TUK offers PhD programmes by research and thesis in all its three Faculties. The programmes are geared towards capacity building for TUK staff and are also open to external applicants from the East African region and beyond.
Recent publications report some of the findings obtained from postgraduate student research.

Research on Broadcast Telecommunication Channels:
W. Ojenge (ojenge@tukenya.ac.ke), W. Okelo-Odongo and P. Ogao, 2015.

Comparing PSO and GA optimizers in MLP to Predict Mobile Traffic Jam Times. International Journal of Computer Science and Information Security (IJCSIS), Vol. 13, No. 10, Pp. 19-30, October 2015. ISSN 1947 – 5500 .

Abstract: Freely-usable frequency spectrum is dwindling quickly in the face of increasingly greater demand. As mobile traffic overwhelm the frequency allocated to it, some frequency bands such as for terrestrial TV are insufficiently used. Yet the fixed spectrum allocation dictated by International Telecommunications Union disallows under-used frequency from being taken by those who need it more. This under-used frequency is, however, accessible for unlicensed exploitation using the Cognitive Radio. The cognitive radio would basically keep monitoring occupation of desirable frequencies by the licensed users and cause opportunistic utilization by unlicensed users when this opportunistic use cannot cause interference to the licensed users. In Kenyan situation, the most appropriate technique would be Overlay cognitive radio network. When the mobile traffic is modeled, it is easier to predict the exact jam times and plan ahead for emerging TV idle channels at the exact times. This paper attempts to explore the most optimal predictive algorithms using both literature review and experimental method. Literature on the following algorithms were reviewed; simple Multilayer perceptron, both simple and optimized versions of support vector machine, Naïve Bayes, decision trees and K-Nearest Neighbor. Although in only one occasion did the un-optimized multilayer perceptron out-perform the others, it still rallied well in the other occasions. There is, therefore, a high probability that optimizing the multilayer perceptron may enable it out-perform the other algorithms. Two effective optimization algorithms are used; genetic algorithm and particle swarm optimization. This paper describes the attempt to determine the performance of genetic-algorithm-optimized multilayer perceptron and particle-swarm-optimization-optimized multilayer perceptron in predicting mobile telephony jam times in a perennially-traffic jammed mobile cell. Our results indicate that particle-swarm-optimization optimized multilayer perceptron is probably a better performer than most other algorithms.

Winston Ojenge et al, 2016. Use of GA-optimized NN to predict DVT-T2 receiver spectrum holes. International Journal of Simulation: Systems, Science and Technology, Vol. 16. DOI 10.5013/IJSSST.a.16.04.12. ISSN: 1473-804x online, 1473-8031 print.

Abstract: Cognitive radio enables opportunistic utilization of under-used spectrum for networks that are overwhelmed. In Nairobi city, mobile telephony networks are overwhelmed while broadcast TV channels lie idle in some parts. Research in cognitive radio concentrates on point-to-point communication and successfully conducts transmitter sensing in order to establish spectrum holes. However, in broadcast communication such as terrestrial TV, transmitter detection is inefficient, as transmitter signals may be present yet that licensed channel is not tuned-into by TV receivers, rendering those frequencies essentially idle. Our paper describes an attempt to use a novel remote technique to detect, model and predict which DVB-T2 channels are not tuned-into by any TV receiver during the worst mobile traffic jam time, within an overwhelmed mobile cell. An MLP that was GA-optimized to an MSE of 0.0046245 indicated predictability of TV-viewing and predicted Channel 514 MHz as being idle during the 5.00-5.03 pm jam time slot.

Winston Ojenge is currently a lecturer in the Department of Electrical and Electronic Engineering, Technical University of Kenya. He has recently completed and submitted a PhD thesis for examination. He can be contacted at ojenge@tukenya.ac.ke

---

Research on Reinforced Polypropylene Composite Materials

Odhong, O.V.E., Muumbo A.M. and Mayaka, A.N., 2016.

