|
Abstracts: Bowman Scholar Research Projects (AY2009)
Jacob R. Cates
Midshipman First Class
United States Navy
3-Paving Constants and the Kadison-Singer Extension ProblemOperator algebras were introduced by Murray and von Neumann in the 1930's. Besides being beautiful mathematical objects, rich with structure, they provide the correct mathematical framework for describing quantum phenomena. An important unsolved problem in operator algebra theory, posed by Kadison and Singer in 1959, asks whether every pure state on the diagonal operators extends uniquely to a pure state on all operators. As shown by Casazza et al., this very abstract problem is equivalent to many concrete problems in math and engineering, including important problems in signal processing.
In 1979, Anderson proved that the Kadison-Singer Extension Problem is equivalent to the (seemingly) more tractable Matrix Paving Problem. The latter asks whether there exists an 0 < ε < 1 and a natural number k ≥ 2 such that every zero-diagonal, norm-one matrix can be partitioned into k diagonal blocks, with each block having norm less than or equal to ε. Over the next decade, many mathematicians tried to solve the Matrix Paving Problem. In particular, Berman, Halpern, Kaftal, and Weiss and Bourgain and Tzafriri made significant progress. Nonetheless, the Matrix Paving Problem remains unsolved, and it is not even known whether k = 3 works (an easy example shows that k = 2 doesn’t work).
Since 2005, Weiss and Zarikian have been computing 3-paving constants in an attempt to show that k = 3 does not work. Prior to their work, the largest-known 3-paving constant was 0.6667. Now it is 0.8615. A 3-paving constant of 1 would show that k = 3 doesn’t work, and would be strong evidence that the Matrix Paving Problem (and therefore the Kadison-Singer Extension Problem) has a negative answer. The current project continues the work of Weiss and Zarikian, with the main goal of computing 3-paving constants for various unexplored classes of matrices.
Indirect Selective Laser Sintering (SLS) of Graphite Bi-polar Plates Faculty Adviser
Assistant Professor Vrej A. Zarikian
Mathematics Department
Marten V. Coulter
Midshipman First Class
United States Navy
for a Proton Exchange Membrane Fuel Cell (PEMFC)This research will involve a study that compares the material properties and performance of conventionally produced graphite bipolar fuel cell plates with those fabricated by a freeform fabrication method, in particular, selective laser sintering. The results of this analysis will contribute to ongoing research into the feasibility of replacing the conventional process of manufacturing graphite bipolar fuel cell plates with parts fabricated using a Selective Laser Sintering machine.
Selective Laser Sintering (SLS) is a type of fabrication that creates parts directly from three dimensional computer models. In this process, powdered material is rolled in layers, 0.003 to 0.007 inch in thickness, over a build area. A 30 watt carbon dioxide laser is used to trace a cross section of the part of corresponding thickness into the powder. The powder particles are fused together as a binder surrounding the powder is melted by the laser. The sintered layer is then lowered into a build chamber and a new layer of powder is rolled over the top. These steps are repeated until the part is complete in the chamber, with all features self-supported by excess powder. In Direct SLS the finished product is simply separated from excess powder and then removed from the chamber. In Indirect SLS the part must undergo post-processes to achieve desired porosity or other desiried material properties.
The Navy has supported the development and testing of freeform fabrication methods like SLS because of many potential benefits. With the ability to fabricate parts quickly from computer drawings, local inventories could be reduced to more common parts. Components specific to a certain system on board a ship or submarine could be generated as needed, saving the time, money, and space that is typically invested in large inventories.
This comparative study will concentrate on differences in material properties of components produced via Selective Laser Sintering (SLS) and properties for components fabricated using traditional manufacturing methods. Specifically, post-processing of selectively laser sintered graphite bipolar fuel cell plates for a Proton Exchange Membrane Fuel Cell (PEMFC) will be further examined and improved upon. Samples will be evaluated by conductivity testing, flexural tests, impermeability testing, and microstructural characterization.
