FACULTY OF PHYSICAL SCIENCES

This study investigated the effectiveness of Aloe vera gel coatings in extending the shelf life and preserving the quality of cucumber, banana, and pepper during storage. The functional group of the experimental samples were carried out using Fourier Transform Infrared Spectroscopic (FTIR) analysis and the chemical composition of Aloe vera was determined using Gas Chromatography Mass Spectrometry (GC MS) whereas the physical parameters such as firmness, weight loss, odour, spoilage, and colour were evaluated, along with chemical parameters including pH, sugar content, and titratable acidity were carried out using standard methods. The results showed that fruits coated with Aloe vera gel exhibited reduced weight loss, slower firmness decline, delayed colour changes, and postponed spoilage compared to uncoated controls. Coated samples also retained higher sugar content and titratable acidity. In addition, while pH generally increased during storage, coated samples showed a lower rate of pH change compared to the controls, indicating delayed deterioration. The FTIR spectra of each fruit samples revealed similar compositions, with only slight variations in intensity. This indicates that while the Aloe vera gel preserved the fruits by forming an external protective barrier, it did not alter or interfere with their internal chemical composition. Among the treatments, the 30-minutes and 60 minutes application generally produced the most effective preservation across the three fruits. These findings highlight the potential of Aloe vera gel as an eco-friendly, natural coating for prolonging the postharvest quality of perishable fruits, thereby reducing food loss and enhancing marketability.

The study focused on the extraction, phytochemical screening, isolation, and characterization of Hibiscus Sabdariffa. The cold extraction was conducted to extract the plant sample using methanol as solvent. The crude extract was concentrated in a rotary evaporator. A portion of the extract was screened for phytochemicals while the other portion was subjected to vacuum liquid chromatography (VLC) , using solvent systems of varying polarity ;from Hexane, Ethyl acetate , Methanol ,and Butanol of increasing polarity respectively .A Brown precipitate was obtained after the concentration of Butanol extract and characterization was done by High Performance Liquid Chromatography (HPLC). Saponins, alkaloids, and terpenes were detected. However, from the HPLC myrtilin, Linalool, lycopene, where present amongst others. The study shows that H. Sabdariffa contains useful phytochemicals which are implicated as medicinal uses.

Tertiary institutions face significant challenges in managing student academic records, often relying on manual or semi-automated systems. These outdated methods frequently lead to delayed result processing, computational inaccuracies, and a lack of real-time access to academic performance data, ultimately hindering institutional efficiency and student development. This project addresses these issues through the design and implementation of a Result-Based Management System (RBMS). The system automates the computation and management of student results, featuring interactive dashboards for students and administrative staff, and generates standardized reports for lecturers, department administrators, and students. Using the Computer Science Department at the University of Benin as a case study, this project demonstrates the critical need for tertiary institutions to adopt automated solutions for tracking and analyzing student performance. The developed RBMS is a robust digital platform that includes modules for calculating Grade Point Average (GPA) and Cumulative Grade Point Average (CGPA), alongside providing real-time academic analytics. Constructed as a web-based application, the system utilizes HTML, CSS, PHP, JavaScript, jQuery, and MySQL. The project methodology follows the Structured System Analysis and Design Method (SSADM), ensuring a systematic approach by deconstructing complex processes into clear, manageable stages. The successful implementation of this system not only resolves the inefficiencies of manual result computation but also serves as a scalable model for digital transformation within academic administration.

