Oprensning af PCE ved kemisk oxidation med kaliumpermanganat

Remediation of PCE by Chemical Oxidation Using Potassium Permanganate

2 Summary and Conclusions

This project is financed by the Technology Development Programme for Soil and Groundwater Contamination, Danish EPA. The county of Funen was the EPA representative in this project.

Prior to the project an in-situ remediation of tetrachloroethylene (PCE) in the unsaturated zone was accomplished at a dry cleaning facility in the city of Odense. The remediation technique used was thermally assisted soil vapour extraction combined with groundwater pumping for contaminant plume control. The unsaturated zone was successfully remediated within one year. However, the saturated zone was still contaminated and the groundwater pumping did not indicate any significant contaminant removal. Remediation of the saturated zone was therefore initiated by in-situ chemical oxidation using potassium permanganate.

Purpose of the Project
The purpose of the remediation activity was to remediate the groundwater contamination in order to eliminate the environmental impact of volatile contaminant transport to the residential areas above the source area. The objective of the Danish EPA technology innovation program has been to collect documentation regarding the effectiveness of the technology for remediating chlorinated solvents in the saturated zone.

Geology and Hydrogeology
The upper quaternary sequence consists of 1-2 meters of fill material, approximately 8 meters of fluvio glacial sand with parts of silty areas and below this, glacial till. The groundwater aquifer is unconfined with a groundwater level 6-7 meters below ground level.

Level of Contamination
Before the remediation activity, concentrations of up to 2,000 µ g PCE/l were measured in the groundwater.

Remediation Strategy
The strategy of the remediation activity was to inject liquid potassium permanganate into the groundwater aquifer at groundwater level upstream the source area. To improve the horizontal distribution of oxidant, groundwater pumping was established downstream the contaminated area in order to pull the oxidant in this direction. A total of 60 m³ 5% potassium permanganate solution was injected.

Monitoring
To evaluate the efficiency of the remediation and the distribution of potassium permanganate in the groundwater aquifer, water samples were analysed for chlorinated solvents and their degradation products, inorganic macro ions, trace metals, bromide tracer and redox conditions. Furthermore, the groundwater colour and changes in colour were noted.

Injection Strategy and Mixing of Chemicals
Liquid solution of potassium permanganate was injected under pressure by using perforated iron lances. This injection method proved successful at the site. A 5% solution of potassium permanganate was used, which is slightly higher than the maximum 4% recommended abroad. However,. using the 5% solution did not cause any problems during injection. In more recent projects, COWI has chosen to use a 2.5% solution, which in bench scale testing is assessed to be sufficient to degrade PCE. Experience from abroad also indicates this.

In this project, all 60 m³ of potassium permanganate were injected as a single batch, but recent experience from abroad suggests the injection to be pulsed injections, preferably with recirculation of pumped up groundwater containing excess oxidant. Reinfiltration might, however, present operational difficulties.

Tracer Addition
The added bromide tracer was assessed to be suitable to evaluate both the horizontal and vertical influence of the oxidant. Together with conductivity measurements and colour observations it has been possible to give a good description of the horizontal and vertical distribution of the oxidant. Without the use of a tracer, it can be difficult to follow the distribution of the oxidant and the effectiveness of the remediation. Therefore, it is recommended to add a tracer to the oxidant solution in future projects.

Distribution of Potassium Permanganate
The horizontal distribution was observed earlier than expected. Only a few days after the injection potassium permanganate had spread more than 15 meters downstream the injection area, which was twice as fast as the estimated oxidant transportation time. The reason for the fast spreading of the oxidant was the downstream groundwater pumping combined with the upstream injection, which increases the gradient. Furthermore, the flow in the more coarse parts of the sandy aquifer was increased.

The distribution of oxidant in the upstream direction was limited due to the downstream groundwater pumping and the injection jets pointing in downstream direction only.

The potassium permanganate was observed spread in both the top and bottom of the aquifer. The vertical distribution occurred primarily in the sandy deposits and to a less degree in the silty parts. The higher density of the injected oxidant solution (app. 1.07 kg/l) has probably had a sinking effect as well.

Remediation Results
The remediation attempt showed a distinctive decrease in the PCE concentrations in both the source area and in the downstream direction. In the source area, the present PCE concentration is less than 10 µg/l, and it is assessed that the remediation has resulted in an almost complete clean up both horizontally and vertically.

Downstream, the dry cleaning facility remediation was a success as well, but remaining contamination is, however, still observed in the silty and fine grained deposits.

Water samples taken in the upper part of the saturated zone show PCE concentrations less than 20 µ g/l in the affected area, which consequently do not constitute any risk regarding the indoor climate in the above apartments. The aim of the project is therefore fulfilled.

Release of Trace Metals
An increased level of chromium was found in the groundwater after the potassium permanganate injection. The increase corresponds to 100 times the normal background level or 5.5 times the regulatory value for chromium in potable groundwater. No other trace metals exceed the regulatory values, but a slight increase in aluminium concentrations was seen due to the injection.

The explanation of the increased chromium concentrations can typically be mobilised metal concentrations due to the oxidation, or impurities in the potassium permanganate product. In this case, it is assessed that impurities caused some of the increase in chromium levels.

It is recommended that the question of trace metal should to be investigated in future projects, also including a study of the impurities of the potassium permanganate product.

Environmental Assessment
In the project, it gave rise to concern whether the potassium permanganate could spread to the nearby stream (Odense Å), which is situated only 250 meters downstream the site. Because of its strong oxidizing capacity, potassium permanganate is harmful to aquatic organisms, and the regulatory value for water recipients is 1 µg/l.

Therefore a risk assessment was made prior to the remediation actions. The assessment indicated that the maximum horizontal distribution of potassium permanganate was 70 meters from the injection point. In short, it was assessed that it was most unlikely that potassium permanganate would reach Odense Å.

The actual distribution of purple coloured groundwater was observed to be app. 40 meters from the injection point, and the strongest purple colour (highest concentration of potassium permanganate) was limited to app. 25 meters from the injection point. The actual observations verify that the injection did not affect the recipient.

Assessment of Applications in Denmark
This project illustrates that remediation using KMnO₄ is an effective supplementary remediation operation in the saturated zone and can be more cost effective and faster than traditional pump and treat activities. The accessible part of the contamination is shown to be degraded very effectively.

Compared to other oxidants such as Fenton‘s Reagent, KMnO₄ will exist and work actively in the sediment for many months. The project illustrated evidence of potassium permanganate by coloured groundwater 10½ months after the injection. This suggests that the method can be used in sediments with lower permeability, e.g. in glacial till deposits. In these sediments, it will be more difficult to obtain a complete clean up, but the most accessible part of the contamination situated in fissures and sand lenses will be oxidised and thereby lower the aqueous equilibrium concentrations in the groundwater.

The overall conclusion from this project is that the use of potassium permanganate as remediation measure in the saturated zone ought to be a serious alternative to other remediation techniques such as pump and treat and air-sparging. It is expected that more projects using KMnO₄ will emerge in Denmark in the future, as seen in USA.