A residential lot impacted by spills from a leaking light heating oil tank was treated with a combination of chemical oxidation and bioremediation to avoid technically challenging excavation. The tank left emptied in the ground was used for slow infiltration of the remediation additives to the low permeability, clayey soil. First, hydrogen peroxide and citrate chelate was added for Fenton’s reaction–based chemical oxidation, resulting in a ca. 50% reduction from the initial 25,000 mg/kg average oil concentration in the soil below the tank. Part of this was likely achieved through mobilization of oily soil into the tank, which was beneficial in regards to the following biological treatment. By first adding live bacteria in a soil inoculum, and then oxygen and nutrients in different forms, an approximately 90% average reduction was achieved. To further enhance the effect, methyl-β-cyclodextrin surfactant (CD) was added, resulting finally in a 98% reduction from the initial average level. The applicability of the surfactant was based on laboratory-scale tests demonstrating that CD promoted oil degradation and, unlike pine soap, was not utilized by the bacteria as a carbon source, and thus inhibiting degradation of oils regardless of the positive effect on biological activity. The effect of CD on water solubility for different hydrocarbon fractions was tested to serve as the basis for risk assessment requirements for authorizing the use of the surfactant at the site.

Fenton’s reaction-based chemical oxidation is in principle a method that can be utilized for all organic fuel residues thus making it a potential all-purpose, multi-contaminant, in situ application for cases in which storage and distribution of different types of fuels have resulted in contamination of soil or groundwater. Since peroxide breakdown reactions are also expected to lead to a physical transport of the target compound, this secondary physical removal, or rebound concentrations related to it, is prone to be affected by the chemical properties of the target compound. Also, since soil conditions are seldom optimal for Fenton’s reaction, the balance between chemical oxidation and transport may vary. In this study, it was found that, with a high enough hydrogen peroxide concentration (5 M), methyl tert-butyl ether–spiked groundwater could be treated even under suboptimal conditions for chemical mineralization. In these cases, volatilization was not only contributing to the total removal but also leading to rebound effects similar to those associated with air sparging techniques. Likewise for diesel, temporal transport from soil to the aqueous phase was found to lead to false positives that outweighed the actual remediation effect through chemical mineralization.

Purpose Soil restoration is still mainly carried out ex situ by excavating and replacing the contaminated soil. In situ remediation would reduce the costs of soil transportation and this way, the problem is not merely transferred elsewhere. The present study introduces a field case where the aged, oilcontaminated soil in a former fuel station in Finland was treated in situ sequentially with different methods. Materials and methods Several approaches, including soil vapor extraction and biostimulation with electrokinetic pumping, were performed in the field. After these treatments, the dense original portion of the soil beneath the gasoline pump location, ca 100 m 3 , was still contaminated with petroleum-derived volatile organic compounds (VOCs), with concentrations of nearly 10,000 mg kg −1 measured at some hotspots. After a period of electroosmotic water circulation, the electrical field (0.5 V cm −1 , DC) was kept connected for 6 months without addition of water, leading to dewatering and warming of the soil. Results and discussion In contrast to the situation with the original wet soil, VOCs, in lab conditions, were found to volatilize very efficiently from the dewatered soil. When the soil vapor extraction treatment was renewed using perforated tubing installed horizontally at ca 1 m depth in the dewatered soil at the contaminated site, the treatment was efficient and the soil was decontaminated in 5 months. The final VOC concentrations were on average 190 mg kg −1 (n = 13) with the highest value of 700 mg kg −1 at one hotspot. After a risk evaluation, the site was concluded to be sufficiently clean for industrial use. Conclusions Since with many former fuel stations, the contamination consists of both volatile fractions that are difficult to degrade by biological means and heavier compounds for which biostimulation is often suitable, a combination of different methods may be worth pursuing.

For more than 20 years we have tested various approaches for enhancement of bioremediation of sites polluted by organic contaminants. Through collaboration with contractors and site owners, more than 10 actual sites presenting typical problems have been targets for testing and optimization, first by laboratory modeling, and then by applying lab experiences in application scale. Samples from the sites were used in controlled laboratory conditions to build micro-and mesocosm-setups in which biological, physical, and chemical treatments were tested and combined, with the main goal of achieving optimal biostimulation and contaminant degradation. As soon as lab results were available, these were utilized for in situ field purposes. Lab and field tests were run in parallel, so that each new challenge in the field treatment generated modifications in the laboratory testing, and each new full scale treatment method was preceded by laboratory modeling. Successful bioremediation was achieved in most of the target cases. Lab testing also created the knowledge when not to use bioremediation, and this can be regarded as one of the utilities of our results. Fresh oil spills resulting from accidents is a new research topic. While old contaminated sites often can rely on an adapted indigenous microbial community, new spill sites may be less responsive to mere biostimulation, and therefore more active treatment measures may be required. The results from this oil spill simulation performed at a lysimeter field are currently being gathered.