The work presents three multiscale models, including sequential, invasive, and concurrent models, for fracture analysis of a crack in a two-phase, functionally graded composite. The models involve statistical description of the particle volume fractions, particle
locations, and constituent material properties; a two-scale algorithm including microscale and macroscale analyses for determining crack-driving forces; and two stochastic methods for fracture reliability analysis. Numerical results indicate that the sequential and invasive multiscale models are the most computationally
inexpensive models available, but they may not produce acceptable probabilistic characteristics of stress-intensity factors or accurate probability of fracture initiation. The concurrent multiscale model is sufficiently accurate, gives probabilistic solutions very close
to those generated from the microscale model, and can reduce the computational effort of the latter model by more than a factor of two.
Ming Yang
Biostatistics Computing shortest confidence interval from an indirect approach
The confidence intervals introduced in most statistical textbooks do not necessarily have shortest interval length. Plenty of work has been done to minimize interval length with coverage probability fixed in advance. Then optimization problem can be converted to solving two resulting equations, including an integral equation. However, solving those two equations simultaneously is difficult and that is why shortest confidence interval is seldom included in textbooks. This article presents an indirect approach to compute shortest confidence interval and we need only solve a single equation. An example is presented to compute the shortest confidence interval of normal variance. Since the computation is greatly simplified and does not rely on a high-speed computer, it is recommended the shortest confidence interval be covered in classroom presentation.
Interaction of Shocks with multi-material interfaces is an important phenomenon in several applications, including high speed-flows with droplets, bubbles and particles, detonations and hypervelocity impact and penetration. To simulate such complicated shock diffraction phenomenon, a fixed Cartesian grid approach in conjunction with level set interface tracking is attractive. In this framework, the challenge lies in accurately and robustly handling shock-interface interactions, particularly if the material is treated in a sharp fashion and is not diffused over the grid as in volume averaging approaches. In this regard, the Ghost Fluid Method (GFM) has been widely used to capture the interface conditions. However, the GFM results in severe over and under heating errors and hence must be supplemented with additional corrective measures to refrain these errors from growing. The Characteristic Based Matching (CBM) was found to be effective in significantly reducing these errors. However, the implementation of the CBM approach is more involved than the straightforward GFM approach. In this work, a simple, easy and an efficient approach is presented which is found to converge at least as fast as the CBM approach. This approach involves the use of a local Riemann solver to correct the over/under heating errors. In the case of the fluid fluid interfaces, it is relatively straightforward to setup the Riemann problem but it is not trivial for the case of solid fluid interfaces. Here, the method has been shown and tested for a wide range of problems and has been found to work effectively.
Martin Thuo
Chemistry pot-in-pot reactions: site-isolation of organometallic catalysts for versatile cascade reactions and recycling
We succesfully occluded Grubbs catalysts in polymeric membranes leading to new reactivities and Selectivities.1,2 Herein we report the macro-fabrication and application of PDMS in pot-in-pot reactions.3,4 These newly discovered tool exploits selective diffusion of small molecules across a polymeric barrier to site-isolate organometallic catalysts, solvents, and/or reagents. This efficient site-isolation allowed us to perform new cascade reactions based on well understood reactions. Our findings will be discussed.
Pradeep Mandapaka
Civil & Environmental Engineering Space-time Scaling Analysis of Rainfall Events over Central United States and Florida
Comprehensive statistical characterization of several rainfall events over central United States and Florida is carried out in terms of first order statistics (e.g., mean accumulation and intermittency), spatial dependence statistics (e.g., correlation and power spectrum), and time correlations for different spatial aggregations. In addition, the study also investigated the rainfall events for the presence of multiscale statistical structure. Rainfall fluctuations in all the events displayed multiscaling behavior, which is summarized in terms of generalized structure function and moment scaling function. For the analysis, customized high resolution radar-rainfall products based on Next Generation Weather Radar (NEXRAD) network Level II data are obtained from the Hydro-NEXRAD system at the University of Iowa. Only the summer season events from the year 2000 to 2007 are considered to avoid any seasonal effects on the results.