Implementing blockchain-based certification systems is pivotal for enhancing the security, transparency, and efficiency of credential verification in academic and professional sectors. This study explores the design and deployment of an NFT-based certificate verification system, leveraging Ethereum blockchain, smart contracts, and decentralized storage (IPFS) to issue, verify, and manage digital credentials. By applying blockchain technology, smart contract logic, and decentralized file systems, this research addresses challenges such as certificate forgery, inefficient manual verification, and limited global accessibility. Key methodologies include the use of Solidity smart contracts for minting NFTs, Web3.js for blockchain interaction, and React.js for user interface development, alongside IPFS for secure off-chain certificate storage. The system is designed to reflect real-world use cases involving students, educational institutions, and verifiers, supporting operations such as certificate issuance, revocation, and verification. Outcomes include instant, tamper-proof verification processes, reduced operational costs, and improved user trust in credential authenticity. By integrating emerging technologies like Web3 and decentralized identity (DID), this research establishes a foundation for scalable, secure, and adaptable certification systems capable of meeting the evolving needs of education and professional validation

Protein folding is an important process that allows a long chain of amino acids (called a polypeptide) to form a specific three-dimensional shape needed for it to work properly. This project studies how proteins fold using the energy landscape model, which explains folding as a gradual movement toward the most stable and low-energy shape. By reviewing studies from 2020 to 2025 and comparing examples of amyloid-β, prion, and αsynuclein proteins, the research shows how changes in the folding process can cause proteins to fold incorrectly. When this happens, they may form clumps, leading to diseases such as Alzheimer’s and Parkinson’s. The findings show that protein folding is not random it is guided by chemical interactions, helper molecules called chaperones, and how easily different shapes can form. Misfolding occurs when proteins get stuck in the wrong shape, creating harmful structures. Overall, this study helps explain why protein folding is so important to human health. It also shows how understanding the energy landscape model can help scientists find better ways to prevent or treat diseases caused by misfolded proteins.

Expanding noncontact techniques for monitoring groundwater level variations to include laser based devices has been limited in the past by the low reflectivity of water to laser light The advances made in the realization of highly sensitive receivers has greatly enabled Laser devices with the prospect for effective noncontact monitoring of groundwater levels. In this work a real-time groundwater level monitoring devices was implemented using a LiDAR-lite v3HP laser-based sensor, an Arduino Uno rev 3 microcontroller, and an ESP8266 Wi-Fi module. Hardware implementation involved interfacing the microcontroller with the LiDAR-lite, the Wi-Fi module, a 16 x 2 liquid crystal display and other necessary basic electronic components. The software implementation involved writing, editing, compiling and uploading codes through the Arduino ide unto the microcontroller. The implemented prototype was powered by a 20,000mAH, 5V power bank and deployed on two artesian wells to collect depth to water surface data which were uploaded automatically to ThingSpeak platform. Measurements at various times taken with the device were compared with manual measurements from a calibrated tape. A correction factor was applied to correct measurement residuals. Results of the validation showed that measured values were uploaded to the ThingSpeak platform an average of 33 seconds which could be reconfigured to longer timeframes. The results were remotely accessed on the platform. A correlation graph of measurement before and after applying the correction revealed a near perfect correlation coefficient of 0.99995 for the LiDAR readings for both pre-correction and post correction measurements; confirming a strong linear relationship with tape measurements. The coefficient of variation, root mean square error and limits of agreement from the Bland Altman’s plots all ascertained the improvement in post-correction measurements.

Expanding noncontact techniques for monitoring groundwater level variations to include laser based devices has been limited in the past by the low reflectivity of water to laser light The advances made in the realization of highly sensitive receivers has greatly enabled Laser devices with the prospect for effective noncontact monitoring of groundwater levels. In this work a real-time groundwater level monitoring devices was implemented using a LiDAR-lite v3HP laser-based sensor, an Arduino Uno rev 3 microcontroller, and an ESP8266 Wi-Fi module. Hardware implementation involved interfacing the microcontroller with the LiDAR-lite, the Wi-Fi module, a 16 x 2 liquid crystal display and other necessary basic electronic components. The software implementation involved writing, editing, compiling and uploading codes through the Arduino ide unto the microcontroller. The implemented prototype was powered by a 20,000mAH, 5V power bank and deployed on two artesian wells to collect depth to water surface data which were uploaded automatically to ThingSpeak platform. Measurements at various times taken with the device were compared with manual measurements from a calibrated tape. A correction factor was applied to correct measurement residuals. Results of the validation showed that measured values were uploaded to the ThingSpeak platform an average of 33 seconds which could be reconfigured to longer timeframes. The results were remotely accessed on the platform. A correlation graph of measurement before and after applying the correction revealed a near perfect correlation coefficient of 0.99995 for the LiDAR readings for both pre-correction and post correction measurements; confirming a strong linear relationship with tape measurements. The coefficient of variation, root mean square error and limits of agreement from the Bland Altman’s plots all ascertained the improvement in post-correction measurements.

