Physical and Mathematical Sciences
Links provided will take viewers to UMaine’s Kaltura video gallery. Presentations available for viewing by UMaine users only are noted (*).
0101. Luminosity Functions of Galaxy Clusters in the Aquarius and Microscopium Superclusters
Graduate Presentation.
Author(s): Logan Hess Mentor(s): David Batuski
Abstract: We examine physical properties of several galaxy clusters located within the Aquarius (AQR) and Microscopium (MSC) super-cluster complexes via their luminosity functions, fitted using Schechter functions. Images of clusters were acquired via the DECam imager of the 4-meter diameter telescope at CITO (Cerro Tololo Inter-American Observatory). We observe six clusters in our sample; four from the Aquarius supercluster (AQR 2541, 2546, 2554, 2555) and two from the Microscopium supercluster (MSC 3695, 3696). Magnitude information and density of member galaxies in each cluster was extracted via the SExtractor software utility. We correct for background by sampling and subtracting under-dense regions of our images. We construct luminosity functions (observations of magnitude vs cluster density) for each cluster in our sample. Schechter functions are fitted to luminosity functions via chi-squared minimization. Fitted Schechter parameters α (characteristic slope) and m* (characteristic magnitude) of samples are compared to parameters derived from other publications of clusters of different environments and redshift.
0102. Exchange Functionals and Basis Sets for Density Functional Theory Study of ZnO Nano clusters in Photocatalytic Reactions
Graduate Presentation.
Author(s): Duwage Perera Mentor(s): Jayendran C. Rasaiah
Abstract: Nano scaled ZnO clusters used as catalysts can push the limits of many industrial applications of chemical reactions in the manufacture of paints, cosmetics, and pharmaceuticals. The clusters act as semiconductors with a wide band gap 3.4 eV at 300 K, and they are prospective photocatalysts in many applications including H2 production in water splitting reactions. In this project, we studied the structural (geometry) and electronic properties (vertical detachment energy and electron affinity) of ZnO monomers and dimers to form several different model ZnO clusters using density functional theory (DFT) with many different exchange functionals and 29 basis sets. We compared the singlet-triplet energy gaps of small ZnO clusters to find the most efficient ZnO cluster and the best theoretical method to investigate their photocatalytic water splitting activity. Our results show that B3LYP/DGDZVP2 level of exchange functional/basis set theory is the most efficient and fastest method to analyze the catalytic activities of ZnO nanoclusters. Comparison of the singlet-triplet energy gap shows that the trimer (ZnO)3 (with 58.66 k cal/mol energy gap approximately equal to the energy of a visible photon at~555 nm) and a HOMO-LUMO gap of 4.4 eV is the best amongst the clusters of different size. We used the Gaussian16 software package (Installed on the UMaine Computer Cluster) for all the calculations.
0103. Characterization and modification of cellulose nanofibrils for green composites
Graduate Presentation
Author(s): Peter Kelly, William Gramlich, Douglas Gardner Mentor(s): William Gramlich
Abstract: Cellulose nanofibrils (CNFs) derived from woody biomass are promising candidates for use as composite reinforcements due to their high strength and inherent biodegradability. Their unique morphology allows them to form networks throughout the composite, assisting load transfer and reinforcing the underlying polymer matrix. The characterization of this hierarchical morphology has proven challenging, due to the orders of magnitude differences in the sizes of fibrils present. A method for the characterization of CNF morphology at multiple length scales has been explored and applied to the production of nanofibrils derived from northern bleached softwood kraft pulp and old corrugated container material. Atomic force, scanning electron, and optical microscopy have been used to characterize fibril dimensions throughout the refinement process to better understand the changes occurring. Additionally, polymeric modifications to the CNF surface, achieved via a grafting-through approach, have been carried out using a library of (meth)acrylate and (meth)acrylamide monomers. These modifications and the effects they have on the morphology of the CNF substrate have been studied and will be discussed.