Brian Barry
Chemistry Low-temperature Solvothermal Synthesis of Single-phase, Phosphorus-rich Transition-metal Phosphides
Transition-metal phosphides (TMPs) are well-known for their capability to form a wide variety of binary phases with varying stoichiometries. These phases can range from metal-rich to phosphorus-rich, and in turn exhibit an extensive assortment of properties including semiconducting, ferromagnetic, catalytic and optoelectronic to name a few. When nanoparticles are produced, the properties can differ from their bulk counterparts and may even be tuned. Because of this, the synthesis of nanoparticles has recently received a great deal of attention. Of the reported TMP phases, only a small fraction have been reported as nanoparticles, nearly all of which are metal-rich or monophosphides. The synthesis and characterization of phosphorus-rich TMP nanoparticles is virtually unexplored. We have found that the reaction of some anhydrous metal chlorides with white phosphorus (P4) in superheated toluene can produce nano and sub-micron particles of CoP3, NiP2 and CuP2.
The paper discusses a probabilistic solution to the fault-containment problem for a class of distributed systems. The solution transforms a self-stabilizing distributed system into a system that is both weakly fault-containing and self-stabilizing, with a randomized scheduler. Weak fault-containment means that from all single failures, the expected recovery time of the transformed version is dependent only on the degree of the nodes. Furthermore, observable changes are confined to only the immediate neighbors of the faulty processes with a high probability. Why should anyone care about an algorithm that allows the neighbors to be contaminated? The answer lies in the small fault-gap. Most solutions to fault-containment that we know of achieve a fault-gap of O(n) or worse. Our solution guarantees that the fault-gap depends only on the degree of the nodes, and is independent of the size of the network.
Si-Chi Chin
Informatics (Information Science) Plural Analysis in Two Linguistic Copora
Text normalization, meaning the conversion of terms to the base forms, is widely applied as a method to cluster words with the same concepts. The conversion of plural forms into singular ones, namely depluralization, is used by many implementations of information retrieval as a minimal text normalization technique. Although some research indicates that such a normalization process would impede word sense disambiguation and text classification, truly thorough research on the distribution of plurals and singulars in a large corpus is absent from current literature. This paper analyzes the usage of plurals and singulars in a Reuters Newswire corpus and Supreme Court Opinion corpus and a to evaluate the potential information loss from depluralization process.
Arda Konik
Biomedical Engineering Development of an Automated [15O]Water Production/Injection System for Positron Emission Tomography
An automated radioactive water ([15O]H2O) production and injection system, controlled by a PC running LabView software, has been developed to minimize the radiation exposure to the PET personnel and to enable single subject studies in which a high number of injections are administered in rapid succession. The system is capable of producing [15O]H2O at desired radioactivity levels (up to 150-180 mCi at 10 µA deuteron beam current) in a very small volume (<2ml) and delivering it at various pump speeds. The system hardware mainly consists of computer controlled solenoid valves and syringe pump, a diffusion unit and a PMT (photomultiplier tube) detector mounted on it.
Zuhong Zhang
Biostatistics A Bivariate Survival Model for the Association of GB Virus Type C Persistence Time and HIV Survival Time
The motivation for this study is through the investigation of the association between infection with a virus, GBV-C virus Type C (GBV-C), and survival time of HIV-infected people. Several recent studies suggest that persistent co-infection of GBV-C is associated with prolonged HIV survival, while several other studies did not find any association and some even concluded that there is a negative association. All previous studies compared the Kaplan-Meier survival probabilities for HIV-infected subjects with or without GBV-C infection. However, the two-sample comparison method used in these studies does not adjust for GBV-C persistence time, which may vary from subject to subject due to its self-clearance nature. The model developed in this paper treats GBV-C diagnostic tests over time as providing interval censored data on the time of GBV-C persistence. An inference procedure is developed to study the association of HIV survival time and GBV-C persistence time in the existence of hybrid censoring. We established asymptotic properties of the association estimate and applied the method to a Muiti-center AIDS Cohort Study.
Yan Feng
Physics and Astronomy Accurate particle position measurement from images
The moment method is an image analysis technique for subpixel estimation of particle positions. The total error in the calculated particle position includes effects of pixel locking and random noise in each pixel. Pixel locking, also known as peak locking, is an artifact where calculated particle positions are concentrated at certain locations relative to pixel edges. We report simulations to gain an understanding of the sources of error and their dependence on parameters the experimenter can control. We suggest an algorithm, and we find optimal parameters an experimenter can use to minimize total error and pixel locking. For a dusty plasma experiment, we find that a subpixel accuracy of 0.017 pixel or better can be attained. These results are also useful for improving particle position measurement and particle tracking velocimetry using video microscopy in fields including colloids, biology, and fluid mechanics.
