M.Ing. / Fouling and more specifically scaling is one of the most discussed and not yet well understood problems facing engineers. Scale formation which has enormous economic and environmental consequences does not only influence heat transfer characteristics and pressure drop through tubes, but influence the volumetric flow rate as well. To solve this problem Physical Water Treatment (PWT) devices have been developed which reportedly reduce scale formation. The efficiency of PWT devices is a controversial subject and many contradictions exits in the claimed effect. Researchers have been trying to prove the efficiency of PWT devices but in most cases the results are typically characterised by low reproducibility. An experimental measuring technique is thus required to measure scale formation and the influence PWT devices has on scaling. This technique must also produce reproducible results to clarify the questions surrounding PWT devices. This thesis is an experimental investigation in which an experimental setup and measuring technique is developed. It is used to evaluate a permanent magnet for the decrease of scale formation in tubes. The volumetric flow rate is used as the indicator of scale formation, which relies on the basic principle that the friction pressure drop increases if scaling takes place. The volumetric flow rate is monitored by measuring the time necessary to fill a container of known volume. The test section consists of three soft drawn copper tubes in which water with a velocity of ± 2 m/s flows. One of the tubes is used to evaluate the PWT device and the other two tubes are used as a control. The water temperature is kept at 53°C. The experiments are done in five phases in which the PWT device is placed onto one of the tubes in phase four and removed in phase five. It follows from the results obtained that a very sensitive experimental setup was designed and built which can be used to evaluate the efficiency of PWT devices. It follows from the variation in time necessary to fill the container in phase four and five that a clear conclusion can not be made. More experiments must be done in which reproducible results are obtained.

Ph.D. / Natural Organic Matter (NOM) consists of a highly variable mixture of products found in various types of water and soils, formed as a result of decomposition of plant and animal material into water. NOM is a precursor for the formation of disinfection by-products (DBPs) during water disinfection. It causes undesirable colour, taste and odour in water. The reduction of NOM affects the capacity of other treatment processes to remove organic micro-pollutants or inorganic species that may be present in the water. The differing levels and composition of NOM in South African water sources in different regions suggests that a water treatment strategy by a treatment plant should address the issues relating to specific NOM and its treatability. Therefore, in order to reduce NOM from water in a water treatment train, the composition of the NOM in the source water must be taken into account, since its composition at any given time may depend on the local prevailing conditions. The primary objective of this study was therefore to characterize NOM present in South African source waters through extensive sampling of representative water types at identified regions in the country and develop a rapid NOM characterization protocol. Water samples were thus collected from eight different water treatment plants broadly located within the five major source water types of South Africa between the periods of February 2010 to June 2011 in five sampling rounds. The sampling campaigns were planned and timed in order to accommodate the dry and the rainy seasons. The water treatment plants sampled were: Loerie (L) Water Treatment Plant, Midvaal (M) Water Treatment Plant, Olifantsvlei (O) Wastewater Treatment Plant, Plettenberg Bay (P) Water Treatment Plant, Rietvlei (R) Water Treatment Plant, Umzoniana (U) Water Treatment Plant, Vereeniging (V) Water Treatment Plant, and the Wiggins (W) Water Treatment Plant. The L water is from a small impoundment, which is fed from the Kouga Dam. The M water is pumped from the Vaal River.

In the biostabilization phase, a BOD₅:N:P loading of 100:3.2:1.1 was found to be "adequate" for treatment, while the standard nutrient loading of 100:5:1 was found to be "excessive". This was evident by the much higher nitrite-nitrate concentration in the effluent of the BOD₅:N:P = 100:5:1 reactor. Organic removal by the first stage units was excellent. BOD₅ and COD removals of at least 99.4 and 96.4 percent, respectively, were achieved under all conditions investigated, except for the two units close to washout conditions (the 5-day sludge age units at 10° and 5°C). Temperature and sludge age also had minimal effects on the removal of metals, except under the two conditions mentioned above; removals were greater than 90 percent, for most of the metals monitored. The reactors only reduced magnesium concentrations by 32.5 to 52.7 percent, mainly because the mixed liquor pH's (about 8.5) were not high enough for magnesium precipitation as magnesium hydroxide. For the lime-magnesium polishing step, samples were dosed with lime to pH levels of 10.0, 10.7, and 11.4. Magnesium doses of 0, 10, 20, 35 and 50 mg/L were then added to the samples at each pH level. In general, removals of impurities were not enhanced significantly by these magnesium additions. This was due, in part, to the initial low concentrations of contaminants; in addition, there already existed greater than 20 mg/L of magnesium in the samples. Aerobic biostabilization, at a sludge age greater than 15 days, at BOD₅:N:P = 100:3.2:1.1, and liquid temperatures of at least 3°C, followed by lime precipitation (to pH greater than or equal to 10.0), is capable of reducing most contaminants of a medium strength leachate (BOD₅ = 8090 mg/L) to levels below local (Province of British Columbia) pollution control objectives. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

This thesis discusses the causes and methods for removing taste and odor compounds from water supplies.

