http://ir.mu.ac.ke:8080/jspui/handle/123456789/26 2026-04-20T09:12:12Z http://ir.mu.ac.ke:8080/jspui/handle/123456789/10122 Title: Development, characterization and evaluation of selected transition metal doped zinc sulphide nanostructure surface layers decorated with graphene for water splitting Authors: Kiptarus, Joan Jeptum Abstract: Water splitting (WS) is the dissociation of Water (H2O) into Hydrogen (H2) and Oxygen (O2). Zinc Sulphide (ZnS) provides an excellent option for the hydrogen reduction cathode in photo electrochemical (PEC) cells for WS. However, its low sensitivity to visible range in electromagnetic spectrum limits its practical appli cability. Few comprehensive studies consider a wide range of transition metals as potential dopants to meet future energy requirements for greater PEC WS. The main objective of this research was to develop, characterize and evaluate the selected Transitional metal (TM) doped ZnS nanostructure (NS) surface layers decorated with graphene (rGO) for WS. The specific objectives were to: simu late the optimal dosage of TM dopants for ZnS nanostructure layers, synthesize TM doped ZnS NS layers decorated with graphene, characterize TM doped ZnS NS layers decorated with graphene and to evaluate the photocatalytic hydrogen production of TM doped ZnS NS layers decorated with graphene. Theoretical f irst principles Ab-Initio calculations based on Density functional theory (DFT) method was employed to examine the electronic structure of ZnS nanostructures (NSs) doped with selected TM dopants including; manganese (Mn), copper (Cu), cobalt (Co) and iron (Fe) in order to modify the structural properties of ZnS NSs. Highly distributed cobalt doped ZnS NSs were effectively fabricated on the surfaces of graphene sheets via simple hydrothermal technique. The structural, electronic and optical properties of the cobalt doped ZnS decorated with graphene (Co-ZnS-rGO-NS’s) were examined using X-ray diffraction (XRD), X-ray pho tocurrent spectroscopy (XPS), Raman spectroscopic (RS), Fourier transmission infrared spectroscopy (FTIR), Scanning electron microscopy (SEM) and Ultra violet visible absorbance spectroscopy (UV-vis). The photocatalytic activity of CoxZn1−xSrGO NS’s at (x = 0, 1, 2, 4 and 6) atomic percentage (atm.%) was determined in lab experiments using water and visible light. The stability of 3d orbital transitional metal dopant (TMD’s)’s in ZnS NSs were shown to be depen dent both on the dopant concentrations and the d orbital character of the TMD’s. Evidently, the 3d orbital TMD’s’s (Cu, Co,Mn and Fe) showed low formation energies and appropriate band edge states due to their low lattice strain, hence absorbed into ZnS NSs. ZnS doped with 4 atm.% of Cu and Co was shown to be optimal for photocatalytic hydrogen generation based on theoretical studies. The f indings of XRD, FTIR, RS, XPS and SEM investigation suggest that graphene oxide (GO) was successfully transformed into graphene sheets, CoxZn1−xSrGO NS’s possessed a crystalline, cuboidal and spheroidal form of structure displaying a paper like appearance. UV-vis spectrophotometric analysis verified a notable rapid increase in transmittance and high transparency (≈ 90%) within (180-800) vi nm wavelength range. Calculations of transmittance spectra revealed a direct allowable band gap range of (1.26-5.46) eV, demonstrating a band gap decrease as cobalt content increased, consistent with theoretical predictions. Furthermore, the optimal cobalt loading of 0.04 atm.% generated a maximum hydrogen yield of 7649µmolh−1 after 720 minutes of Ultra Violet (UV) light exposure, indicating that the ZnS NSs’s electronic and optical characteristics were influenced by their stability with respect to dopant concentration. In conclusion, the results show that improved transfer of photo-generated electrons, increased surface area and better dispersion-absorption properties all contributed to higher photocatalytic hydrogen generation activity. The study recommended synthesis optimization for commercially viable technology. 