Date of Graduation


Document Type



The concept of using prefabricated vertical drains (PVDs) for remediation of contaminated sites with fine-grained soils is examined. The PVD system is used to shorten the drainage path of the groundwater flow and promote subsurface liquid movement expediting the soil flushing process. The use of PVDs in the current state of practice has been limited to soil improvement. The use of PVDs under vacuum conditions is investigated using study soils consisting of 100% sand and 80% sand/20% kaolinite mixture. PVD parameters evaluated included: Well Resistance, Zone of Influence, and Smear Effects. Applicability of the PVD soil flushing was evaluated through pilot-scale contaminant recovery experiments, and development and verification of a prediction model. The PVD Well Resistance indicated linear relationships between PVD Headloss and Extracted Fluid Flowrate, and PVD friction factor (f) and Reynolds Number (R{dollar}\\sb{lcub}\\rm e{rcub}).{dollar} For the MD-88 PVD the slope of the best-fit line ranged between 2 to 3 for PVD widths ranging 5.0 to 10.0 cm. The 10 cm wide MD-88 showed f = 0.2 at R{dollar}\\sb{lcub}\\rm e{rcub}{dollar} of 10{dollar}\\sp{lcub}-3{rcub}{dollar} compared with the MD-7404 showing f = 2 for R{dollar}\\sb{lcub}\\rm e{rcub}{dollar} 10{dollar}\\sp{lcub}-4{rcub}{dollar} to 10{dollar}\\sp{lcub}-3{rcub}.{dollar} For the 8020 soil, the ratio of the zone of influence divided by the PVD equivalent diameter (circumference/{dollar}\\Pi){dollar} ranged between: 3.2 to 8.4 for no mandrel disturbance; 6.3 to 10.4 for pushed mandrel; and 4.2 to 11.6 for the driven mandrel. In the case of 100% sand, no "well" defined PVD zone of influence was measured under the range of test vacuums. The zone of influence in 80/20 soil did not show a corresponding increase in aerial extent as the vacuum pressure was increased beyond a critical depth. Pilot-scale tracer flushing tests demonstrated that longer flushing times were necessary to achieve consistent clean-up levels as the soil dry unit weights increased from 1.57 to 1.67 g/cm{dollar}\\sp3.{dollar} For in situ hydraulic conductivities (k) less than 10{dollar}\\sp{lcub}-2{rcub}{dollar} cm/s, an 80% reduction in tracer concentration was reached after 40 to 50 minutes. For in situ k of 10{dollar}\\sp{lcub}-4{rcub}{dollar} cm/s, the 80% reduction required 20 to 60 hours. For the 80/20 soil, doubling the initial concentration doubled the time required to reach 80% removal. The Model accurately predicted clean-up times consistent with pilot-scale results.