What is thought to be the first major remediation project in the UK to use soil washing is on schedule for completion in mid-September. The process separates and removes fine grained contaminated material from coarse grained material, which is then cleaned. The washed soil can then be largely re-used as fill on site, offering an economical and environmentally friendly alternative to conventional methods of remediating contaminated land.
The first UK trial has been at Basford gasworks in north Nottingham. The works produced town gas from 1841 until 1972, when the 7.76ha site was largely demolished, leaving just two gas holders and some ancillary buildings. Since then it has been used as a maintenance and distribution point by Transco.
Town gas was produced from coal and oil. It created by-products for local industry including coke, motor benzol, sulphuric acid and ammonium sulphate, but its legacy of contamination included coal tars and liquors, petroleum hydrocarbons and spent oxide.
At Basford, it has also left many geotechnical problems such as loose fill and buried railway lines, structures and tanks. To complicate matters, some of the aggregate used in the concrete for the gasworks is polluted with cyanide, the backfilled original course of the River Lean runs nearby, and there was a small tar factory in the south west corner of the site.
Basford is part of BG Property Div- ision's rolling programme to reclaim former gasworks. It owns between 900 and 1,000 potentially contaminated ex-gasworks sites and spends about £60M every year on remediation. Last year 100 sites were cleaned up.
While previous investigations showed the site did not present a risk to nearby residential areas, they did identify the need for remedial action to protect the underlying Sherwood sandstone aquifer, a source of potable water.
BG decided the best course of action was source removal of contaminants rather than containment. With up to 7m thick deposits of mainly alluvial gravels, with some peat in the south and demolition fill, geology was considered suitable for soil washing and the size of the problem meant this method would probably be cheaper than removal to landfill. About 80% of material put through the washer is cleaned and re-used while just 20% has to be disposed of off site.
The method's principle is that most contamination, particularly organic, forms only a surface coating on soil grains. Fine grained material such as clay or silt, with a high surface area to volume ratio, is potentially more contaminated than a sand or gravel. The process separates fine contaminated particles from the coarse sand and gravel fractions which are cleaned by an attrition scrubber that removes contamination from the surfaces of the larger grains. Absorbent natural organic material like wood and peat is also removed.
While the cleaning processes are not new, the system has not been used at any great scale in the UK and its efficiency has rarely been quantified, so a major element of the project was to optimise process steps and accurately assess effic-iency. A pilot study was carried out, partly to select and scale the most approp- riate unit processes so that an inte- grated plant could be built, and to allow the joint venture of manufacturer Linatex (UK) and operator Heijman Milieutechnik to see if it was commercially viable.
BG com- missioned con- sultant Parkman Environmental to devise and super- vise a suitable investigation and subsequent re- mediation strat- egy (agreed with the Environment Agency) using soil washing. A detailed site investigation, which included 350 trial pits on a 10m by 10m grid and geotechnical and chemical testing of more than 2,000 samples, allowed a site classification of soil and con- tamination to be made. Material was placed in 11 categories (five 'clean' and six 'dirty') based on the soil washing processes required. Some boreholes were sunk into the sandstone to check groundwater contamination, which, according to BG Property Division construction manager Steve Wallace, is not extensive.
This information was used to build a detailed ground model that clearly defined the boundaries of the different material and allowed a selective digging strategy to be formed. This involves main contractor VHE Construction excavating batches of nine cells covering a total of 30m square and, with some degree of material mixing, keeps all the different processes of the plant working at the best possible rate, with no part idle. Optimum efficiency of the whole plant is 50t per hour.
Processing steps are in ascending cost and are designed to reduce the volume of material passed on to the next step. Close control should recover material suitable for re-use with the minimum degree of processing. As the boundaries between clean and contaminated soils are known to a high degree of accuracy, cross-contamination is avoided, wastage minimised and the maximum amount of clean material recovered.
Targets set for the site were to remove 95% of the key contam-inants (PAHs) and between 85% and 90% of phenols and ammonia. Some of these targets have already been beaten.
'The quality of the cleaned products is very high,' says Wallace. 'Possibly too clean,' he jokes, adding that by 'tweaking' the process it should be easier to more accurately clean the soil to agreed level and make the method more economical.
Site works started in May 1997 when the plant was set up, and cleaning started in earnest that August. In all, 150,000m3 of soil will be excavated over the course of the £4.6M contract, planned to take 76 weeks and now expected to finish a couple of weeks earlier in mid-September.
One of the most obvious cost savings over conventional dig and dump is the reduction in vehicle movements from an estimated 18,000 to 5,000, because less contaminated material has to be removed and less clean material brought in. This, says Wallace, also minimises disruption to residents.
Further savings are made by mixing and re-using the cleaned coarse and fine soils as fill. Recovered ash and clinker is screened to remove foreign material and then rescreened to obtain a 10mm to 40mm fraction used in steel manufacture, and fines for making breeze blocks.
But, Wallace says, other factors partly offset these savings. Up front site investigation costs are greater, design input is much higher and the contract period is predicted to be 166% longer than a comparable dig and dump contract. Set-up costs are also very high and economies of scale are important. 'Smaller sites may not have sufficient room to allow processing and may not justify the mobilisation of a soil washing plant.'
However, mobilisation costs should fall once the method is more common in the UK and experience will help engineers decide whether or not the method is suitable at an early stage.
Wallace believes the need for pilot studies remains a major barrier to the introduction of this and many other remediation technologies to the UK. Proving the viability of a method has some risk attached, he says, as there is a a danger of investment in pilot studies being lost if it proves to be unsuitable.