| Front page | | Contents | | Previous | | Next |

Afprøvning af jordvask

Summary and conclusions

The present report evaluates the results from a full-scale project concerning cleaning of contaminated soil in a soil washing plant. K.K. Miljøteknik AS rented the plant for cleaning of soil that was stored in Rødby. The soil washing plant was rented from the German Company Werner Frantzen.

Six trials with cleaning of soil were carried through, and the last trial was completed with an intensive soil sampling, due to an agreement of financial support from the Danish Environmental Protection agency. Erik K. Jørgensen AS was responsible for the sampling and the following report. Unfortunately the German Company went bankrupt after the sixth trial, and all the following trials with the plant were cancelled.

The principles of soil washing are described in chapter two, and very briefly summarised, the soil washing technique is by mechanical influence to fractionate the soil into different dimensions dependent on the size of the soil granulate. By means of screening, washing in water drums, separating in hydro cyclones or separating by screw or settlement the soil is fractionated into six different fractions of size. The large fractions (as stones, fragments, gravel and sand) are expected to be recycled to the construction industry after soil wash. The fine fractions and the organic fractions have to be either cleaned or deposited on a dumpsite. The water in the soil washing plant is recycled and should not contain any contaminants, because soil washing primarily is intended for cleaning of soil contaminated with heavy metals or fuel oil components.

All in all 3600 tonnes of soil collected from five different contaminated sites were cleaned at the soil washing plant. The type and concentration of the contaminants and the historical background for the sites were all different from each other. The collected soil types chosen, were two common fill layers from the City of Copenhagen containing heavy fuel and lead, soil from a former factory for paint and lacquer preparation, soil from a former battery factory and soil from a foundry.

An important part of the project was to make a proposal as to what kind of analyses will be necessary to pre-characterise the soil to be able to evaluate the cleaning potential of the soil to soil wash treatment. The quantity of the fine fraction appeared to have a great influence on how much of the cleaned soil, that ended up in fractions to be recycled. The best way to pre-characterise the fine fraction of the soils was determined to be by a particle size distribution combined with an analysis for the content of natural organic material.

Cleaning of 3600 tonnes of soil resulted in 680 tonnes of soil cleaned to a satisfactory level. If the cleaning of soil should be considered a success, the concentrations of the contaminants in the fractions to be recycled should be lower than the level permitted when depositing the soil at a landfill or a similar place. At the same time the part of the soil in the fractions to be recycled should be big enough to guarantee the economy in the project. In the last three trials of cleaning soil the part of the fractions to be recycled was as high as 65 %.

In the cleaning of soil containing free lead several problems arose. One serious problem showed to be due to the act, that the free lead was separated together with the fractions of the same size as the free lead. If the plant should be able to handle this kind of contamination, it would be necessary to incorporate a unit to sort the soil according to the specific gravity. This kind of unit was not accessible at the actual plant, but it would be possible to incorporate.

Another problem by cleaning especially fills layers were the lump of slag in the soil that was separated together with the fractions to be recycled. The following analysis of the fraction of stones showed up to contain both heavy metals and tar components instead of being clean as expected. Slag is normally not a problem to reuse beneath roads or other building projects, and the fractions to be recycled from the soil wash containing slag should also be possible to use. It is therefore necessary to change the control analyses to another type of analyses, instead of analyses for total hydrocarbon compounds or heavy metals, analyses to control the leaching from the soil would be better.

In several trials the cleaned soil ended up containing more contamination than before soil wash. This is expected to be due to the inhomogeneous distribution of contamination in the soil before cleaning compared to the very homogeneous soil after cleaning where the contamination also is more even. This has given some difficulties when evaluating the results of this project, but generally it is chosen to ignore this complication while it is estimated, that no contamination can be added to the soil from a soil wash plant.

There was no significant difference between the results by cleaning the soil for heavy metals or cleaning the soil for heavy fuel by soil wash. Both kinds of contaminants are evaluated to be very suitable for treatment in a soil washing plant.

It was tried to evaluate if any contamination has moved from the bigger (stone) fractions to the organic or fine fractions by comparing the results of the analyses before and after cleaning in the soil wash plant. While only one trial was carried through with intensive sampling it is not possible to make any clear conclusion on this hypothesis.

The consumption of resources is illustrated in chapter 4. By cleaning 3600 tonnes of soil the consumption of oil was 12.000 litres and the consumption of water was 1500 m³. This corresponds to app. 3,3-litre oil/tonne and 0,42-m³ water/tonne cleaned soil with an expense on app. 25 Kr/tonne. This is evaluated to be an acceptable cost level.

In the light of the expected costs for depositing the left over fractions (fine and the organic fractions) it is estimated, that the maximum cost has to be between 300 and 500 Kr/tonne soil if it should be economical attractive to clean the soil by soil wash. K.K. Miljøteknik AS had paid the German Company Werner Frantzen 300 Kr/tonne for cleaning the soil.

In the light of the finished trials, soil wash is evaluated to be a potential relevant way to clean soil in Denmark. It is gathered though to be necessary to improve the results of the soil wash for instance by adding other kind of units to treatment of special types of contaminants.