Nicole Kallemeyn
Biomedical Engineering Toward Patient-Specific Finite Element Mesh Development and Validation of the Spine
Finite element (FE) analysis is a useful tool to study spine biomechanics; an accurate model, once validated, has the potential to yield clinically relevant results which will aid in the treatment of spine pathologies. The anatomy of the spine is very complex, making it difficult and time consuming to create a geometrically accurate FE mesh. Most current models are based on simplified or generalized geometry, and are not created on a patient-specific basis, which is of importance. Using an extension of existing multi-block techniques, we propose novel methods for patient-specific FE mesh development of the spine, which enable the user to create spine FE models much more quickly than traditional methods. We have created a CT based cervical spine functional spinal unit (C45) FE model that to date has been validated by comparison to experimental data in the literature. Future work will include specimen-specific validation of our models.
Piotr Domaszczynski
Civil and Environmental Engineering Hydro-NEXRAD: System Overview
Hydro-NEXRAD is a prototype system that allows hydrologists to obtain user-specified rainfall data for their research. These data are based on observations collected by the national network of WSR-88D radars, known as NEXRAD. Users interact with Hydro-NEXRAD through a web-based interface that has map-based components for spatial navigation, calendar- and time series plot components for temporal navigation and a menu-based component for selection of processing options. Through the interface, users browse the Hydro-NEXRAD metadata and select data of interest. As the system is approaching the point of being fully operational, the authors and a group of test users have evaluated several aspects of the system. Metadata remains very important for the system functionality. Radar-based, basin-based and point (for selected set of rain gauge locations) metadata serve multiple purposes: 1) enable users to efficiently search for subsets of data (SQL query, visual inspection), 2) provide information on quality of the collected data archive (missing or corrupt data), 3) and have a scientific value (basin-based metadata has a potential to be used as a precipitation input to hydrologic models). Author provides an overview of the Hydro-NEXRAD system.
Gary Pierce
Biomedical Engineering Isokinetic 3D Shoulder Strength Assessment: Preliminary Internal External Rotation Strength Surfaces
This work experimentally measures maximal shoulder joint torque at several isometric positions and angular velocities to develop 3D surfaces of peak torque as a function of position and velocity. We will specifically focus on internal and external rotation strength, as this is important for rotator cuff stability at the shoulder joint. This testing is being conducted using both isometric (static) and isokinetic (dynamic) contractions. The testing is currently ongoing; preliminary results will be presented. We hypothesize that the resulting 3D internal rotation strength surface will differ from that of external rotation: peak torques will occur at different angles, the shapes of the strength surfaces will vary due to the differing muscle biomechanics of the shoulder rotators; and internal rotation torque will be greater than external rotation torque, based on previous research. Currently, shoulder strength data is only available in 2D as torque-velocity or torque-position curves, without considering their potential interactions. This study will provide the first 3D surface of torque, as a function of joint position and velocity, for the shoulder joint.
Debra Ely
Biomedical Engineering Study of Microvolt T-wave Alternans as a Predictor for Sudden Cardiac Death
Heart disease is the number one killer among men and women in the United States. Microvolt T-wave alternans (MTWA) has been identified as a precursor to ventricular arrhythmias (VT) and ventricular fibrillation, which are predecessors of sudden cardiac death (SCD). MTWA is a noninvasive and inexpensive risk assessment used to identify a patient at risk for SCD. We looked at the effect of drugs, VT, and ligation in regards to alternans ratio to help identify mechanisms related to MTWA. Data showed that an hour after occlusion infarct size stabilized as a result of ligation and endothelin consistently increased alternans ratio when administered.
Srinivas Tadepalli
Biomedical Engineering Cervical Laminoplasty Construct Stability: A Finite Element Study
Cervical laminoplasty is a widely used surgical technique to treat spinal cord compression caused by cervical spondylotic myelopathy, ossification of the posterior longitudinal ligament, congenital spinal stenosis, or a combination thereof. Cervical laminoplasty permits extensive canal decompression while maintaining cervical alignment and stability. The single-hinge laminoplasty technique consists of opening the lamina from either the left or right side, with the contra lateral side acting as a hinge. It still remains unclear as to what extent the lamina must be opened so that the spinal canal is widened for good results. The finite element (FE) method is an ideal tool to determine the biomechanical response of the instrumented cervical spine. A FE model of the fifth cervical(C5) vertebra was created to study the biomechanical stability of a posterior single-hinge cervical laminoplasty via a plating system. Our novel modeling techniques provide a means to capture the complex irregular geometry of cervical spine on a patient-specific basis that will ultimately allow the surgeons to plan patient-specific care. Thus they can better evaluate alternate surgical procedures and treatments thereby contributing to improved surgical outcome.