0104. Application of Machine Learning Techniques to Classify and Identify Galaxy Merger Events in the CANDELS Fields
Graduate Presentation
Author(s): Alex Koch Mentor(s): David Batuski
Abstract: A random forest classifier written in the Python coding language was applied to images of a subset of galaxies from the Cosmic Assembly Near-infrared Extragalactic Legacy Survey (CANDELS) to classify merger and non-merger events. 226 merging and 226 non-merging galaxies were selected from four CANDELS fields, totaling a combined 452 galaxies for training and validation. The classifier was trained on a set of parameters measured for each galaxy, including mass, star formation rate, galactic half-light radius, as well as Concentration and Asymmetry measurements, for 339 galaxies using five-fold cross-validation. The remaining 113 galaxies were used for testing and validation. The classifier performed with a mean accuracy of 92.31% and a precision of 93.32.
Additionally, preliminary work has been completed training a Computer Vision convolutional neural network to analyze and classify images of merger and non-merger events in the same fields. Due to the small number of merger events, data augmentation has been done to increase the dataset significantly to boost performance. Currently, the classifier performs with an accuracy of 85.66% and a precision of 85.23, although optimization of the classifier is ongoing.
0105. Surface Modification of Cellulose Nanofibrils Using Water Soluable Monomers and Subsequent Surface Polymer Characterization
Undergraduate Presentation
Author(s): Peng Cheng, Peter Kelly, William Gramlich Mentor(s): William Gramlich
Abstract: While efforts have been taken to introduce more biodegradable plastics into the commercial space, many modern biodegradable plastics lack the rigidity necessary to substitute the nonbiodegradable plastics in use today. Cellulose, a naturally abundant reinforcing agent, cannot normally be effectively integrated into plastics due to poor interfacial interactions. Our research shows that cellulose nanofibrils (CNF) can undergo surface modifications by attaching various polymers to CNF, which may improve interfacial interactions between CNF and plastic materials. To further study these interfacial interactions, a method to detach the polymers on modified CNF must be investigated in order to streamline surface polymer analysis. This project seeks to synthesize fresh polymer-functionalized CNF using various water soluble monomers, then develop and enact a surface polymer characterization method to further research polymer-functionalized cellulose. A two-step grafting-through copolymerization scheme is used to synthesize polymer-functionalized CNF. Experiments show that poly(n-isopropyl acrylamide), polyacrylamide, polymethacrylamide, poly(acrylic acid), poly(methacrylic acid), and poly(2-hydroxyethyl methacrylate) can be successfully grafted onto CNF. Methods to separate grafted copolymers from polymer-functionalized CNF under basic and acidic conditions are under investigation.
0106. Analyzing Nonsmooth Sensitivities in Graph Centrality Problems
Undergraduate Presentation.
Author(s): Jonathan Donnelly Mentor(s): Peter Stechlinski
Abstract: Agent based ratings are an incredible mechanism to evaluate the importance of a given agent in many network structures. Peer-to-peer (P2P) networks, which distribute the burden of software sharing across multiple computers, have consistently risen in prominence and rely on such schemes to provide security. With richly interconnected, but decentralized structures, P2P networks often operate based on the trustworthiness of each agent in the network in a democratized way. Typically this is accomplished using algorithms based on matrix theory, such as the EigenTrust algorithm which is built from Google’s PageRank search engine algorithm. In this scheme, agents assign individual trust scores to one another, which then propagate through the network and eventually generate a global trust score for each user; these individual trust scores may be assigned as negative (distrust) or positive (trust). Eventually, a consensus is reached between agents which corresponds to the principal eigenvector of a matrix representing the network. This consensus value, however, may be vulnerable to manipulation; because network members provide the baseline ratings, a group of malicious actors could carefully choose the scores they assign to reduce the score of a target actor. We are interested in analyzing the sensitivity of these trust scores to changes in the network, to determine the viability of such attacks. We apply recently developed theory called lexicographic differentiation to compute parametric sensitivities, which can be used in, for example, optimizing the network against malicious actors who aim to engineer the network.
