feasibility study ISCO for CHC contaminations

The application of permanganates to oxidize entrapped organic chlorinated hydrocarbons (perchloroethene and trichloroethene, PCE and TCE) was investigated during a feasibility study and funded by the BWPLUS research program. Research work depended on batch, columns and 2D Flume experiments for three different kinds of soil (silt, fine sand mixture, coarse sand mixture). Main scope were the effects and the consumption of permanganate (NOD) by the soil matrix containing natural organic matter (NOM), the change and restoration of the hydraulic conductivity caused by the reaction-specific manganese iron production and the reduction of the NAPL oxidation reaction rate by precipitated manganese iron on the reaction interface.

The higher the amount of fine particles (silt) the higher is the consumed specific mass of permanganate. For sandy material a NOD
(g MnO4- per g TOC) between 0.6 - 0.7 was determined, for silty material NOD was between 1.8 - 2 g MnO4/g TOC of soil. Consumption data are site specific and should be determined in controlled batch tests lasting for about 3 weeks.

The oxidation of soil matrix (reaction of quasi first order) shows a soil matrix specific but low amount of fast-oxidizable organic carbon and a major content of slow-oxidizable organic matter. Site-specific amounts are determined in batch tests. The reaction rate of the oxidation of contaminants is between 10 - 100 times higher than the rate to oxidize the soil matrix. For high initial contaminant concentrations (approx. 12 g NAPL per kg soil) up to 25% of the total permanganate mass was consumed by the soil matrix during long term column experiments whilst the residual emplaced contaminants were completely oxidized.

Depending on soils grain size the hydraulic permeability will be reduced by a factor 2 - 20 by manganese iron precipitation; the smaller the grain size the smaller the reduction of permeability. By the formation of manganese iron layers the reaction rate is significantly reduced for direct interfacial reactions (residual or contaminant pool) by a factor of 10 -100. The reaction rate of the oxidation of solved contaminant was not affected by precipitation processes. Rebound effect occurred every time when permanganate release was stopped until the complete oxidation of the contaminant mass. Therefore an intermittent permanganate release or infiltration is advised for a control of the remediation process and its success as well as to ensure contaminant mass transfer and oxidation. In relation to a hydraulic remediation (reference column experiments) ISCO resulted in a 10 times faster contaminant mass removal.

The complete oxidation of a DNAPL pool under nature like flow and soil conditions (coarse sand, 2D Flume experiment) was not achieved during reasonable time demands due to precipitated manganese layer formation. Mass oxidation was 80% of TCE, 70% of PCE. During field application the risk of bypassing the contaminant source zone using a high concentrated permanganate solution is evident. A density driven sinking of a permanganate solution for a concentration of more than 2 g/l was observed. Using numerical simulation might be used to determine the optional concentration of permanganate solution with a defined density driven flow behavior.

Klaas, Norbert
Braun, Jürgen

Trötschler, Oliver
Hetzer, Steffen

VEGAS

From: 2005-10-01

To: 2006-12-31

Baden-Württemberg-Programm Lebensgrundlage Umwelt und ihre Sicherung (BWPLUS)