Poonam Kaushik
Chemistry A sacrificial approach towards the synthesis of reactive targets via template-directed solid-state synthesis
We are involved in developing a synthetic method that utilizes templates to direct the reactivity of molecules within hydrogen-bonded assemblies in solids. We have shown that co-crystallization of resorcinol with trans-1,2-bis(4-pyridyl)ethylene produces hydrogen-bonded assemblies that undergo photochemical [2+2] reaction to produce the corresponding cyclobutane product with a controlled stereochemistry, namely rctt-1,2,3,4-tetrakis(4-pyridyl)cyclobutane. This approach has been used to construct biologically-relevant ladderanes as well as cyclophanes. Owing to the architecture of molecular assemblies in our studies, the molecules constructed via template-directed solid-state syntheses are pyridine derivatives. Although templates have been shown to provide a general means to control bimolecular reactivity in the solid state, the stability of pyridines toward chemical modification can limit the products in downstream synthetic applications. To explore a possible alternative to pyridines, we have investigated benzothiazoles as hydrogen-bond acceptor groups. The benzothiazole ring has been used as a formyl group equivalent in solution-phase organic synthesis. This presentation will illustrate our results in using the benzothiazole functionality as a functional group alternative to pyridines in template-controlled solid-state synthesis to obtain easily-modifiable targets.
Soojeong Kim
Applied Mathematics 4-string tangle analysis of DNA-protein complexes based on difference topology
Difference topology is a methodology to derive the number of DNA crossings trapped in an unknown protein complex. By this method, Pathania et al revealed the topological structure within the Mu protein complex which consisted of three DNA segments containing five nodes. In their experiments, they used a site-specific recombinase which is known as Cre. Cre mediates DNA exchange by rearranging target sites of the DNA segments. The initial DNA conformation is unknotted. After Cre recombination, the products are knots or catenanes. Recently, Darcy et al. proved that the five-noded conformation is the only biologically reasonable structure of the Mu protein DNA complex. We address the possibility of protein complexes that binds four DNA segments. By the property of Cre, we can assume that after Cre recombination, the topology of a DNA-protein complex would be a knot or catenane. The latest results of the mathematical tangle model for this case will be discussed.
Xuefeng Zhao
Mechanical Engineering Pointwise Identification of Elastic Properties in Nonlinear Hyperelastic Membranes
We present an innovative method for characterizing the distributive elastic properties in nonlinear membranes. The method hinges on an inverse elastostatic approach of stress analysis that can compute the wall stress in a deformed convex membrane structure using assumed elastic models without knowing the realistic material parameters. This method of stress analysis, in conjunction with a suitable method for strain data acquisition, lays a foundation for pointwise identification of stress-strain properties in membranes of general convex shape. Based on this inverse method, an experimental methodology is developed to characterize the pointwise property of nonlinear elastic membrane. The results show that the method can effectively delineate the distributive elastic properties in membranes.
David Rotsch
Chemistry Synthesis of Polynuclear Eu(II) Complexes: MRI Contrast Agents for the Detection of Reactive Oxygen Species
There has been an imperative need in the field of medicine for a noninvasive method for the detection and study of reactive oxygen species (ROS). Some ROS have been implicated in aging, cancers, and reperfusion injury in acute myocardial infraction. The incorporation of redox active contrast agents (CA) and magnetic resonance imaging (MRI) may prove to be an effective and noninvasive way to measure ROS levels. Current CA do not possess the ability to change their oxidation states. It is the purpose of this research to synthesize stable redox active CA. Europium, a lanthanide has similar MRI detectability as gadolinium (the active site in current CA) and is redox active. Europium has two accessible oxidization states and each state demonstrates a different MRI signal strengths and thus present potential ROS MRI CA. Herein we discuss the synthesis, findings, hopes and setbacks of high nuclearity clusters as MRI CA.
Dongsu Kim
Civil and Environmental Engineering Three Dimensional Representation of River Hydrodynamics using Acoustic Doppler Current Profiler
The objective of this paper is to highlight the applications of recent Acoustic Doppler Current Profiler (ADCP) for measuring three-dimensional river channel variables. In order to conveniently process, store, and visualize this abundant ADCP information a customized visualization and processing software, labeled AdcpXP, was developed. The present paper also describes initial efforts for developing a multi-dimensional geospatial and temporal data model for representation of river geometry and flow characteristics. The developed Geo-data model, labeled Arc River, is a customized data repository to assimilate various types of three-dimensional hydrodynamic datasets acquired with ADCPs into geodatabase schema on the top of any available relational database management system (RDBMS).
