TTU Published Journals @ Volpe Library
Not a member yet
971 research outputs found
Sort by
*WINNER* A study on the Spectrophotometric Analysis of Hg(II) using Dithizone under Conditions Pertinent to Hg(II) Reduction in Aquatic Systems
No full textThe reduction of mercuric mercury (Hg(II)) in aquatic systems contributes to the transformation, transportation, and fate of mercury in the environment. Solar radiation has been identified as the driving force of Hg(II) reduction to volatile dissolved gaseous mercury (DGM) in aquatic bodies and is hypothesized to be linked to processes mediated by dissolved organic carbon (DOC). Superoxide has also been hypothesized to mediate Hg(II) photoreduction in aquatic systems but the proposed mechanism has recently been rejected in literature. We investigated the applicability of the dithizone method for spectrophotometric analysis of Hg(II) under different environmental conditions pertinent to the study on aquatic Hg(II) reduction, particularly superoxide-mediated Hg(II) reduction. We studied the effects of organic acids (e.g., citrate, cysteine), pH, and reagents used to generate superoxide (e.g., xanthine, xanthine oxidase). Our study showed that some organic acids lowered the sensitivity of the method but the calibration curves in all scenarios retained good linearity. We concluded that the mercury-dithizone method was valid for the study of superoxide mediated Hg(II) reduction. Our preliminary study of this reduction was performed in the absence of light and under controlled conditions (e.g., pH 7.2, 25°C, and 5 µM initial Hg(II) concentration). We observed a decrease of the absorbance at 496 nm (i.e. reduction of Hg(II)) with various rates of superoxide production and various concentrations of organic acids. This preliminary research suggests that superoxide could be an important intermediate for Hg(II) reduction even in the absence of sunlight
*WINNER* Creating Color Flame Candles as an Alternative to the Rainbow Flame Test
No full textThe rainbow flame test is a visually appealing chemical demonstration that showcases atomic emission spectra, but it can be very dangerous and has caused injuries due to accidents. Recently updates have been made to increase the safety of the demonstration, but it can still prove to be inaccessible to groups without access to proper safety training, certain scientific equipment, and supplies. The purpose of this research is to create easy to make candles that produce colored flames which can be used over long periods of time in a safe manner in classroom and educational settings. Cotton and wood candle wicks are soaked in salt solutions containing different metal cations that are known to produce colored flames and then sealed with wax prior to making the candles. Some wicks are sealed with wax and others are not to determine the effectiveness of sealing the salts into the wicks prior to making the candles. A portion of the sealed candles are coated with a wax containing the salt to allow the salts to soak into the wick as they burn. Other wicks are prepared by spraying the solution on the wicks and allowing them to dry between applications to build up the amount of the salt on the wicks. If successful, this work would allow a new way for the rainbow flame test to be conducted in a safe and accessible manner for a variety of audiences in scientific and non-scientific settings
Future Sales Predictions on Russian Electronics Shops
Get PDFSpeculations about future sales for a given company or store and for given products in those stores have been critical to successful trade going far back in human history. However, until much more recently this was more art than science as the availability of mass sales data was much more limited. Today, we have access to nearly boundless amounts of relevant data and incredible computational tools with which to work with it and to improve upon these speculations. Given the daily historical sales data provided by the Russian firm 1C used in the Kaggle challenge “Predict Future Sales,” we seek to improve on existing sales predictions across items and stores that may or may not be chains. This dataset provides information about many individual store locations as well as 11 fields for products and sales per shop per day from January 2013 to October 2015. To this end, we will use the statistical software R and possibly machine learning methods to generate a month’s worth of sales predictions ahead of store restocking. We will identify common patterns across stores such as sales data for the same or similar product as well as attempting to identify yearly trends in sales. As it is a Russian dataset, we will seek to overcome lingual and cultural barriers faced by international data scientists in industry as we pursue this goal
Chess Openings and Ratings
No full textThe purpose of our research is to determine what chess opening works best for a given rating, so players can give themselves an edge and win more games. In chess, a person’s skill can be quantified in a number called a “rating”. Most chess websites use this rating to match players together with similar ratings to create an even match. Based on this skill rating, we want to determine what openings work best for different ranges of ratings. This is important because some openings have a more potent effect in lower ratings, but are not as effective in higher ratings. We compare different openings and their win rate together across different buckets of ratings to find out which opening, or openings are most effective at which rating
Malware Classification Using Deep Learning in Cloud Environments
No full textCloud infrastructure is vulnerable to malware due to its exposure to external adversaries, making it a lucrative attack vector for malicious actors. A datacenter infected with malware can cause data loss and/or major disruptions to service for its users. This work analyzes and compares various deep learning and machine learning methods within the scope of malware classification. The classification is based on behavioural data using process level performance and system wide performance metrics including cpu usage, memory usage, disk usage etc. These machine learning models are designed to extract features from data gathered from live malware running on a real cloud environment. Experiments are performed on OpenStack (a hypervisor) testbed designed to simulate cloud environment scenarios. Comparative analysis is performed for different machine learning models
Dynamic Address Validation Array (DAVA): A Moving Target Defense Protocol for CAN bus