2025-01-01T00:00:00Z http://ir.mu.ac.ke:8080/jspui/handle/123456789/10114 Title: Technical assessment of large-scale integration of Solar electrification in energy systems in kenya Authors: Dominic, Ondieki Samoita Abstract: Kenya has witnessed a significant increase in electricity demand, reaching 1.5 GW in 2022 compared to a production of 12.65 TWh. This growth is primarily driven by population expansion and industrialization. However, continued reliance on fossil fuels remains environmentally unsustainable. To address this, the Kenyan government has set a target of achieving 100% renewable energy integration by 2030, with a strong emphasis on solar and wind energy. With its abundant solar resources, Kenya has the potential to generate more solar power than its total electricity demand. This thesis investigates the feasibility and impact of large-scale integration of solar power systems into Kenya’s energy mix. EnergyPLAN tool was employed to simulate hourly energy production and demand, enabling a comprehensive assessment of the technical, economic, and environmental implications. Cross-sectoral analysis was conducted to evaluate interdependencies and sectoral dynamics. A novel Whale Optimization Algorithm (WOA) based Maximum Power Point Tracking (MPPT) algorithm was developed in MATLAB and benchmarked against conventional methods, including Incremental Conductance, Fuzzy Logic, and Particle Swarm Optimization (PSO). Simulation results showed a 32% increase in solar power capacity—from 212.5 MW (6.8% of total generation) to 4,601 MW—at an annual cost of KSh 145.5 billion, compared to KSh 186.9 billion under the baseline scenario. With further solar power integration, optimal generation reached 10.01 TWh (39.56% of total), while renewable electricity output increased from 11.90 TWh to 19.76 TWh. CO₂ emissions dropped significantly from 0.134 Mt to 0.021 Mt, and total annual production costs decreased to KSh 134.3 billion. These findings demonstrate that optimized solar power integration offers substantial benefits in cost savings, emissions reduction, energy security, and system reliability. Sectoral Innovation System (SIS) analysis revealed that global cost declines primarily drive solar power adoption, with minimal local adaptation needed. The proposed WOA-based MPPT algorithm achieved a tracking efficiency of 99.95% with a steady-state error of 0.04%, outperforming PSO (99.7% efficiency, 0.2% error). Although PSO successfully tracked the global maximum power point, its dynamic response was inferior to that of the developed WOA-based MPPT system. 2025-01-01T00:00:00Z http://ir.mu.ac.ke:8080/jspui/handle/123456789/10029 Title: Sustainable bioethanol production from Zambian corn stover Authors: Mwanakaba, Cosmas S. Abstract: Commercialization of second-generation bioethanol production is hindered by the lack of sustainable, cost-effective, and environmentally friendly pretreatment technology. The use of Deep Eutectic Solvents (DES) is a promising alternative. This study aimed to optimize DES pretreatment of Zambian corn stover to maximize bioethanol production. The specific objectives were to determine engine performance and emissions of bioethanol/gasoline blends; ascertain the ideal conditions for cellulose yield, enzymatic hydrolysis, and bioethanol generation; and conduct a techno-economic feasibility study of major scale DES-based bioethanol production. The factors studied during pretreatment included time (6–15 hours), temperature (60°C–150°C), choline chloride to lactic acid ratio (1:2, 1:6, and 1:10), and substrate-to-solvent ratio (SLR) (1:08–1:32). Hydrolysis was conducted at temperatures between 45°C and 50°C for 60– 72 hours. Optimization of pretreatment and hydrolysis was performed using Central Composite Design (CCD), Response Surface Methodology (RSM), Artificial Neural Networks (ANN), and Gradient Boosted Regression Trees (GBRT). Mathematical models were developed to estimate cellulose and fermentable sugar yields. The optimal pretreatment conditions:105°C, 10.5-hour reaction time, and a 1:6 ChCl:LA ratio yielded a 46.1% cellulose recovery, with model predictions achieving 43% (quadratic) and 46.1% (GBRT) at R2 values of 91% and 80%, respectively. Optimal enzymatic hydrolysis conditions enzyme loading of 10 mg per gram of biomass, 50°C, and 72- hour reaction time resulted in a fermentable sugar yield of 78%, validated through High- Performance Liquid Chromatography (HPLC). Fermentation using Saccharomyces cerevisiae produced bioethanol with an 80% yield, confirmed via Gas Chromatography- Mass Spectrometry (GC-MS). Distillation was conducted at 78.5°C using a computer- controlled bioethanol process unit. Through laboratory-level distillation, 2.82 g of bioethanol was obtained, leading to a final production volume of 3.57 L. Bioethanol/gasoline blends (G100, E10, E20, E30, and E40) were tested on an Atico computer-controlled hybrid test bench engine. Brake power and brake specific fuel consumption (BSFC) results were 31.42, 32.72, 34.03, 30.11, and 28.8 kW and 0.2706, 0.2516, 0.2333, 0.2765, and 0.3194 kg/kWh for G100, E10, E20, E30, and E40 blends, respectively. E20 provided the best balance between performance and emissions, increasing brake thermal efficiency (BTE) by 7.4% while reducing carbon monoxide (CO) and hydrocarbon (HC) emissions by 21% and 26%, respectively. Higher ethanol blends (E30 and E40) further reduced emissions but required modifications in ignition timing and fuel injection for optimal engine performance. A techno-economic analysis (TEA) assessed the feasibility of scaling up DES-based bioethanol production for a 50,000-liter capacity plant. The DES process was found to be 27% more cost-effective than conventional methods due to the recyclability and biodegradability of lactic acid and choline chloride, reducing overall fuel costs. A life cycle assessment (LCA) showed a 32% reduction in greenhouse gas emissions compared to fossil fuel-based gasoline. The results confirm the potential of DES-based pretreatment to enhance bioethanol production and improve economic viability. 