Summary and conclusions, Miljøstyrelsen

Passiv ventilation til fjernelse af PCE fra den umættede zone - Hovedrapport

Summary and conclusions

The detection frequency of chlorinated solvents in the Danish ground water is still increasing, and consequently there is a growing interest to investigate the effectiveness of new as well as existing remediation techniques to reduce the content of these components. Several pilot tests of the new remediation technique passive ventilation (PV) have shown promising results. Originally, this method was developed in USA, and here it was tested in a number of different types of pollution, using different well configurations. To document the technique under Danish conditions, the technique is now being tested at 4 localities, all former dry-cleaning sites polluted with PCE. The project is a co-operation between the counties of Storstrøm, Frederiksborg and Ribe under the Danish EPA’s technology demonstration programme). At all 4 localities, a sandy and highly permeable unsaturated zone polluted with PCE was found under a cover layer of clay. This permeable and confined strata resulted in an extensive and wide spread soil gas pollution.

Using PV, the natural pressure gradients between the atmosphere and the unsaturated zone are utilised to force the polluted soil gas to the ground surface through wells screened across the unsaturated zone. The method can be used to reduce volatile components, i.e. PCE, and other chlorinated compounds, more volatile fractions of petrol products, including MTBE, as well as methane from waste dumps. Alternatively, it will be possible to direct the airflow down wards in wells, which allows adding atmospheric air containing oxygen into the unsaturated zone. Hereby, the aerobic micro-biological degradation of oil components is stimulated, and the method is called "passive bio-venting".

In Denmark, the variations in the atmospheric pressure are primarily caused by the passage of weather systems with high and low pressures, and secondary by variations in the temperature of the atmosphere. Generally, there are only small variations in the atmospheric pressure. However, these variations are spread over extensive areas and contains a huge amount of energy. The size of the vertical pressure gradient in the subsurface, depends to a certain extent on the thickness and permeability of the overlying soil, and this is the reason why the method is dependent on a low permeable layer of clay near the surface. The pressure gradient across the low permeability layer , the permeability and thickness of the unsaturated zone to be ventilated, are the controlling parameters for the size of airflow to be achieved. The theory and controlling mathematical equations as well as relevant physical parameters behind the PV technique are described in this report. It is further shown step by step how these parameters can be estimated from pilot tests. Furthermore, it is shown how the average airflow from a passive ventilating well can be calculated from a historical time series of atmospheric pressure and physical parameters from pilot tests.

A proto type system for PV has been developed, consisting of a well screened across the unsaturated zone, an GAC-unit (Granular Activated Carbon) for cleaning of the soil gas before discharge, and a one-way valve allowing only discharge of soil gas from the well. The pressure loss of the individual components has been calculated and measured in a laboratory, and subsequent measurements at the individual sites have documented that the pressure loss through the total system is very low (< 0.5 mBar), and generally only contributes to an insignificant reduction of the natural airflow. The developed GAC-unit can restrain approx. 450 g PCE, before a break through can be detected. All components applied so far have proven to be durable, and it is assessed that one yearly inspection should be sufficient. However, in case of very high concentrations of PCE in the soil gas discharged, inspection and exchange of the GAC-unit every 6 months can be necessary.

The average airflow over a year from a PV well varies from 0.2 – 1.1 m³/h. The number of discharge periods is approximately 180/year, with an average duration of 13-25 hours. The maximum duration of a discharge period is approx. 4.5 days, and the highest peak airflow registered being approximately 32 m³/h. There is a continuous discharge of soil gas 40-50% of the time, equivalent to a statistically positive differential pressure in 50% of the time. The maximum differential pressure measured is approx. +/- 11 mBar, and was registered during an extremely rapid fall in the atmospheric pressure. The largest flows are registered at the locality in Askov, and the total airflow from the system’s 9 filters is approx. 0.2 mio. m³/year, corresponding to an average flow of 23.7 m³/h. A total peak flow for the system of approximately 250 m³/h, which equals the amount for a typical active SVE-system driven by electrical vacuum pumps, was registered during an extremely rapid fall in the atmospheric pressure. The total amount of soil gas discharged at the individual sites over 2 years indicates, that the pore volume turnover rate for the unsaturated zone is between 25 and 100 times.

At the individual localities, the start concentrations of PCE in the soil gas air have shown an average of 100 – 300 mg PCE/m³. During the 18-24 months of operation, there has been a reduction in the average concentration of 50-85%, but with great variations between the individual filters. Thus some of the filters show reductions up till 96%. On an average, the concentrations in the individual localities have been reduced to 30 – 120 mg PCE/m³. The lowest level achieved in Askov, where the largest flow, but also the lowest pore volume turnover rate was measured. The development over time of the level of concentration is exponential. Thus, the average half-life period of the decrease in the concentrations for the Askov locality can be calculated to be 7 months, whereas the half-life periods for the other localities are estimated at approximately 28 months. The horisontal extent of areas with high PCE concentrations (>50-100 mg PCE/m³) has been significantly reduced, and this indicates a clear effect of the operation. Areas with low initial discharge concentrations show only small absolute reductions in the concentrations. In a few filters there is, however, increasing concentrations towards the end of the measurement period. This is due to a rising groundwater level and thereby a reduced thickness of the unsaturated zone, and as a consequence of this, a less or total lack of airflow.