0107. The Impact of Natural Rubber as a Toughening Agent on the Strength, Degradability, and Toxicity of an Algae-Based Bioplastic
Undergraduate Presentation
Author(s): Megan Driscoll Mentor(s): William Gramlich
Abstract: In recent years, the negative impact of petroleum based plastics on human health and the environment has become increasingly clear. Due to mismanaged waste systems, billions of tons of plastic waste has infiltrated both terrestrial and marine ecosystems across the globe. These plastics leech harmful substances into the environment, cause animal ingestion and entanglement, and break down into microplastics that are then biomagnified up the food chain, resulting in human consumption. However, the use of plastics have been revolutionary in science, medicine, and daily life, even benefiting the environment in many ways. Therefore, rather than ridding the world of plastics altogether, a promising solution has come in the form of bioplastics, or plastics made from renewable materials. Current bioplastics, such as PLA, are limited in application due to their brittle nature. Additives, like glycerol plasticizers, are often added to offset brittleness, but sacrifice strength in return. In place of glycerol, Natural Rubber, an elastomer, can be used as a toughening agent to benefit both the durability and strength of the plastic. This project looks at the impact of Natural Rubber on the strength, degradability, and toxicity of an agar-based bioplastic. Thiol-norbornene modifications were used to couple the agar and the natural rubber. Six film formulas were made to understand the impact of each step of the modification process. The viability of the films was determined using tensile strength analysis, a four-week degradation study, and a bioassay using Artemia sp. as a model organism for short-term toxicity.
0108. Discerning the Distribution of Dark Matter via Weak Lensing in Extremely Dense Supercluster Environments
Graduate Presentation.
Author(s): Sarah Rice Mentor(s): David Batuski
Abstract: It is well known that, on large scales, the Universe is composed of a vast network of galaxy filaments and voids. Since dark matter interacts gravitationally, it stands to reason that it should be distributed in a similar manner as luminous matter. In this project, we analyze the shapes and orientations of faint (very distant) galaxies to construct weak lensing maps of two extremely overdense supercluster regions using data gathered from our 2015 observation program. The images analyzed to create these maps were taken using the 500 megapixel Dark Energy Camera on the 4-meter telescope at Cerro Tololo Inter-American Observatory in Chile.
We then compare each map to a model of the lensing expected from the galaxy clusters within each of our two supercluster images. The results thus far show statistically significant agreement for the orientations of the lensing effects between the data and the model. This suggests a lack of possible massive dark matter filaments connecting the galaxy clusters in each supercluster region.
0109. Cyanide Detection in Water Using Infrared Spectroscopy
Undergraduate Presentation.
Author(s): Kristen Dennen, Nayeem Ibnul Mentor(s): Carl Tripp
Abstract: Cyanide is an extremely toxic ion which often contaminates drinking water from industries such as mining, steel and iron manufacturing, and wastewater treatment. If ingested or inhaled, cyanide prevents cells from using oxygen, leading to cell death. Health effects can range from dizziness to loss of consciousness and respiratory failure. Long term exposure can lead to heart and brain damage. The Environmental Protection Agency (EPA) sets the limit for cyanide in drinking water at 0.2µg/mL, or 0.2ppm. The goal of this project is to develop a method to detect cyanide at concentrations below the 0.2ppm limit, while also allowing samples to be tested in the field, quickly and cost efficiently. In this method cyanide is precipitated from a water sample using silver nitrate, and a given volume of the sample is passed through a membrane. This membrane concentrates the solid cyanide from the effluent. An infrared spectrum is then recorded directly off the membrane. A band on the spectrum corresponding to cyanide measures how much infrared radiation was absorbed by the solid cyanide. Using this absorbance value, Beer’s Law calculations can be performed to determine the concentration of cyanide that is present in the water sample.