Faculty Advisers
Professor Angela L. Moran and Professor Richard E. Link
Mechanical Engineering Department
Jeffrey R. Denzel
Midshipman First Class
United States Navy
Studies in the Photolysis of Nitroaromatic Compounds:
2,4-dinitrotoluene and 2,6-dinitrotoluene Determination of Rates and ProductsNumerous studies have been conducted concerning nitroaromatic compounds in fresh water or ground water. However, very little research has been done with nitroaromatic compounds in sea water. Nitroaromatic compounds are used as propellants by the military and are dangerous to both humans and the environment. In recent months, the National Oceanic and Atmospheric Administration has detected a significantly high level of two nitroaromatic compounds in waters of Ordnance Reef Wai’anae, HI. Out of fourteen nitroaromatic compounds screened for in the reef waters, only 2,4-dinitrotoluene (2,4-DNT) and 2,6-dinitrotoluene (2,6-DNT) registered a concentration above the detection limit. Even more alarming is the report that solid nuggets are washing ashore constituting solely of these two compounds and binding material.
The purpose of this experiment will be to determine the rate of decomposition of 2,4-DNT and 2,6-DNT due to sunlight in natural waters. The depth in the reef ranges from 24 feet to a maximum 300 feet and sunlight penetrates much of the water column, consequently the sun may have a significant impact on the disappearance of these two compounds through photochemical reactions. Conditions of the reef can be replicated in the lab using a SunTest CPS+Ò solar simulator equipped with various cutoff filters to select different regions of the solar spectrum. Additionally, this experiment will investigate the relation of several common aqueous substances present in marine water systems to the reaction rate. These substances will include, in no particular order, hydronium, nitrate, and carbonate. This will be done in order to determine if the presence or absence of a substance effects the disappearance of 2,4-DNT and 2,6-DNT. Moreover, a comparison of the photolysis rate in fresh water and sea water will be performed by comparing half-lives. The results in fresh water will be compared to the findings of previously establish research such as that of Zepp, et al. Methods of experimentation will closely resemble those used by Denzel, et al. For instance, irradiated samples will be analyzed by High Performance Liquid Chromatography (HPLC) in order to determine the degradation rate of the two chemicals. Lastly, if sufficient time is available, photolysis products will be isolated from salt water and analyzed by Liquid Chromatography Mass Spectrometry (LCMS) in order to determine their chemical identity. The necessity of this final stage of the project is due to evidence that nitroaromatic compounds can degrade into more dangerous compounds known as azo-compounds and azoxy-compounds.
Faculty Advisers
Professor Daniel W. O'Sullivan and Assistant Professor Dianne J. Luning Prak
Chemistry Department
Jacob P. Dobisesky
Midshipman First Class
United States Navy
Analyzing the Mixed Radiation Field Environments of Naval Aviators and AircrewSince the creation of our galaxy, the atmosphere has been constantly bombarded with galactic cosmic rays, or high energy, charged particles that go through various decay processes on their descent toward earth. These processes have created a time-varying, mixed radiation field throughout the entire atmosphere, which has shown to be strong enough to damage biological cells. As a result, national and international agencies have established guidelines and limits on the amount of radiation commercial pilots may receive. The United States has not yet created a limit for this type of radiation. This project will utilize a tissue-equivalent proportional counter to measure variations in total dose and dose rates that are experienced at various altitudes and locations in the atmosphere. The eventual goal in this area of research is to develop an accurate prediction code for radiation, accounting for solar variations, magnetosphere effects, and secondary particle build-up. This project will also seek to evaluate the military applicability and accuracy of current commercial, predictive codes that estimate radiation exposure received during flight, to form a more complete understanding of the complex radiological environment experienced by aviators and aircrew.