Improving Impact Strength Recovery of Fractured and Healed Rice Husks Fibre Reinforced Polypropylene Composites).
Int. Journal of Engineering Research and Application www.ijera.com ISSN : 2248-9622, Vol. 6, Issue 10, ( Part -2) October 2016, pp.76-83 www.ijera.com 76|P a g e DOI: 10.5923/j.cmaterials.20170702.01

Abstract: Rice husks fibre reinforced polypropylene composite (rhfrpc) is a natural plant fibre reinforced polymer composite having advantages of high strength, light weight and affordability. They are commonly used for light load structural and non structural applications. They are mainly used as particle boards, for fencing post, roofing tiles, for interiors of car and aircrafts among other usages. This material once cracked by impact forces cannot be repaired using traditional repair methods for engineering materials such as metals or other composites that can be repaired by welding or by patch repair methods respectively, thus a method of repair of rice husks fibre reinforced polypropylene composites by refilling the damaged volume by injection of various healing agents has been investigated. The composite coupons were produced by injection moulding, cooled sufficiently and prepared for charpy impact tests. Test results for pristine coupons were a maximum of 48 J/mm2. The destroyed coupons were then subjected to healing in a fabricated healing fixture. Healing agents such as epoxy resin, ethyl cyanoacrylate, and tannin gum have been investigated for their use as possible healing agents to fill the damaged volume and perform healing action at the fractured surfaces. The impact test results were recorded and compared with those of unhealed pristine coupons. The recovered strengths were a maximum of 60 J/mm2 translating into a 125% impact strength recovery, and this is good enough for the healed composites to be recommended for reuse in their second lives of their respective original functions.

Odhong O. V. E., Muumbo A. M., Mayaka A. N. 2017. Recovering Impact Strength of Fractured Rice Husks Fibre Reinforced Polypropylene Composites Using Healing Agents
International Journal of Composite Materials 2017, 7(2): 37-45.
DOI: 10.5923/j.cmaterials.20170702.01.

Abstract: Plant fibre reinforced polymer composites, of which rice husks fibre reinforced polypropylene composite is an example, are gaining prominence in engineering applications. Rice husks fibre reinforced polypropylene composite is ecofriendly, affordable and it is commonly used for light load structural and non structural components as well as structures. When fractured by rapidly or gradually applied loads, the rice husks fibre reinforced polypropylene composite cannot be economically repaired using traditional repair methods such as patch, bolted or scarf repair which would be applicable for leaf based as well as seed hair based fibre reinforced polymer composites. In this research, repair by refilling the damaged volume by injection of healing agents has been investigated. The repair helps in recovering part of initial impact strength lost during destructive testing. The composite test pieces were produced by injection moulding, cooled and tested for impact strengths. Impact strength values for pristine test pieces were a maximum of 61 J/mm2. The destroyed test pieces had their fractured surfaces healed by injecting various healing agents and retesting the samples destructively after healing. The impact strengths for healed test pieces were recorded and compared with those of pristine test pieces. The recovered impact maximum strength after healing and retest reached 60 J/mm2 translating to 98.36% of the initial impact strength. The injection of healing agents can therefore restore rice husk fibre reinforced polypropylene composite strength for reuse in light load structural applications.

Edward Odhong is currently a lecturer in theFaculty of Engineering and Technology, Multimedia University of Kenya, and a registered PhD candidate in the Department of Mechanical and Mechatronic Engineering, Technical University of Kenya. He can be contacted at odhongedward@yahoo.com.

---

Research on Biodiesel Production and Energy Usage

Jeremiah O. Masime, Eric Ogur, Betty Mbatia, Joseph Lalah, Austin O. Aluoch and Geoffrey Otieno 2017. Extraction and Characterization of Yellow Oleander (Thevetia Peruviana) Seed Oil, Meal and Analysis of Eggshells’ Catalyst as Potential Feedstocks for Biodiesel Production in Kenya. Journal of the Kenya Chemical Society 10-1 (2017), 58-69; Available online at www.kenchemsoc.org.

