«Strengthening the Nation through Diversity, Innovation & Leadership in STEM San Antonio,Texas · October 3-6, 2013 Get Connected! Connect with the ...»
University of California, Berkeley, Berkeley, CA.
The combination of two different metals to form bimetallic nanoparticles can result in catalysts with improved reaction activity and selectivity. Conventional synthesis procedures, however, seldom distribute metals uniformly or require catalytically poisonous additives to prevent agglomeration. Impurities and compositional non-uniformity have prevented rigorous conclusions about the effects of alloying. We propose a colloidal synthesis method that has been successful in producing uniform bimetallic AuPt and AuPd nanoparticles (3-4 nm diameter). These particles are prepared using ethanol or ethylene glycol as reducing solvents and polyvinylpyrrolidone or polyvinylalcohol as protecting polymers. Such materials contain only C, H, O, and N, which can be removed by postsynthetic O3 treatment without significant change in size or composition. The formation mechanism was studied using in situ UV-visible spectroscopy and TEM. Results suggest a galvanic exchange formation mechanism. This mechanism involves the surface atoms of the seed nanoparticle (Pt or Pd) reducing Au3+ ions in solution onto the particle surface as Au0. This redox reaction is thermodynamically favored due to the higher standard redox potential of Au relative to Pt and Pd.
Pt2+ or Pd2+ ions generated by the reaction are then redeposited onto the surface by the reducing solvent. Developing rigorous conclusions about the size and composition effects of bimetallic clusters is key to designing valued catalysts for environmentally relevant reactions such as methanol oxidation in fuel cells and CO oxidation in automobile catalytic converters. Mechanistic knowledge of how to produce uniform clusters, such as those made in this study, is crucial to making such conclusions.
350 ORAL ABSTRACTS
ENHANCED CHARGE STORAGE CAPACITY OF CONDUCTING POLYMER ELECTRODES BY THE
ENTRAPMENT OF P-BENZOQUINONEMargarita Arcila-Velez, Mark Roberts.
Clemson University, Clemson, SC.
There is currently a strong need for high energy/high power storage devices for small- and large-scale applications such as transportation, load leveling, storage for intermittent energy sources, and portable technologies. The performance of current energy storage technologies (batteries and supercapacitors) falls well short of requirements for using electrical energy efficiently. Pseudocapacitors, a supercapacitor type, have the ability to store great amounts of charge due to the inherent properties of their faradaic storage mechanism. In these supercapacitors, the bulk redox material stores energy by a transfer of charge between the electrolyte and the electrode. Furthermore, lowcost, environmentally friendly, and highly abundant materials are necessary for feasible stationary energy storage.
Electroactive conductive polymers have the potential to address these major issues, along with having good intrinsic conductivity ranging from a few S cm−1 to 500 S cm−1 in the doped state, fast doping and de-doping processes, and ease of manufacture, particularly as thin films. Conducting polymer electrodes, however, still require improvement of their electrical charge capacity and stability. In this work, we report an increase in capacitance and charge capacity of conducting polymer electrodes through the entrapment of an abundant prototypical organic redox molecule such as 1,4-benzoquinone. This 1,4-benzoquinone is entrapped within a polypyrrole network through electrochemical polymerization, showing chemical and physical interaction with the polymer structure. This work will enable the development of supercapacitor materials that combine the high storage capacity properties of redox molecules with the high conductivity and capacitance of intrinsically conducting polymers such as polypyrrole.
A COMBINED ENCRYPTION AND AUTHENTICATION SCHEME FOR SECURE FIELD-PROGRAMMABLE GATE
ARRAY DESIGN AT REDUCED LATENCYJuan Portillo, Bao Liu.
University of Texas at San Antonio, San Antonio, TX.
Existing schemes for securing a field-programmable gate array (FPGA) design include encryption and authentication in separate steps wherein the FPGA chip contains a private key for decryption and a public key for authentication.
These cryptographic keys are subject to side channel analysis and other hardware security attacks. In this paper, we propose a combined encryption and authentication scheme, where an FPGA designer encrypts the bitstream based on its digital signature such that an FPGA chip only needs to contain a public key to verify the digital signature and correctly decrypt the bitstream. A supply chain adversary achieving the public key will not be able to program a new circuit on the FPGA chip. Our scheme is compared against the ElGamal scheme using ModelSim cycle accurate simulation to determine latency. Our scheme achieves a 12% decrease in latency. Our experimental results show that the proposed scheme achieves enhanced security against supply chain adversaries at lower execution latency.
We present a path planner that uses variational methods to construct 3D, locally optimal, collision-free paths for a curvilinear device that must reach multiple targets while satisfying a bounded curvature constraint. This planner is analyzed in the context of steerable medical needles used for minimally invasive surgeries. Our planner uses partial path retracing and produces paths that consistently cut less tissue than paths obtained without partial path retracing.
We capture the shape of the needle’s path and retracing information using a binary tree structure. Our approach accommodates topological changes in the tree structure by performing an initial node swapping optimization followed
by a shape optimization step. Results are shown for scenarios with 3 targets and compared with other planning approaches.
ENVIRONMENTAL ENGINEERINGRoom 213B
ESTIMATION OF PREFERENTIAL CONTAMINANT TRANSPORT IN KARST GROUNDWATER SYSTEMS USING
MIXED MODELS AND CONSERVATIVE TRACER APPROACHESAngel Anaya, Ingrid Padilla.
University of Puerto Rico at Mayagüez, Mayagüez, PR.
