Treatment of very soft clays via expansive geopolymer injection
Figure 1 Area of Subsidence
Background
Geobear was requested to stabilise an area of ground at a site in Scotland, covering approximately 600 m2, which was suffering from subsidence (Figure 1). The subsidence was thought to be associated with water leakage from a deeply buried sewer at the site location. The ground movement formed a shallow depression which was up to 640 mm deep compared to surrounding levels. Figure 2 shows a digital model of depression.
Site Investigation
The general area around the depression was investigated and found to comprise loose granular made ground to a depth of around 2.0 m, which was underlain by very soft/soft becoming firm silty clay to a depth of around 6.0m. Within the centre of the depression, the natural soils were found to be even weaker and characterised by very soft silty clay. In some locations, the Dynamic Penetration (DP) testing recorded 0 blows for 100 mm penetration with the rods sinking under their own weight.
Figure 2 Digital model of subsidence
A ground model was designed to show the variation of undrained shear strength with depth within and outside the subsidence zone for design purposes. The model was built by correlating the DP blow counts with Standard Penetration Test (SPT) and triaxial cell test results from the undisturbed samples collected.
Figure 3 shows the variation in undrained shear strength across the site, with the centre of the depression represented by test location BH3 (DP2/DP3), where the shear strength is noticeably lower.
Treatment Design
Figure 3 Variation in shear strength prior to treatment
The process for the treatment of soft clays is based on fracturing the clays and permeation of geopolymer along with the fractures. The pressure required to fracture the clay can be determined using undrained shear strength and overburden pressure. The length of the fracture can be estimated from fracture toughness and lift/expansion pressure of the geopolymer. The final density of the geopolymer in the cracks controls the strength of the hardened geopolymer. There is a relationship between the strength of the hardened geopolymer, the mass injected and the relative improvement in shear strength of the soft clay.
For this project, clay treatment injection depths of 1.75m and 3.75m were selected as representing the zone where the most improvements was required. Injection locations were set out at 1.0 m centres on the surface, on the basis of the calculated hydraulic fracture length in the clay. The mass of the geopolymer to be injected was calculated to be 22.5 kg per injection point per injection level (1.75m and 3.75m). Initial modelling indicated an improvement of shear strength of around 600% should be achievable based on the design.
Results
Figure 4 Results of geopolymer treatment
The measured improvement in shear strength was determined via post-treatment DP testing, with blow counts converted to undrained cohesion via the correlations established prior to testing in the site investigation phase, the results are shown on Figure 4.
The results showed a substantial improvement in shear strength, from near zero to 18 kN/m2 in the weakest soil zone. The average improvement across the treatment depth (1.75m to 3.75m) was 780%, which was even higher than predicted. The results confirmed that injection of expansive geopolymer was effective to about a depth of 1.0m below the specified injection level.
Conclusion
The results of the project confirmed that very soft clays can be effectively treated by expansive geopolymer injection with improvement in shear strength in excess of 600% achievable. The ground at the site in question was effectively stabilised and no further subsidence occurred.