The optimization of frames per second (FPS) in camera systems is a critical factor in enhancing the efficiency and reliability of modern surveillance applications. FPS directly influences image clarity, motion detection accuracy, and the overall responsiveness of monitoring systems. High FPS ensures smoother video streams, enabling precise identification of fast-moving objects, while low FPS conserves bandwidth and storage resources. This study explores the balance between performance and resource management in surveillance environments, focusing on adaptive FPS optimization techniques. Key strategies include dynamic frame rate adjustment based on scene activity, integration of artificial intelligence for motion-triggered recording, and hardware-software co-optimization to reduce latency. By aligning FPS with contextual surveillance needs—such as crowd monitoring, traffic control, and perimeter security—camera systems can achieve improved situational awareness without excessive computational or storage overhead. The findings highlight that intelligent FPS optimization not only enhances operational efficiency but also strengthens the scalability and sustainability of surveillance infrastructures.

The process of estimating property prices in Nigeria remains largely manual, subjective, and
inconsistent, as real estate valuation often depends on personal experience, incomplete records,
or unstructured web data. These limitations create inaccuracies that affect buyers, sellers,
investors, and policymakers. To address these challenges, this study aims to develop a more
reliable and data-driven method for house price prediction using machine learning techniques.
The objectives include collecting and preprocessing a multi-state Nigerian dataset, engineering
relevant features, developing multiple predictive models, and evaluating their performance to
identify the most accurate approach.
A dataset consisting of 28,915 property records scraped from PropertyPro was cleaned,
normalized, and transformed through techniques such as one-hot encoding and Mutual
Information Regression. Four machine learning models, which are Linear Regression, Decision
Tree, Random Forest, and XGBoost, were implemented and evaluated using MAE, MSE, and R².
Quantitatively, the models achieved R² scores of 0.475 (Linear Regression), 0.433 (Decision
Tree), 0.517 (XGBoost), and 0.520 (Random Forest). Qualitatively, Random Forest
demonstrated superior ability to capture nonlinear patterns, handle noisy data, and generalize
across diverse property locations. The best-performing model was deployed through a Streamlit
web application to enable real-time house price prediction for end users.
Despite its success, the study is limited by the absence of macroeconomic variables such as
inflation and interest rates, which play a substantial role in housing market dynamics.
Additionally, challenges in obtaining comprehensive and well-structured real estate data across
all Nigerian states constrained the breadth of the analysis. Future research should integrate
geospatial and macroeconomic datasets to further enhance prediction accuracy and model
robustness.

Lifetime distributions are fundamental tools in statistics and probability theory, widely used to model the time until an event of interest occurs, such as the failure of a system, the death of an organism, or the occurrence of a financial default. This study focuses on reviewing three key lifetime distributions: The Exponential, Weibull, and Gamma distributions with an emphasis on their mathematical formulations, properties, and practical applications. The Exponential distribution, characterized by its memoryless property and constant hazard rate, is simple but limited in its applicability to more complex scenarios. The Weibull distribution, with its flexibility in modeling increasing, decreasing, or constant hazard rates, is highly versatile and widely used in reliability engineering. The Gamma distribution, capable of modeling multi-phase processes, is particularly useful for analyzing skewed data and complex lifetime behaviors. This study contributes to the theoretical understanding of lifetime distributions and provides practical guidelines for their application in fields such as reliability engineering, survival analysis, and risk assessment. Recommendations for future research include exploring additional distributions, analyzing larger datasets, and applying advanced estimation methods to further enhance the accuracy and applicability of lifetime distribution models.