Karst groundwater systems are highly productive and provide an important fresh water resource for human development and ecological integrity. Their high productivity and anisotropy are often associated with conduit flow and high matrix permeability. The general objective of this work is to characterize transport processes in conduit and diffusion-dominated flow under base flow and storm flow conditions. This work focuses on the development of geohydro statistical tools to characterize flow and transport processes under different flow regimes and their validation with conservative tracers. Laboratory-scale geo-hydrobed models were used for this purpose. Experimental work entailed making a series of point injections in wells while monitoring the hydraulic response in other wells. Statistical mixed models were applied to spatial probabilities of hydraulic response to determine the best spatial correlation structure to represent paths of preferential flow in the limestone units under different groundwater flow regimes.
Transport experiments were also conducted using a CaCl2 tracer to validate the preferential flow path estimates and characterize dual porosity transport. Results show that the system can be used to represent the variable transport regime characterized by conduit and diffuse flow in the karst systems. Initial hydraulic characterization indicates a highly heterogeneous system resulting in large preferential flow components. Temporal distributions of Cl- in the model show varying responses according to the location of the sampling cluster and support the preferential flow network distribution generated from the statistical modeling, allowing identification of transport mechanisms. [This work is supported by the National Institute of Environmental Health Sciences (Grant Award P42ES017198).]
MATERIALS ENGINEERING (INCLUDING CERAMICS/TEXTILES)
COMPARISON OF COMPUTATIONALLY SIMULATED FISSION PRODUCT DISTRIBUTION WITH CORRELATIVE
CHARACTERIZATION TECHNIQUES IN SURROGATE NUCLEAR FUEL MATERIALSBilly Valderrama1, Hunter B. Henderson1, Lingfeng He2, Todd R. Allen2, Michele V. Manuel1.
University of Florida, Gainesville, FL, 2University of Wisconsin-Madison, Madison, WI.
1 During the fission process, uranium dioxide (UO2) fuel material is irradiated, forming fission products. The addition of these fission products alters the path phonons travel in UO2, adversely altering the heat transfer properties of the fuel material and overall fuel performance. To improve performance, a deeper understanding of the role of insoluble fission products, such as xenon (Xe), during microstructural evolution is critical. In this investigation, atom probe tomography (APT) is combined with transmission electron microscopy (TEM) to provide insights into the segregation behavior of fission products. Connecting these techniques with computer simulations of fission product distribution provides a more complete understanding of fission product migration during use in the reactor. In this investigation, Xe-ionirradiated cerium dioxide (CeO2) was used as a surrogate for irradiated UO2. CeO2 is a model material with properties similar to UO2 such as lattice parameter, fluorite crystal structure, thermal conductivity, and irradiation behavior.
Although each technique individually has limitations, greater insight into material behavior can be gained when pairing
experimental and computational techniques. This work will show the importance of combining correlative experimental and computational techniques to enhance the understanding of fission product behavior in nuclear fuel.
MECHANICAL ENGINEERINGRoom 208
DESIGN OF NETWORKED SYSTEMS USING DISTRIBUTED DELAYSMarcella M. Gomez, Richard M. Murray.
California Institute of Technology, Pasadena, CA.
The effect of delays in dynamics is understudied; correspondingly, current results for control of delayed systems are conservative and until now have been sufficient for the problems at hand. We believe a future need for extended work in systems with delays will arise in large-scale network systems due to the amount of information being processed and the potential for large distances between plants and controllers distributed through the span of a city for example.
It is a difficult task in itself to design distributed or decentralized controllers for nondelayed systems. It would be beneficial to deal with delays in a decentralized fashion independent of controllers. Thus far, we have investigated methods of analyzing stability and performance of systems with multiple and stochastic delays for scalar systems.
We have shown the potential stabilizing effects of adding delays in parallel to already delayed feedback systems using the Nyquist stability criterion. Furthermore, if delays are present, they are often not fixed, so we investigated stability of feedback systems with stochastically varying delays in discrete time. We arrived at a condition for pointwise asymptotic stability of these stochastic systems under the condition the delays are identically independently distributed at each time step. We explore how to extend and apply these results to the design of network systems with delays. Given the topology of a network system, we will investigate how to add the minimal number of edges to increase robustness, improve performance, and perhaps even achieve stability if the system is unstable.
A PERIDYNAMIC MODEL OF DIFFUSIVE FLUID FLOW THROUGH A DEFORMABLE POROUS MEDIAJason York, John Foster.
University of Texas at San Antonio, San Antonio, TX.
There are several engineering problems, such as hydraulic fracturing and carbon-dioxide sequestration, in which cracks in a porous media are initiated and driven by high pressure fluid flow. In addition to the challenges involved in accurately modeling such fractures, classical continuum mechanics introduces a singularity at the crack tip, which has proven to be problematic in predicting nonplanar crack propagation. The peridynamic formulation of solid mechanics is a relatively new, nonlocal approach that is continuing to display its potential in modeling material failure.
Recently, peridynamic theory has been applied to fields other than solid mechanics, such as heat transport and fluid flow. This work incorporates a nonlocal formulation of diffusive mass transport with the existing nonlocal formulation of solid mechanics. Preliminary results display the capability of modeling coupled, damage-induced flow and failure propagation in a reactive porous media. (Partially funded by NSF LSAMP-BD 1249284.)
MEDICINE (E.G., DENTISTRY, SURGERY, OPTOMETRY,VETERINARY) Room 212B
LONG-TERM OUTCOME OF BRAIN STRUCTURE IN PRETERM INFANTS: EFFECTS OF LIBERAL VS.
RESTRICTIVE TRANSFUSIONSAmanda Benavides, Peg Nopoulos.
University of Iowa, Iowa City, IA.