Faculty Advisers
Professor Martin E. Nelson
Mechanical Engineering Department
Visiting Professor Vincent L. Pisacane
Aerospace Engineering Department
Captain John W. Nicholson, USN
Weapons and Systems Engineering Department
Robert N. Harris III
Midshipman First Class
United States Navy
The Response of the MCNP5 Model for the DT-702 Thermoluminescent Dosimeter
to Rare Radioactive ParticlesThe objective of this research project is to develop and validate an accurate model of the Navy standard DT-702 thermoluminescent dosimeter (TLD) using Monte Carlo Nuclear Transport Code. A TLD is used to measure the radiation absorbed by personnel working in a radiated space. The DT-702 TLD contains four manganese-copper-phosphorus (MCP) doped LiF chips that are held on an aluminum card contained in a plastic holder. On the plastic holder, above each chip, is a different filter that is used to block different types of radiation.
Previous scholars, Travis Albright (2006), and John Hayashi (2007), modeled the TLD in an irradiation chamber at the National Institute of Standards and Testing (NIST). An experiment was conducted using a Cesium-137 source to irradiate ten TLDs to 500 mrems. Next the room, TLD and card holder were all modeled using Solid Works. Using Sabrina and Moritz, the model was translated into MCNP. However, due to the complexity and eccentricities of MCNP5, the model was full of errors.
The first task of this project will be to fix the errors in the NIST Cs-137 model. After fixing the errors, the model must be validated. Using the MCNP *F8 energy deposition tally the number of rads deposited per particle can determined. This value can be compared to the original test values obtained at NIST. Another method that will be used is to compare the energy deposition on each chip, with and without the chip filters in place. The values of energy deposition on the each chip without the filter in place should be approximately the same.
Once the Cs-137 model has been validated, it be tested using a Co-60 source. This will be accomplished by altering the source definition card of the Cs-137 model to account for an equal probability of a 1.173 MeV or 1.3325 MeV gammas being emitted. And finally, new batch of TLDs will be irradiated at NIST using their Co-60 source. The overall goal is to create a model of the DT-702 that can be used to accurately predict the TLDs response to any type of radiation.
Faculty Adviser
Professor Martin E. Nelson
Mechanical Engineering Department
William F. Jenkins II
Midshipman First Class
United States Navy
Discrete Active Disruption of Underwater CommunicationThis research project will serve to examine the feasibility of disrupting underwater communications in a manner which is both effective and discreet. Advances in signal processing and channel equalization in the underwater medium are yielding continuously more robust methods of underwater communications. As these technologies are introduced into the fleet in coming years, it is of strategic import that research be undertaken to counter this relatively new and emerging method of communication.
The immediate goal of this research is to construct a set of waveforms capable of disrupting – or jamming – waveforms used in the St. Margaret’s Bay UNET-08 experiments in June 2008. In UNET-08, various waveforms (QAM, PSK, CDMA) transmitted by a towed transducer were received by a moored array of hydrophones and subsequently recorded on hard drives aboard an adjoined buoy. These two groups of data allow for a comparison of waveforms as they were transmitted and as they were received. It is from these two groups of data that various analyses will be conducted to determine an ideal balance between sufficient interference and satisfactorily low signal-to-noise ratio input levels. Through examination of the UNET-08 waveforms in the time and frequency domains, the project will aim to determine methods which are suitable for disrupting the waveforms by, for instance, identifying and exploiting harmonic patterns.
Development of a Robotic "Sniffer-Dog" for Radioisotope Detection Faculty Advisers
Commander Edward J. Tucholski, USN
Physics Department
Commander Carl A. Hager, USN
Oceanography Department
Christina J. Moore
Midshipman First Class
United States NavyThe goal of this project is to make a High Pure Germanium (HPGE) radionuclide detection system ‘smart’. This project involves integrating a commercially available HPGE system with an autonomous robot, which could then be used to inspect an area for radioisotopes that might be used in a radiological dispersal device (RDD) or are classified as Special Nuclear Material (SNM), which is used as fuel in nuclear power plants and nuclear weapons. In the current security climate, such a machine would be extremely useful.