Abstract: Biodiesels have a number of advantages such as reduced emission, improved combustion, although their energy efficiency is still lower than that of petro diesel. However biodiesel technology can provide a sustainable energy solution in decentralized systems such as large scale agroforestry farms and reduce dependence on petroleum diesel. Thevetia peruviana (Yellow Oleander) seed oil was extracted using n-hexane as a solvent in a Soxhlet extractor. The percentage yield was 61.78 ± 1.53 %, which makes yellow oleander seed a good source of the vegetable oil. The extracts were then subjected to physico-chemical analysis using ASTM and EN standard methods. The composition of the incinerated waste eggshells catalyst, yellow oleander oil and meal were ascertained using XRF analysis. The calorific values of the meal and oil were found to be 29.8986 and 38.6470 MJ/kg, respectively. The physico-chemical parameters for the biodiesel tested included specific gravity, free fatty acid, saponification value, peroxide value, iodine value and the yield. The spectroscopic data of the oil were obtained by GC/MS, FTIR and UV Visible spectroscopic analysis. The eggshells also registered relatively high levels of CaO, MgO and Al2O3 and other metals, which makes it a good heterogeneous catalyst. The studied physicochemical characteristics of the yellow oleander oil extract compared favorably with the conventional oil found in the Kenyan market.
Jeremiah O. Masime, Eric Ogur, Betty Mbatia, Austin O. Aluoch and Geoffrey Otieno 2017. Optimization and Thermodynamics of the Extraction of Yellow Oleander Seed Oil Using Soxhlet Extractor. Journal of Scientific Achievements, January 2017, Vol. 2, No. 1, Page: 43–50; Available online at www.jsciachv.com
Abstract: Optimization of yellow oleander (Thevetia peruviana) oil extraction from yellow oleander seeds was done using a Soxhlet extractor. The following parameters were studied in order to measure their influence on batch extraction using petroleum ether (60 – 80℃): particle size, solid-solvent weight ratio, the effect of particle size and time, and temperature on yield, the maximum yield, and optimal conditions were obtained when using petroleum ether at a temperature of 343K with a contact time of 36 hours, average particles size >1 mm, and meal to solvent ratio of 1:6. The thermodynamics study of the extracted oil showed that ΔEa, ΔG and ΔH were positive while ΔS was found to be negative, indicating that yellow oleander oil extraction process is not a spontaneous process at all temperatures.

Jeremiah Masime is currently serving as a part-time lecturer in the Department of Chemical Science and Technology, Technical University of Kenya, where he is also a registered PhD student. He can be contacted at: masimejo@gmail.com

---

Research on Nanotechnology

James Jorum Owuor, Florence Oloo , Martin Ongas,, Caroline Kirimi,Wesley Nyaigoti Omwoyo, and Jeremiah Waweru Gathirwa (2016). Development and validation of a LC/MS method for the quantitation of nanoformulated Primaquine in whole blood and plasma of mouse model
International Journal of Life-sciences Scientific Research (IJLSSR) (in press).
Abstract: A Liquid chromatography-mass spectrometry (LC/MS) method was developed and validated for the quantitation of the antimalarial drug, Primaquine (PQ), in whole blood and plasma. The analyte was extracted using a protein precipitation method followed by chromatographic separation on a Waters Xterra, RP C8,2.5µm, 50mm x 4.6mm analytical column with a mobile phase consisting of  A: 0.5% Formic acid in 20mM NH4COOH, B: Methanol pH adjusted to 3.0 with FA at a ratio of 3:7 (v/v), delivered at a constant flow rate of 0.5 ml/min. Mefloquine (MEF) was used as the internal standard. Compound reaction monitoring was performed using 260.3 Da for precursor ion and175. 2and 379.2 Da for product ions for the quantification of PQ and 379.2 Da for precursor ion and 175.2 and 379.2 Da for product ions for the quantification, respectively. Calibration curves were constructed over the concentration range 16.7–4300 ng/ml. The mean intra- and inter-assay accuracy values for the analysis of PQ in WB was 104% (%CV = 5.6) and 98.6% (%CV = 5.7), respectively. The mean intra- and inter-assay accuracy values for the analysis of PQ in plasma was 92.7% (%CV = 3.7) and 93.7% (%CV = 5.4), respectively. No significant matrix effect was observed during the method validation. The validated method was applied to an absorption study in mice, to determine and compare PQ concentrations in whole blood and plasma samples. Results of the statistical analysis using a linear mixed effects growth curve model concluded that there was no significant difference (p-value = 0.688) between WB and plasma PQ concentrations. This method utilizes a small sample volume of 20 µl, facilitating low blood collection volumes and a short chromatographic run time of 3 min which allows for high sample through put analysis.
James Jorum Owuor, Florence Oloo , Kivunzya Mwaiwa,Wesley Nyaigoti Omwoyo,, and Jeremiah Waweru Gathirwa (2017). Comparison of freeze drying and spray drying to obtain primaquine-loaded solid lipid nanoparticles. International Journal of Life-sciences Scientific Research (IJLSSR) (submitted).