Get PDFThis paper presents Dynamic Address Validation Array (DAVA), a novel moving target defense protocol for the Controller Area Network Bus (CAN bus). DAVA's primary goal is to mitigate the common CAN bus vulnerability of an unauthorized entity misappropriating components in the vehicle through sniffing and reusing ECU IDs for replaying messages. Using a dynamically allocated array stored in the ECU that is updated and validated frequently, DAVA limits an attacker's ability to reuse ECU IDs for replay attacks. The protocol strives to be minimally invasive and lightweight for application in CAN bus while still being secure. This paper discusses the DAVA protocol, a proof of concept implementation, and initial performance measurements. This paper explains how DAVA is able to provide a robust security framework for CAN bus without the need for a large amount of storage or CAN bus standard modification
Global Survey and Distribution of Pennsylvanian and Mississippian Microbial Mounds
Get PDFMicrobial mounds, including Waulsortian and Waulsortian-like mounds, are lithified structures composed of carbonate compounds and ancient microbes that aided in the production of those compounds. They commonly developed in shallow sea environments of the Pennsylvanian (323 to 299 Ma) and Mississippian (359 to 323 Ma) era strata due to the photosynthetic tendencies of cyanobacteria and its environmental symbionts that require marine environments (i.e.: phylloid algae). The lithification of these microbes can give insight to biosignatures left by this process and can therefore inform our understanding of microbial mounds on Earth and, potentially, on other planets. This study serves to build a global database of microbial mounds from these geologic eras, which span 60 million years. This database will include geographic data from scientific studies of microbial mound structures from the last 31 years (1990-2021). We sift through these studies to find the specific locality of when and where the specimens are found and then pinpoint the location on Google Earth. With these locations pinpointed, we can construct paleogeographic maps and compare the ecology, location, and sedimentary character of each mound. This is done in order to find similarities in conditions on ancient shallow marine slopes and determine the fundamental controls on mound formation. Understanding the conditions necessary for microbial mound growth is a first step toward predicting where mounds may have developed elsewhere in the Solar System
*WINNER* Flash Flooding Prediction of Cummins Falls State Park
No full textCummins Falls State Park has become a popular attraction since being named as one of America’s Secret Swimming Holes by the Travel Channel. Located in southern Jackson County, the park is visited by thousands of people each year. While a beautiful place, the falls also are subject to flash flooding like the arroyo flash floods of the western U.S. In response, the state has installed gauging and rainfall monitoring in the watershed. Our research focuses on comparing observed data with model output using the USACE’s Hydraulic Engineering Center River Analysis System (HEC-RAS). The objective is to determine how much rainfall can cause dangerous increases in stream level, and how much time elapses between a rainfall and the increased stream level. Observations from the rain gauges indicate that rainfall amounts over 0.04 inches in 5 minutes can cause unsafe rises in the stream depth. The time from rainfall at the far reach of the watershed to the Cummins Bridge Road station appears to range from 2-3 hours. The time between the stream rising at the West Fork gauge and the East Fork Gauge to reach the bridge is about 45 minutes. Further modeling of the watershed may provide more details of stream behavior and answers to questions of rainfall and timing
*WINNER* Stress Resistance As a Potential Mediator for the Effect of Self-Efficacy on Depression
No full textIndividuals with higher self-efficacy and greater stress resistance have lower depression. (Ehrenberg et al., 1991; Bergeman & Deboeck, 2014). Self-efficacy is an individual’s belief and confidence regarding their abilities to achieve a goal (Bandura, 1977). When subjects consider themselves not capable of doing a task, they are more likely to experience fear and give up (Bandura, 1982). When people see themselves capable of handling a situation, they show better persistence in the face of an obstacle (Bandura, 1997). We hypothesized that increasing people’s self-efficacy would lead to an increase in their stress resistance, which we predicted would mediate a decrease in depression. 150 students participated in the study through Qualtrics. The results were analyzed. Independent samples t-tests indicated that the manipulation increased Social Self-Efficacy by a slightly significant margin (p = 0.04) but did not significantly increase the targeted variable General Self-Efficacy (p = 0.53). The conditions also did not produce differences in Perceived Stress or Depression (p’s > .05). Therefore, we analyzed the data collapsed across condition to test if the predicted correlational associations were found. General Self-Efficacy was negatively correlated with Perceived Stress and Depression, which were positively correlated with each other (p’s < .05). Following the Barron and Kenny (1986) method, we tested for mediation using a linear regression and Sobel test; the results supported Perceived Stress as a mediator of the association between General Self-Efficacy and Depression. Although our manipulation was not successful, the correlational analysis supported further investigation into this model
Optimization of Mass Transport within Direct Formic Acid Fuel Cell Anode Catalyst via Pore Formers
No full textBatteries have become a necessity for today’s ever-increasing demand in portable power, but lengthy recharging times, degradation, and limited charge capacity hinder the batteries’ efficiency. Direct formic acid fuel cells are a sustainable alternative to batteries due to their high efficiency, instantaneous fueling times, and 24/7 operating time capabilities. However, mass transport limitations due to two-phase flow (gaseous carbon dioxide product and liquid formic acid reactant) plague the fuel cell’s efficiency due to the anode catalyst layer’s small pore size (~20 nm). This two-phase flow must be optimized to reach peak cell performance. This research aims to optimize the two-phase flow by incorporating a magnesium oxide pore-former at varying wt% (0-30 wt%), increasing the pore size from ~20 nm to ~50 nm, and creating a porous templated anode catalyst layer. This porous anode catalyst layer will optimize the two-phase flow of the reactant and product while maintaining the proton conduction, mass transport, and electron conduction. These porous templated anode catalyst layers have showed increased electrochemical surface areas and improved cell performances compared to non-templated anode catalyst layers