Designing and building a ‘smart’ machine that uses a HPGE detector presents a challenge for finding the right balance between speed and accuracy. The longer the inspection time, the more accurate the search will become, but the longer it will take to complete. A properly designed system would have a ‘search and detect’ mode which, when activated, would cause it to survey an area. At the end of the survey it would have reported the location and identity of every radionuclei of interest in the area to the handler. It would do this by either rotating the detector to survey the room and or using locomotion to get it closer to sources to improve signal quality. Stronger signals allow faster identification of an isotope with strong statistical confidence. Some key tasks in achieving project goals are to prepare the platform; mount the detection system; connect the detection system to a processor; program a method of bearing detection for 1D detection; and integrate this with a search pattern for 2D Detection.
Faculty Advisers
Professor Martin E. Nelson
Mechanical Engineering Department
Assistant Professor Randy P. Broussard
Weapons and Systems Engineering Department
Professor Mark J. Harper
Mechanical Engineering Department
Ryan J. Pifer
Midshipman First Class
United States NavySearch for Triaxial Deformation in Tantalum-167
The nuclear structure of tantalum-167 will be investigated by producing high-spin states through the use of a heavy-ion reaction. Emitted gamma rays will be detected in coincidence and these data will be analyzed at USNA. The nuclear “wobbling mode” is predicted to occur in tantalum-167 and may be observed through an analysis of the energy levels populated in the reaction. In particular, gamma rays associated with the proton (π) i13/2 sequence will be sought as this sequence is expected to be critical component for the wobbling mode to exist.
Nuclei can have many different shapes with various symmetries. Theory also suggests that a nuclear shape with no symmetry can exist and is said to have triaxial deformation. A stable triaxial deformed nucleus will produce a “wobbling” mode when it rotates. This is similar to spinning an asymmetric top and observing its wobbling motion. Only a few examples of this exotic mode have been observed in lutetium isotopes with neutron numbers of 90, 92, 94 and 96. Theoretical calculations indicate that the neutron number of 94 may be the main cause for the asymmetric shape that leads to a wobbling mode, which is the neutron number for tantalum-167. If found this would be the first isotope outside of lutetium isotopes to show this behavior.
To create these spinning nuclei two lighter elements are collided with a particle accelerator at very high energies creating angular momentum in the resulting nucleus. After a few neutrons are ejected, gamma rays are emitted to slow down the spinning nucleus, and these gamma rays can be detected using a germanium spectrometer. In the case of tantalum-167 a beam of 51V will strike a stationary target of 120Sn, forming 171Ta. The beam energy will be selected such that four neutrons will boil off most frequently giving 167Ta. This experiment was conducted at Argonne National Laboratory in March 2008 by Dr. Hartley using the ATLAS linear accelerator and the Gammasphere spectrometer (110 spherically arrayed γ-ray detectors). Data produced by this experiment will be analyzed, in collaboration with graduate students at other universities involved with the experiment. After the data are analyzed, one of two conclusions will be reached. If there is no wobbling sequence observed, then the particle-hole excitation theory that has been used to explain that wobbling will only occur in lutetium will be validated and no further investigation into other elements possessing a triaxial deformed nucleus will be performed. However, if the wobbling mode exists, it will be the first case outside of a lutetium isotope to have this occurring and investigation into the influence of the proton level density for wobbling will be conducted.
Faculty Adviser
Associate Professor Daryl J. Hartley
Physics Department
Bradley J. Ulis
Midshipman First Class
United States NavyAlternate Biometric Algorithm Processing Using Parallel Logic
in Field Programmable Gate Arrays (FPGAs)Within recent years, iris recognition biometrics (recognizing individual human beings based on iris characteristics) have become increasingly popular since the introduction of more reliable techniques for documenting minute details of the human eye. Iris recognition algorithms, however, require very powerful computing capabilities, making iris recognition systems difficult to deploy. Current Iris recognition processing relies on general purpose processors that are designed for a broad range of applications that include running a wide range of software and operating systems. This project seeks to design single-purpose hardware specifically for an existing iris recognition algorithm in an effort to minimize system foot print and reduce algorithm execution time.
Faculty Advisers
Assistant Professor Ryan N. Rakvic
Electrical and Computer Engineering Department
Assistant Professor Randy P. Broussard
Weapons and Systems Engineering Department
Last updated 27 October 2008
Back to Top