Removal of oxyfluorfen from spiked soils using electrokinetic soil flushing with linear rows of electrodes

• The herbicide oxyfluorfen can be transported by electrokinetic processes in EKSF.
• The dragging of oxyfluorfen to cathode wells is the main EK mechanism in EKSF.
• EKSF attains a 26.8% improvement in the removal of oxyfluorfen after 34 days.
• EKSF yields a soil pH gradient as a consequence of the acidic and basic fronts.
• The removal of a given herbicide by EKSF strongly depends on its solubility in water.

This study focuses on the evaluation of the electrokinetic soil flushing (EKSF) strategy to remediate soil following a simulated spill of the herbicide oxyfluorfen. EKSF is attained by placing (in the soil mockup) two rows of electrodes of different polarity facing each other. The results are compared with those obtained in a reference experiment in which the same spill was simulated and no remediation actions were taken. In addition to the daily monitoring of the most important parameters in the flows, after the remediation test, a post-mortem analysis was performed to obtain a 3-D map of the pollutant distribution. Those results demonstrate that despite the hydrophobic character of oxyfluorfen, it can be efficiently transported by EKSF. Hence, after 34 days of treatment, a 26.8% improvement in the removal of oxyfluorfen was achieved (explained in terms of the effect of the electric field on the pollutant) compared with the reference experiment in which only volatilization can explain the removal of the herbicide. Comparison of the removal of oxyfluorfen by EKSF with that of 2,4-D (studied in a previous study) demonstrates that comparable dragging to the cathode and volatilization are obtained. However, the lower efficiency of the transport of oxyfluorfen by gravity fluxes and electromigration (explained because it is contained as micelles) yielded worse performance of EKSF for this water-insoluble pesticide and hence less efficient remediation. This contradictory result reveals the importance of tests at large-scale facilities such as that used in this work to predict the performance of real systems in future full-scale applications.

Figure optionsDownload as PowerPoint slide