0110. Colloidal interactions to generate latex coated pigment for paper coating applications.
Graduate Presentation.
Author(s): Pradnya RaoMentor(s): Carl Tripp, Doug Bousfield
Abstract: Latex binders are the costliest components in the coating. A cost-effective paper production approach made industry efficient utilization of binder without compromising with the coating strength and paper quality. In this proposed work, we are implementing an innovative industrially scalable approach of layer by layer techniques using electrostatic interactions to form a pigment binder composite in the solution phase as a conceivable method for lesser and effective use of the binders. Coating formulations are generated using cationic precipitated calcium carbonate (PCC) pigments as a core and it is mixed with anionic styrene-butadiene (SB) latex binders to form monolayer adsorption of binder particles on the pigment. The migration of latex binder in coatings generated on various substrates under various drying conditions was measured using Raman spectroscopy and compared with reference coatings consisted of anionic pigment and anionic binder. For most cases, the new system showed reduced latex binder migration than those obtained with the reference coating. The coated papers were also measured for strength, opacity, gloss, water drainage rate, and porosity. Little difference is seen in the picking strength of the coating and gloss compared to coatings prepared with standard formulations. Water drainage rate, opacity, and porosity were higher for latex coated pigment coatings than the reference standard coating; this increased porosity is likely due to the strong electrostatic interactions present in the coatings.
111. UMCS Lab Help Center
Undergraduate Presentation.
Author(s): Jeremy Hutchinson, Chad Berry, Nate Gagnon, Cooper Dahlberg, Isaac Sparks-Willey Mentor(s): Penny Rheingans
Abstract: In order to assist with the operation of the Boardman Hall Computer Science lab, we at Diamond Tech are in the process of developing a web application that will allow students to request help, both in person and remote through zoom or discord, from lab staff in a structured and organized way. The Computer Science lab, located in Boardman room 138, is a common use lab available to all computer science students. The lab is often utilized for working on assignments or personal projects as a group, and computer science TAs and professors are available throughout the week to provide feedback and support when needed. Our application, the UMCS Lab Help Center, will assist lab staff by keeping track of requests for help, and data about requests will be stored and analyzed to provide insight into when and with what computer science students require help. The Help Center will help lab staff assist students as well as inform computer science instructors when a large number of students from their course are seeking help. We are developing the Help Center as part of the computer science capstone program. We are using an agile development process that resembles the V-model of software development, meaning development takes place over two week periods followed by a meeting to discuss progress. Documentation and planning began with an overview, and as development progressed the scope focused down to specific architecture planning. After planning was complete, we moved to implementation, starting with a very focused scope and then integrating each component with the others. The architecture of the Help Center is being developed following the model-view-controller model which organizes components of the application into three separate categories that only interact when necessary. Components of the system that involve user interaction are separated from internal representation of concepts, which are again separate from the final view shown to the user.
0113. Synthesizing hydrogels from norbornene-functionalized hydroxypropyl cellulose.
Undergraduate Presentation.
Author(s): Lauren Andrews Mentor(s): William Gramlich
Abstract: Hydrogels are three-dimensional networks of crosslinked polymers that display a high water affinity, allowing for absorption of large amounts of water into the polymer network. While already having numerous applications in industries such as medicine and cosmetics, common methods for synthesizing hydrogels often require non-sustainable materials and may have toxic byproducts formed during gelation. Using cellulose, which is the most abundant biopolymer on Earth, to create hydrogels solves these problems by using a sustainable, biologically inert starting polymer that produces no harmful gelation or degradation byproducts. The goal of this research, under the guidance of Dr. William Gramlich in the Department of Chemistry, is to synthesize hydrogels from norbornene-functionalized hydroxypropyl cellulose (cHPC) by altering the concentration of the crosslinking agent and then conducting rheological studies on the gels. The results of mechanical testing will determine if the mechanical properties of cHPC hydrogels can be precisely tuned by changing these initial reaction conditions.