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of a Master of Science degree in Chemistry. University of the Witwatersrand, May 2013 / The pollution of water sources by chemical and biological species has created a serious water crisis all over the world. Such pollution has placed severe strains on the limited water sources resulting in the spread of waterborne diseases, which continue to be the leading causes of deaths in developing countries. Pollution by organic species still poses a serious health and environmental problem. Attempts to mitigate this problem are on-going and a number of methods are employed currently. Activated carbon and reverse osmosis are some of the current techniques that are used for the removal of organics in water. However, these techniques are limited in the removal of pollutants at lower concentrations (ng/L). Recent studies demonstrated the efficient removal of organics by nanoporous cyclodextrin (CD) polymers, a class of nanomaterials with great potential in absorbing organic pollutants from water. This project reports on the synthesis of β-cyclodextrin (β-CD) based polymer nanocomposite materials (nanocomposites) that have been blended with nitrogen-doped carbon nanotubes (N-CNTs) and silver (Ag) nanoparticles for water treatment. Prior to this study, the synthesis of these nanocomposites has been based on a conventional method that involves heating the reactants in a round bottom flask for 16 – 24 h. In this study a new method that is efficient, greener and time saving is reported. This facile method involved synthesizing the polymer nanocomposites under microwave irradiation wherein complete synthesis of the polymer nanocomposites was achieved in 10 min. N-CNTs were first synthesized via modified chemical vapour deposition method (CVD) using a 10wt% Fe-Co/CaCO3 catalyst. The N-CNTs were found to contain ~ 2% nitrogen by CN and XPS analysis. The N-CNTs were of high purity and were oxidized with acid functional groups (-COOH, -C=O, -OH) using nitric acid under reflux. Zeta potential studies indicated that the quantity of acid functional groups increased with increase in acid treatment time. The functionalised N-CNTs (fN-CNTs) were then decorated with Ag nanoparticles using microwave irradiation and further polymerized with β-CD using hexamethylene diisocyanate (HMDI) as the linker in an industrial microwave under an inert gas atmosphere of N2. Two types of polymer nanocomposites were synthesized namely, N-CNTs/β-CD and Ag/N-CNTs/β-CD. Different synthesis parameters such as microwave power and time were varied during the synthesis of these composites to study their effect on the result materials. Different level of power, 400 W, 600 W and 800 W were tested and surface area and morphology data indicated that all these powers can be used in synthesising the polymer composites. The optimum power used was 600 W, which gave highly porous, less densely packed morphology and a higher surface area of the polymers. The synthesis time was varied for 10 min, 15 min and 30 min. An irradiation time of 10 min was found to be sufficient for the synthesis of the nanocomposites. The polymers showed an efficient removal of p-nitrophenol, bisphenol A and trichloroethylene (TCE) from spiked water as confirmed by UV-Vis spectroscopy and GC/MS analysis.

Thesis (M.Sc.(Civil Engineering))--University of the Witwatersrand, Faculty of Engineering, 1989.

Drinking water samples provided by Orlando Utilities from various locations and from Florida Technological University were analyzed for organohalides. Compounds tentatively identified were 0.0024 mg/l DDT, 0.003 mg/l phosdrin, 0.00036 mg/l BHC, and 0.000095 mg/l endosulfan. These concentrations are well below recommended limits for drinking water. Two water samples were mixed with 10 mg/l of 20-40 mesh granular activated carbon, agitated for 2 minutes and then allowed to settle in the flask for 45 minutes. A reduction in the organohalide concentration varied from 25 to 98 percent of the original concentration. Also, a solution of seven common pesticides of 0.1 mg/l each were mixed with 1, 5, 10, 25, and 50 mg/l of carbon. The results indicate a reduction of 99 percent or better for all pesticides with 5 ppm carbon concentration.

This study investigated the effectiveness of the magnesium coagulation process in reduction of trihalomethane (THM) precursors. The water was obtained from Lake Washington, a highly colored potable water supply which is used by the City of Melbourne, Florida. The THM concentrations in the finished water at Melbourne currently exceed the THM standard of 0.1 mg/l. For Lake Washington water, treatment varies according to seasonal changes in water quality. During the dry period, the recycle magnesium does not perform effectively as a coagulant for THM precursor removal. This is because of the high levels of magnesium in the late water at this time. During the wet period, when magnesium concentrations are very low, it is much more effective. Magnesium sulfate was found to be effective in reducing THM precursors for both types of water. There was a direct correlation observed between THMFP, TOC and color observed in treated water samples. The lack of effectiveness in THMFP, TOC and color removals by recycle magnesium was not found to be due to a lack of magnesium precipitation. Variations in rapid and slow mixing times had no effect on THM precursor removal, but did affect settling of the floc. Addition of alum as a polymer at high pH values was also very effective in reducing the THMFP, TOC and color, and increasing floc sedimentation.

A kaolin in tapwater suspension was treated either with alum or one of two cationic polymers. The resulting suspension was filtered through various configurations of felt cloth filters. The kaolin suspension was adjusted to 100 JTU. Treatment by polymer consistently lowered the filtered effluent to less than 1 JTU. Jar tests were utilized to determine optimum coagulant dose for the tests. The formulation of a complete filter cake in the filter appeared to be the determining factor in the efficiency of water clarification by the filter. The literature lacked references to this concept as applied to water supply problems. The results of this report suggest that continuous filtration should now be examined.

Lake restoration procedures are reviewed with the objective of making recommendations toward future lake restorations for central Florida lakes. The progress of drawdown for restoration is examined. A total phosphorus model is developed for Lake Eola to predict the success of the drawdown. No one restoration procedure should be applied to every lake in central Florida. Each situation must be studied separately. It appears that lake drawdown was the best technique to use for Lake Eola. However, the stormwater inputs and free-flowing drains must be purified to insure the success of the drawdown or the Lake will return within the immediate future to its previous eutrophic state.