The evening meeting began with presentations from Headley Greaves of Buxton Lime Industries, Murray Reid of the Transport Research Laboratory and Alex Kidd of the Highways Agency. They were followed by Stephan Jefferis of Environmental Geotechnics, who opened the discussion with a short commentary on swell tests.
The most common stabilisers are lime, cement, PFA and granulated slag. While their use often rapidly improves the soil's properties, there are problems associated with each method.
The presence of sulphur compounds in stabilised soil can cause a wide array of problems, the most common of which are swelling and heaving. A high profi le example is a recently opened section of the A10 Wadesmill Bypass where 35% of the subgrade and capping layers needed to be replaced because of heave from sulphate bearing soils (GE December 04).
The sustainable approach Greaves began with a description of what soil stabilisation is, how it works and why it should be used, with particular emphasis on the environmental and economic benefits of the method.
Greaves said there were numerous reasons for using stabilising techniques. While sustainability and environmentally friendly approaches were not often seen as good economics, stabilising practices were now seen as meeting environmental and financial targets, he said.
The implementation of the Landfi ll Tax in 1996, coupled with spiralling fuel bills, meant the case for soil stabilising could not be stronger. This was refl ected in the rapid increase in its use (Figure 1).
With so many projects demanding the reuse of onsite material, it had become imperative that existing material could be modified into a suitable soil, Greaves said. Soils could be stiffened by mixing with lime, cement, pulverised fuel ash (PFA) or ground granulated blast furnace slag (GGBS).
For large scale projects, most lime stabilising was done with specialist equipment. The most common of these was the rotovator (Figure 2).
This machine had an on-board lime hopper. A computer controlled and monitored the rate of lime spreading and the machine was able to mix lime with a soil layer up to 300mm thick in a single pass. As soon as mixing started, reactions began at molecular level.
The fi rst reaction occurred when the lime started to absorb water (up to 32% of its own weight) and started to give off heat. This heating also drove off more water. Within two hours, modifi cation of the soil properties began.
Modifi tion was the second stage and resulted in a dramatic change in physical properties. The material became less plastic, the soil appeared drier and the soil particles began to break down. Once this occurred, compaction was easier to achieve (Figures 3 and 4).
The fi nal phase was the long-term stabilisation process. This involved a chemical reaction with the lime that resulted in silica and aluminate gels forming in a cementitous reaction.
All of these reactions resulted in material with bulk fill improvement of CBR (5-8%) and capping layer improvement of up to 15% CBR.
If this was not a suffi cient improvement in material properties, cement could also be used. If soil had been pretreated with lime, the use of cement could result in sub-base material with CBR values over 50% higher, greater stiffness, frost resistance and increased trafficability.
Both PFA and GGBS were as effective as cement but took much longer to react with soil. This could be benefi cial, however, as it meant the material could be stockpiled for several weeks and remain workable.
Sulphur compounds in soil stabilisation Murray Reid of the Transport Research Laboratory (TRL) outlined the problems sulphur compounds could cause to stabilised soil and how best to control the associated risks. New tests and protocols were discussed, including guidance on what planning is required.
For all the benefi s that were achieved with lime stabilisation, there was some risk involved. The most critical threat, Reid explained, was the adverse effects caused by the presence of sulphur compounds in the soil.
These could pollute drainage, attack metals and concrete in buried structures and attack soils, resulting in heave in stabilised material.
Heave could occur as a result of naturally occurring pyrite reacting with lime or cement and water to produce ettringite or thaumasite.
Both of these had the potential to expand by up to 60%. This expansion was well documented, from high profi le projects such as the M40 in the late 1980s, and more recently, the A10 Wadesmill Bypass.
The challenge facing engineers, Reid said, was how to best limit the risk.
Guidance for dealing with sulphates in soil was published in HA74/95 Treatment of fill and capping materials using either lime or cement or both, in reaction to the M40 heave.
This was updated to HA74/00, which not only detailed problems that were likely to occur, but also how to test for sulphate correctly.
It also gave guidance on tolerances.
Specifi ations for highway works were included with detailed acceptable limits for acid soluble sulphate and testing methods to be used.
Before 2001, Reid said, standard sulphur tests were slow, difficult to perform and not particularly accurate. Most laboratories realised this and were using a range of automated methods developed in-house.
TRL trialled 18 new tests from which four were chosen as the most suitable, based on speed, cost and accuracy. The new range of test methods was published in 2001 by TRL in the document TRL447 Sulfate specification for structural backfills.
Following on from the new tests, new limiting values were established (Figure 5).
Further work on sulphates in soil was under way, Reid said, with BRE planning to release a revised version of its Special Digest 1 (SD1 Concrete in aggressive ground) (GE January 05). In particular, there were plans to revise the water soluble sulphate thresholds to make allowances for sulphate in groundwater. As a result the limiting values in HA74/00 would require redefinition, he said.
The way forward?
The Highways Agency (HA) has spent a great deal of time and effort providing guidance for soil stabilisation. The agency's Alex Kidd outlined the guidance available in the HA and European codes, and gave his views and observations on the future of soil stabilisation.
While a wide range of documents were available for providing information on sulphates, there was a very high level of contradiction across several topics, he said.
Some texts (The Thaumasite Expert Group Report and HA74) stated that glacial tills were generally free from sulphates, whereas others (BRE SD1 and Ciria 504 Engineering in glacial tills) said tills might contain sulphates and there was high horizontal and vertical variability.
As well as differences in the guidance documents, the distribution of sulphur compounds, namely sulphides and sulphates, was equally inconsistent. This appeared to be the most signifi cant risk when conducting preliminary studies and this was exactly what new guidelines aimed to address, Kidd explained.
In December 2004, Mott MacDonald was appointed to review HA74. Its primary task is to examine the implications of TRL447 and various European standards and to examine the recommended frequency of testing and acceptable limits.
It is hoped this revision will serve to clear up confl icting practices and to provide clearer guidance.
Arup, TRL and BRE have been appointed by the HA to review the structural applications of stabilised soil and the uses of additives in high-sulphur soils.
Discussion Stephan Jefferis of Environmental Geotechnics opened the discussion with a brief presentation on the applicability of swell testing. He said oxidation could only occur to a signifi cant extent when the material was unsaturated - which it would be during dry weather, when most UK stabilisation work was carried out.
However, water was required for ettringite and thaumasite to form.
As a result, heave of lime stabilised material would occur only after a dry period (for oxidation) and at the onset of wet weather (for ettringite and thaumasite formation).
Greaves suggested only a small amount of pyrite needed to be oxidised to cause problems.
Dave Tonks of Edge Consultants asked if the new guidelines reduced the range of materials that could be used and would reduce the reuse of onsite material, which would reduce sustainability. Reid replied there would be scope for an appropriate judgement to be made on a site-bysite basis.
Dennis Higgins of the Cementitious Slag Bakers Association noted that HA74 did not cover the use of slag. He asked if this would be addressed in the review. Kidd said slag was covered in the EU standards and the review was still open to additional input.
Independent consultant Peter Kerns commented that carbonate material should be tested volumetrically and that the optimum temperature for formation of Thaumasite was 5 oC, so testing should be done at that temperature. He suggested some of the new tests still required development.
Ian Laws of BRE said groundwater was not covered in HA74.
He asked if this would be addressed during the review. Reid agreed this was a critical factor that needed to be considered and said it would be addressed in the next revision.
Simon Weilham of Mouchel Parkman asked if there was any effect of vegetation on stabilised soil or vice versa.
Kidd said this had not been looked at but the review period for HA74 was still open and any options would still be considered.