machine (JVM), which runs portable Java bytecode on a given machine.
This project will focus on developing a multi-platform application that supports communication between two or more hosts across a network , using the TCP/IP protocol- development will be made at the applications level of the TCP/IP reference model. The advantages of such an application include ease of maintenance (given the straightforward sockets implementations in Java), and portability. The latter is of extreme significance, given the often native machine specific programming involved in network applications.
Issues other than writing the application itself which will be involved in the project include researching the TCP/IP reference model, designing and implementing an intuitive user interface and investigating issues related to further uses of networked applications.
WIRELESS LAN's
Shahbaz Dakhan, James Muturi, Timothy Gruszynske, Hashim Dempsey
Electrical and Computer Engineering Department
Wireless LAN's are a relatively new technology that offers several advantages over the existing wired networks. Our problem was to investigate the feasibility of this technology; And if feasible, which of the available wireless technologies we would use in a model setting such as IIT.
There are basically two wireless LAN's standards. One of them is based on transmission in the radio waves frequency range which generally provides us with a transmission rate of about 2Mb per second. It is relatively cheap, widely used but is prone to a wide range of interference sources. Secondly, there is a LAN standard based on transmission in the microwave frequency range which offers us a greater level of security as well as a data transmission rate of 20Mb per second. However, the latter wireless standard is at the moment quite expensive and not as widely used as the former but its transmission rate compares favorably to that of the normal wired Ethernet, which is about 20Mb per second. Due to its relative newness, we cannot vouch for its reliability.
Based on cost and reliability concerns, we determined that a full deployment of wireless system is not feasible, but we recommend a trial deployment in certain areas to complement the existing wired network. At the moment, these areas are not served by the wired network as it would not be feasible to do so since the cost would be far much greater than the benefits derived from this service. Hence, the wireless network with its low cost of installation and flexibility would help "wire" previously "unwireable" areas.