Formatted Title
An Adaptive Approach to Implementing an ISCO Soil Mixing Project at a Former Oil Depot in Sweden
Background/Objectives
A former oil depot located on a man-made peninsula in a large inland lake in western Sweden was remediated using a combination of excavation and in situ chemical oxidation (ISCO). The remediation effort targeted four areas with varying concentrations of petroleum hydrocarbons in soil and groundwater. Surficial soil was excavated down to the groundwater table at 1 to 1.5 meters below ground surface and a blend of Klozur® SP (sodium persulfate) activated with hydrated lime was applied to the excavation base and blended into the saturated zone using an excavator with a mixing attachment. An adaptive approach was employed, where ongoing geochemical monitoring of treated cells guided the next steps. This allowed for timely validation of the planned dosing assumptions, allowing for adjustments during the course of the field installation.
Approach/Activities
A total of 96 tonnes of Klozur® SP and 33 tonnes of hydrated lime was estimated to treat the four target areas at the site (total of 1,100 m3 of soil). The planned Klozur® SP dosages ranged from 2.5 to 8.4% by soil mass, calculated based on average baseline concentrations ranging from 660 to 3,700 mg/kg TPH. Maximum TPH concentrations were measured at 11,000 mg/kg in one isolated area, which was excavated and sent for off-site disposal. Each soil mixing area was subdivided into smaller treatment cells, adjusted based on the planned dosage. Following the completion of an area, groundwater was analyzed for pH, ORP and electric conductivity to evaluate distribution and whether conditions appeared favorable for continued treatment. Soil samples were also collected and screened with a PID meter to get a rough estimate of treatment progress. Select groundwater samples were sent out for direct persulfate analysis. This combined data set was then used to assess the need for remixing a cell, making pH modifications / add more Klozur® SP or deeming a cell complete.
Results/Lessons Learned
Following the initial reagent soil mixing application, pH measurements ranged from highly elevated (>11) to acidic (<6) only days after the application. Low pH measurements appeared to align with a rapid persulfate consumption and low ORP values, indicating that the persulfate was mostly spent. Where this also coincided with low PID measurements, no additional reagent application was deemed necessary. Other cells still showed PID values above 400 ppm, prompting additional reagent applications.
Preliminary results from post application confirmatory soil and groundwater sampling to a large extent confirmed the earlier indications based on PID measurements, with treatment efficiencies generally ranging from 87 to 95% reduction in soil TPH relative average baseline concentrations and between 93 to 97% reduction relative maximum baseline values.
One cell measured significantly higher PID measurements during the confirmatory sampling relative preliminary screening values, coupled with some TPH fractions remaining above guidelines. This coincided with this area having the largest pre-treatment variability in TPH measurements (average of 966 mg/kg and maximum of >5,000 mg/kg TPH at baseline). A lesson learned may therefore be to tighten the sampling grid and increase safety factors in areas with larger variability.