2025-01-01T00:00:00Z http://ir.mu.ac.ke:8080/jspui/handle/123456789/10022 Title: Process simulation and machine learning modeling of biomass wastes co-gasification for syngas and biochar production Authors: Bongomin, Ocident Abstract: The urgency of climate change has accelerated research into renewable energy, including biomass co-gasification. While biomass waste offers a sustainable resource for producing syngas and biochar, the conversion process remains complex due to variability in feedstock properties, reactor design, and operating conditions. Additionally, traditional experimental and mechanistic modeling approaches are often time-consuming, costly, and limited in generalizability. This study addresses this gap by integrating experimental characterization, process simulation (PS), and machine learning (ML) to enhance understanding and prediction of autothermal biomass co-gasification outcomes. The primary objective of this research is to develop predictive models to optimize syngas and biochar production. Specifically, the study characterizes the physico-chemical and thermo-kinetic properties of five biomass feedstocks (coffee husks, groundnut shells, macadamia nutshells, rice husks, and tea wastes); develops a PS model to represent biomass co-gasification dynamics; develops and validates ML models to predict co- gasification outcomes; assess model robustness using new biomass blends; and evaluates the impact of model interpretability techniques on feature importance rankings. Proximate analysis method followed ASTM E1131-08, while ultimate analysis was conducted using a Carbon-Hydrogen-Nitrogen-Sulfur (CHNS) analyzer. Thermogravimetric analysis under a nitrogen atmosphere was employed to study thermal degradation, and kinetic parameters were estimated using the Coats-Redfern method. Aspen plus PS method was used to simulate a pilot-scale downdraft gasifier. ML models (Random Forest, Artificial Neural Networks, Gradient Boosting Regression, Support Vector Regression, and SuperLearner ensembles (SLE)) were developed and validated in MATLAB. Robustness was tested by validating the models with new feedstock blends. Interpretability was evaluated using permutation importance, Gini importance, and partial dependence plots. Proximate and ultimate analysis results revealed variability among feedstocks. Volatile matter (63.96% ±3.57%) indicated high syngas and tar potential, fixed carbon (19.62% ±2.69%) contributed to char formation, and carbon content (47.69% ±4.80%) suggested high energy conversion efficiency. Thermo-kinetic analysis showed peak devolatilization temperatures between 345°C and 380°C, activation energies ranging from 39 to 46 kJ/mol, and Gibbs free energy values between 151 and 162 kJ/mol, indicating favorable decomposition behavior. PS model achieved high accuracy with temperature deviations of 2°C (pyrolysis), 4°C (combustion), and 7°C (reduction), and syngas yield deviations of 0.21 Nm3/kg (Equivalence Ratio (ER) 0.17) and 0.34 Nm3/kg (ER 0.29). Sensitivity analysis showed that increasing ER enhanced hydrogen concentration by 10–15% and reduced carbon dioxide by 20–25%. All ML models performed well with Coefficient of Determination (R2) > 0.90 and Root Mean Square Error (RMSE) < 5%, confirming their effectiveness. Robust analysis showed that SLE maintained superior generalizability (R2 > 0.75, RMSE < 5%) for predicting char and syngas yields. However, hydrogen and cold gas efficiency predictions were less robust, indicating needs for more diverse datasets to improve generalization. Interpretability analysis identified ER and steam-to-biomass ratio as key predictors, with macadamia nutshells playing a critical role in enhancing char yield. In conclusion, this study demonstrates biomass waste co-gasification using integrated experiments, simulation, and ML, providing insights for sustainable energy. The findings recommend applying the developed ML models in academic research and practical gasification systems to support reliable process prediction and decision-making 2025-01-01T00:00:00Z