Natalie Ulrich
Chemistry Synthesis of D-Ring Analogues of the Schweinfurthins
The schweinfurthins comprise a small family of natural products isolated from the African tree Macaranga schweinfurthii at the National Cancer Institute (NCI) in the late 1990s. Most of these compounds exhibit both potent and differential cytotoxicity in the NCI s 60-cell line assay. Of particular interest is the schweinfurthin's unique pattern of cytoxicity toward specific cell lines, a pattern which shows no significant correlation to the profile of any known anticancer agent. This pattern may be indicative of a novel mechanism, a novel molecular target, or both. For the purpose of elucidating the mechanism of action and molecular target of the schweinfurthins, a number of new D-ring analogues have been synthesized. Progress on the synthesis of several schweinfurthin F analogues featuring D-ring modifications, work aimed at biotinylated derivatives for affinity bioassays, and preliminary data from in vitro testing will be presented.
Karna Barquist
Chemistry Amino functionalization of silicious zeolite materials
Zeolite materials were functionalized with amine groups and used for adsorption of chromate anions out of aqueous solutions for use in possible environmental applications.
Robert Williams
Civil and Environmental Engineering Multi-Scale Modeling of Non-Linear Anisotropic Elastoplastic Behaviour in Woven Fabrics
In this research on inelastic mechanics of textile fabrics we describe results obtained with both computational and experimental methods. The mechanical behavior of woven fabrics is of interest because in stiff, bulky protective clothing systems they can adversely impact human performance. With good micro-mechanically based models for such fabrics, their impact on human performance can be quantified. A hierarchical multi-scale modeling approach is employed to investigate the mechanics of protective fabrics and develop suitable constitutive models. Beginning at the length scale of individual fiber diameters (tens to hundreds of microns), unit-cell methods are used to study frictional stick-slip interactions between fibers which give rise and contribute to the nonlinear and dissipative characteristics of textile fabrics. After assimilating these micro-scale interactions into mesoscale material models, unit-cell methods are used again on the length scale of millimeters to study the mechanical interactions between yarns under biaxial stretching and shearing. Finally, an anisotropic elastoplasticity model for fabrics that integrates key structure-dependent behaviors from the micro- and mesoscales is implemented, and its resulting behaviors are compared with experimental results.
Anton Petushkov
Chemistry Zeta Potential Studies of Functionalized Silicalite-1 Nanocrystals
Zeolites are inorganic crystalline materials, whose framework consists of silicon, oxygen and aluminum atoms. The zeolite framework is characterized by a system of interconnected micropores and cavities typically ranging from 3 to 15 Angstroms. Zeolite surface consists of silanol groups that can be functionalized with a wide variety of organics to tailor the surface properties of nanoparticles towards specific applications.One of the challenges in using nanoparticles is that their colloidal solutions are often metastable and the particles can form aggregates. The stability of zeolite nanoparticles can be improved by charging their surface, which can be monitored by measuring the zeta potential.The objective of this work was synthesis and functionalization of silicalite-1 nanoparticles with organic groups, namely amine and carboxy. Zeta potential studies were then carried out to provide more insight in the stability of nanozised crystals in the range of pH typically used in drug delivery.
Peter Likarish
Computer Science B-APT: a Bayesian Anti-Phishing Toolbar
Identity theft is one of the fastest growing crimes in the nation and phishing is a primary vector facilitating identity theft. In this paper we present a Bayesian Anti-Phishing Toolbar (B-APT) which identifies phishing websites. Bayesian filters have performed superbly in content-based spam filtering and we adapted a Bayesian filter to detect phishing attacks in the web browser. Our results show that B-APT effectively detects phishing attacks (in our experiment B-APT detected 100% of phishing sites), has very few false positives and has low resources requirements.
Natalia Alexeeva
Chemistry Near-infrared microspectroscopic analysis of rat skin tissue heterogeneity
Sixteen rat skin tissue samples are characterized semi-quantitatively. Despite the complex composition of skin, its combination spectra (4000-5000 cm^-1) can be modeled with six components: water, collagen type-I, keratin, fat, a constant and a slope terms. Linear fit is performed with absorbance spectra of these pure components to yield sets of six fractional coefficients for each measured location. A two-dimensional map of each component can be created to represent its relative distribution in each particular sample. Reference point in the center of the sample is periodically measured to track dehydration, change in tissue thickness and temperature with time. Variations in chemical composition due to different image location are at least ten times greater than the temporal changes at the reference location. One-way ANOVA suggests that variation in tissue composition between animals is larger than between locations within a single sample.