Abstract: Spray drying and freeze drying are well established processes for solid lipid particles nanoformulation. Encapsulation of drugs into nanoparticles is receiving increased interest . In this study the spray drying and freeze drying of different nanosized Solid Lipid Nanoparticle (SLN) and the physicochemical characteristics of the obtained particles (size distribution, morphology, surface area, and drug release parameters) are studied. Primaquine loaded Solid Lipid Nanoparticle(PQ-SLN) dried by both methods was studied. The influence of temperature, pressure and nozzle diameter on the morphology and characteristics of spray dried PQ-SLN was observed while the effect of air pressure and nozzle diameter on the morphology and properties of freeze dried PQ-SLN was evaluated. The influence of drug loading of the starting nanoparticle was also studied. It has been demonstrated that these processing parameters have practically no influence on the particle morphology and properties, and the only parameter determining the particle characteristics is the drug content of the nanoparticle, either in the spraying or in the freezing process. Spray drying leads to a monomodal distribution with higher particle size, while freeze drying produces more porous particles, with a bimodal intra-particle distribution. The drug release profile of spray-dried SLN is better than that of the freeze-dried SLN. As a result, the characteristics of the spray-dried SLN are better than freeze dried SLN.

James Jorum Owuor is a Tutorial Fellow in the Department of Chemical Science and Technology, Technical University of Kenya. He is a registered PhD student in TU-K and his thesis is on nano-formulation of drugs. He can be contacted at: jjamesowuor@gmail.com.

---

Research on Pesticides

Boniface Mbithi Muendo, Zachary Moranga Getenga, Joseph Owuor Lalah, Shem Oyoo Wandiga, Michael Schmid, Michael Rothballer, Ulrike Dorfler (2017). Hexazinone degradation by a Bacillus sp. and Staphylococcus sp. isolated from pineapple and sugarcane cultivated soils in Kenya. Toxicological and Environmental Chemistry journal (in press).
Abstract: A study was undertaken to characterize bacterial isolates capable of degrading hexazinone from tropical soils with long term application history in Kenya. Hexazinone is an s-triazine herbicide widely used for control of weeds in sugarcane and pineapple farms in Kenya. Soil samples from farms with prior exposure to hexazinone were screened for enhanced degradation by adapted microbes. The non-sterile soil samples taken from sugarcane and pineapple cultivated farms which had been exposed to hexazinone for over 15 years were found to degrade 82.2 % and 93.4 % of 5 mg/l initially applied hexazinone, respectively, in a laboratory experiment after 146 days of incubation. In contrast, non sterile soil samples from control farms with no previous hexazinone application degraded 48.8 % and 36.8 % of the initially applied hexazinone, respectively, after 146 days of incubation. From the laboratory studies, two bacterial strains capable of hexazinone degradation were isolated from sugarcane and pineapple-cultivated soils. A comparative sequence analysis of 16S rRNA coding genes, showed highest similarities of 97.5% to Staphylococcus gallinarum and 99.8% to both Bacillus toyonensis and Bacillus thuringiensis. Of the 5 mg/l initially applied hexazinone, the two isolates degraded up to 38.4 % and 53.2 % of the herbicide, respectively, after 46 days. The native strains isolated have a great potential in bioremediation of hexazinone-contaminated soil sites.

Bonface Mbithi Muendo is working with the Kenya Bureau of Standards (KEBS). He is currently a PhD student in the Department of Chemical Science and Technology, Technical University of Kenya. He can be contacted at: kaske18@yahoo.com.