Construction projects involving the installation or repair of subsurface structures or utilities often require dewatering to induce a temporary lowering of the local groundwater elevation to facilitate construction. In the event that a known contaminant plume is present in an adjacent area, this dewatering may inadvertently draw the contaminant into the previously uncontaminated work area. Uretek Holdings, Inc. has developed its Injectable BarrierSM to be installed prior to dewatering exercises to provide a groundwater cut-off by reducing the potential movement of groundwater due to the hydraulic gradient induced by dewatering. A benefit of Injectable BarrierSM as compared to conventional methods of hydraulic control is that excavation is not required prior to its installation and no excess soils are generated through its installation. Injectable BarrierSM is a proprietary process registered with the United States Patent and Trademark Office by Uretek Holdings, Inc. Since methodical in-situ testing of the effectiveness of the Injectable BarrierSM has not been performed to date, it was the focus of this research to test the performance of the barrier under in-situ conditions utilizing a subsurface environment indicative of a West-Central Florida location. A testing plot to perform this research was selected on Hillsborough County property in Tampa, Florida which provided both a relatively shallow groundwater elevation in addition to a clay confining layer at a relatively shallow depth, making this an ideal location for testing the performance of the Injectable BarrierSM. After establishing the native conditions through baseline pump testing and repeating the testing procedure following the installation of the Injectable BarrierSM, a quantification of the reduction in hydraulic conductivity was achieved. Pumping tests were performed on the Injectable BarrierSM at its standard spacing as well as modified versions of the barrier with variation in the lateral spacing to include 6 foot, 4 foot, 3 foot, and 2 foot injection patterns to determine if a modified injection process could improve its performance. The 3 foot lateral spacing corresponding to the standard Injectable BarrierSM process indicated a 20% reduction in the hydraulic conductivity following its installation. By performing a small scale excavation following the completion of all pumping tests, it was discovered that the dispersion of the material in the subsurface appeared insufficient to provide the coverage needed to establish a barrier capable of further reducing the local hydraulic conductivity, especially at the shallowest injection depth of 3 feet below land surface (ft bls). It is concluded that modified amounts of injected material, closer lateral injection spacing, and potentially modified injection temperatures and component ratios could increase the effectiveness of the Injectable BarrierSM.

Thesis (MEng)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: The purpose of this study is to construct and test a surface tension driven water pump. The surface tension driven water pump is a passive water pump which uses a similar mechanism to that of trees to pump water. This study was conducted at the Department of Mechanical and Mechatronic Engineering at the University of Stellenbosch. For the study an extensive literature survey was done encompassing aspects such as water properties, surface tension (basic principles, capillary forces, temperature and contaminant effects, wettability), bubble formation (nucleation theory and tensile strength of water) and, finally water and mineral transport in trees (plant structures and mechanisms, limiting factors, misconceptions and organic substance transport). Previous work by botanists who demonstrate the transpiration mechanism needed for water transport in trees was also considered. The study further required the development of a theoretical thermal-hydraulic model to simulate the pumping performance for the surface tension driven water pump. The developed water pump was also experimentally tested with particular focus on design improvement, pumping performance, pump behaviour, potential pumping head as well as water collection capability. The experimental data was statistically analysed by multi-linear regression. Both the experimental data and statically generated predictions were compared to the theoretical thermal-hydraulic model. The results show that a working surface tension driven pump was constructed. Evaporation rates of up to 400 mL/hr.m2 were obtained, with pumping head heights reaching up to 1.8 m and a