Over the two years of operation at the 4 localities, the total removal of PCE is 2-3, 5, 8, and 50-60 kg, respectively. The highest rate of removal is seen in Askov. To this shall be added a small amount of degradation products from PCE, corresponding to approx. 1-5% of the amount of PCE removed. After 2-4 years of operation, the removal rate/year for the systems is estimated at 0.1 – 1 kg PCE/year, and calculated from the existing data, this roughly corresponds to the estimated seepage to the unsaturated zone. Thus, it can be concluded that the existing systems in the long term will be able to retain the present reduced level of concentration, and potentially further reduce the concentrations, and hereby reduce the mass flux to the ground water considerably.

At the locality in Askov, it is estimated that the mass flux to the ground water has been reduced during the 2 years of operation in that the concentration of PCE in the ground water directly below one of the passive ventilating wells has decreased from 300 µg/l till 20 µg/l. This is assessed to be in good coherence with the fact that more than 50 kg PCE has been removed from the unsaturated zone. A similar tendency in the ground water has been observed at another project locality.

At the locality in Fakse, measurements have been made in the limestone, through two wells. The average airflow measured is of the same size and with the same discharge pattern as the filters in the sand layer at the site. One of the wells has been installed with a mini 12V vacuum pump, driven by a system of solar cells and a small windmill on the ground. The average flow/year has been approx. 1 m³/h, and from this well and an additional well, a total amount of approx. 0.25 kg PCE have been removed. This pump has yielded 5 times the amount discharged by passive flow, and shows that it is possible in a simple way to increase the natural flow by means of electrical energy from other types of renewable energy sources. The system has been very stabile, but the price (DKK 65,000) for the system is, however, relatively high compared to the effect achieved. The relative contribution of the solar cells is 80% of the total effect, and thus it is assessed that potential future systems should be driven by solar cells directly connected to the pump in order to avoid unnecessary electronic installations, and at the same time obtain considerable savings.

An expected "radius of influence", corresponding to a well distance of 15 m, was applied in the design at all the project localities. The actual distance applied between the individual wells was, however, from 7 – 20 m, the largest distance in Askov and the smallest in Fakse. In all localities, the concentrations are found to be decreasing over time, but there are no simple coherence between the well distances or the pore volume turnover rate and the effect achieved. If the "radius of influence " is calculated as the horisontal distance, from which a PV well pulls in a volume of soil gas corresponding to an average discharge event, the "radius of influence" can be estimated at 1 m (Fakse) to 4 m (Askov). Using the same method, it has been calculated that one PV well – during the longest discharge event (by volume) – has pulled in air from between 3 m (Fakse) to 10 m (Askov). Generally, the tracer tests indicate a larger transport distance that the calculations above, which is probably due to a certain heterogeneity in the vertical permeability of the unsaturated zone.

By design of future systems, a preliminary phase using active ventilation should be carried out in case the start concentrations are higher than approx. 50-100 mg PCE/m³. Hereby the potentially mass accumulated in the unsaturated zone is considerably reduced. Then, when the removal rate has stabilised, normally after a few weeks or months, the active ventilation is changed to passive ventilation. For sites with a relatively high natural airflow (>2 m³/h), a well distance of 15 m is recommended, but in case of a natural airflow between 0.5 and 2 m³/h, a well distance of 10 m will be adequate. In case the natural airflow is extremely low, a well distance of down to 5 m will be acceptable (<0.5 m³/h). Generally, it is recommended to place the PV wells in the source area of the overlying low-permeable material, in order to capture the PCE close to where it is transported into the unsaturated zone.

The total construction cost for a standard system with 5-6 PV wells, similar to the system implemented in the localities in Allerød, is approx. 200,000 DKK per site, excl. V.A.T. Dependent on the exact configuration, a PV well to 20 m below surface, incl. GAC-unit, will amount to approx. 35-40,000 DKK, excl. V.A.T. In case there are existing wells, which can be converted into PV, the expenses will be limited to a well, including GAC-unit, which is estimated at approx. 15-20,000, excl. V.A.T. For the system in Askov with discharge through the roof, no GAC-units and incl. 6 wells, the total costs were 190,000 DKK, excl. V.A.T., and thus not much different in price from the system designed with GAC-units. For the PV system (excl. The active system driven by solar cells, etc.), implemented in Fakse, with 8 filters in the sand layer and 2 in the limestone, the total costs were 260,000 DKK, excl. V.A.T. A pilot test including PV can be carried out for the amount of 45-90,000 DKK, excl. V.A.T., dependent on the level of detail, and whether there is an existing suitable well on the site. The design of a standard PV system can be carried out for an amount of 20-40,000 DKK, excl. V.A.T.

The operation costs for the standard system including GAC-units are relatively low, as it will be sufficient with only one inspection a year. In connection with this inspection, measurements of the present concentrations in each filter can be made, and GAC-units with break through can be replaced. This yearly visit, incl. change of 6 GAC-units, and a short report can be carried out for an amount of approx. 45,000 DKK, excl. V.A.T., dependent on the amount of GAC-units to be changed. If the concentration measurements are excluded, the operation costs can be reduced to approx. 20-35,000 DKK/year, excl. V.A.T.