maximum pump functional lifespan of 13 days. The results further suggest that there is a good correlation between the various statistical fits and the experimental data. The developed theoretical thermal-hydraulic model was also found to be in good agreement with the experimental results. A sensitivity analysis of the theoretical and statistical models showed that the statistical models fairs poorly under extrapolation. Additionally, the mechanistic causes of pump failure as well as the effect of heat and pumping head on water pumping performance were identified. Thereafter, the water collection efficiency was established to be 98% on average. Further testing revealed that the pumping performance of larger area or multiple grouped “leaves” are less accurately predicted with the theoretical model than a single “leaf”. In conclusion, the results provide some support that the surface tension driven pump may be used as a water transport system in an artificial photosynthesis project, if the functional lifespan of the pump can be greatly improved. It is recommended that a more rigid hydrophilic material be used in the “leaf” interface and that multiple narrower conduits be used instead of a single larger pipe. Additional future work may include the development of pit-like structures to prevent air spreading throughout the system as well as a simple mechanism for evaporative control. / AFRIKAANSE OPSOMMING: Die doel van hierdie ondersoek is om 'n oppervlakspanning-aangedrewe waterpomp te bou en te toets. Die oppervlakspanning-aangedrewe waterpomp is ‘n passiewe waterpomp wat gebruik maak van ‘n meganisme soortgelyke aan dié van bome om water te pomp. Hierdie ondersoek is by die Departement Meganiese en Megatroniese Ingenieurswese by die Universiteit van Stellenbosch uitgevoer. Vir die ondersoek is 'n uitgebreide literatuurstudie gedoen wat aspekte soos water eienskappe, oppervlakspanning (basiese beginsels, kapillêre kragte, die uitwerking van temperatuur, onsuiwerhede asook benatbaarheid), lugborrelvorming (kernvormingsteorie en die treksterkte van water) en uiteindelik water- en mineraalvervoer in bome (plantstrukture en -meganismes, beperkende faktore, wanpersepsies en die vervoer van organiese stowwe) insluit. Vorige navoring deur plantkundiges, wat die watervervoermeganismes in bome demonstreer, is ook in ag geneem. Die ondersoek het die ontwikkeling van 'n teoretiese termies-hidrouliese model ingesluit, wat gebruik is om die oppervlakspanning-aangedrewe waterpomp se werking te voorspel. Die waterpomp is ook eksperimenteel getoets met die fokus op ontwerpverbetering, pompwerkverrigting, pompwerking, potensiële pompopvoerdrukhoogte sowel as die waterversamelingsvermoë. Die eksperimentele data is statisties ontleed deur middel van meervoudige liniêre regressie. Beide die eksperimentele data en statisties-gegenereerde voorspellings is vergelyk met die teoretiese termies-hidrouliese-model. Die resultate toon dat 'n werkende oppervlakspanning-aangedrewe pomp gebou is. ‘n Verdampingstempo van tot 400 mL/hr.m2, pompopvoerdrukhoogte van tot 1.8m en 'n maksimum funksionele pompleeftyd van 13 dae is bereik. Die resultate dui verder daarop dat daar 'n goeie korrelasie tussen die verskillende statistiese lynpassings en die eksperimentele data is. Die teoretiese termies-hidrouliese-model wat ontwikkel is, toon 'n goeie ooreenkoms met die eksperimentele resultate. 'n Sensitiwiteitsanalise van die teoretiese en statistiese modelle het getoon dat die statistiese modelle swak voorspellings maak as geëkstrapoleerde data gebruik word. Verder is die meganismes wat pompweiering veroorsaak, die effek van hitte asook die effek van pompopvoerdrukhoogte op die pomp se werkverrigting geïdentifiseer. Daarna is die doeltreffendheid van waterversamelingsvermoë vir die waterpomp vasgestel op gemiddeld 98%. Verdere toetse het getoon dat die pompwerkverrigting van groter gegroepeerde "blare" minder akkuraat met die teoretiese model voorspel word as vir 'n enkele "blaar". Ten slotte: Die resultate toon dat die oppervlakspanning-aangedrewe waterpomp as 'n water vervoer stelsel gebruik kan word in 'n kunsmatige fotosinteseprojek, indien die funksionele leeftyd van die pomp verbeter kan word. Dit word aanbeveel dat 'n sterker hidrofiliese materiaal in die "blaar"-koppelvlak gebruik word en dat verskeie nouer leipype gebruik word in plaas van 'n enkele groter pyp. Bykomende toekomstige werk kan die ontwikkeling van put-agtige strukture insluit wat die verspeiding van lug deur die hele stelsel voorkom, sowel as 'n eenvoudige meganisme wat die verdampingstempo beheer.