Hydro-mechanical properties of lime-stabilised shrinkable UK clays

Lead Research Organisation: London South Bank University
Department Name: Fac of Eng Science & Built Env

Abstract

Some clay soils of high plasticity show large volumetric changes in response to seasonal cycles of wetting and drying. Such clays are called shrinkable or expansive clays. In the UK, a range of shrinkable clays can be found in South and East England. Due to their presence, seasonal variations in moisture content can cause annual surface movements as high as 50mm. This can cause significant distress and damage to civil engineering structures (in particular light buildings and pavements). Previous research has estimated that the damage caused by shrinkable soils may be higher than damage due to any other natural hazard, including earthquakes and floods. This can result in millions of pounds of damage to homes every year, as several millions of homes are affected, and has also led to a rapid increase in insurance claims for 'subsidence damage' to low-rise buildings, principally housing, in the UK. Maintenance and repair cost of the affected infrastructure (pavement, railway tracks etc) can also be very high and may incur severe traffic delays.One method to improve the engineering properties of the shrinkable soil and prevent damage to structures is to stabilise the soil using chemical additives and in particular lime. The technique is particularly attractive, as it allows the improvement and use of the soil in-situ. For this reason, it is often a more economic (in terms of both time and money) and environmentally-friendly alternative to the conventional methods, which involve removal and disposal of the unsuitable soil to landfills and its replacement with imported quarried aggregate material. When the lime-treated material is in place it is subject to environmental effects, including seasonal cycles of wetting and drying. This implies that during its design life, the lime-treated soil will be at different states, from being fully saturated with water, to almost completely dry. When the soil is in a partially saturated state, its behaviour and properties are fundamentally different from those when it is fully saturated. It is therefore important to know its properties and behaviour in both states. Such a complete picture of the lime-treated soils is currently lacking.With the scope of an increased use of the technique in a wider range of projects and applications, the need has emerged for predictive numerical tools able to describe the behaviour of lime-treated soils for a variety of problems and conditions (including partially saturated soil conditions). Such tools are not available at the moment for this type of soil. The aim of this research is to develop computer software which will allow predictions to be made. This will be based on new experimental data on lime-treated UK clay soils, which will allow to create an appropriate model for the behaviour of the lime-treated soil. An additional novelty of the research is that both fully saturated and partially saturated soil conditions will be considered. The predictive tool will enable engineers to perform sophisticated analyses and to make better informed decisions on the behaviour of lime-treated materials in a variety of applications. The research is timely as with extreme weather conditions becoming more common due to climate change, engineering problems due to cyclic shrinkage/expansion of soils are expected to rise in the years to come. Moreover there is currently an increased environmental and economic pressure to upgrade material by stabilisation of in situ soil, as an alternative to its export to landfill and replacement by imported granular fill. In particular, the EU Landfill Directive on the Landfilling of Waste (Council Directive 1999/31/EC) which has also been transposed to Britain, is based on the policy objective to manage resources in a more sustainable manner. Improving the in-situ soil with lime is a sustainable solution, as it saves construction time and money (including landfill taxes) and reduces emissions from vehicles.

Publications

10 25 50
publication icon
M. Mavroulidou (2009) A study of the laboratory measurement of the Soil Water Retention Curve in 11th International Conference on the Environmental Science and Technology (CEST2009), 3 - 5 September 2009, Chania, Greece

publication icon
M. Mavroulidou (2013) Hydro-mechanical properties of lime-treated London Clay in Proceedings of the 18th International Conference on Soil Mechanics and Geotechnical Engineneeing

publication icon
Maria Mavroulidou (Author) (2012) An investigation of the effects of cementation and suction on lime treated London Clay in Unsaturated Soils 2011: Theory and Practice

publication icon
Mavroulidou M (2013) Water Retention and Compressibility of a Lime-Treated, High Plasticity Clay in Geotechnical and Geological Engineering

publication icon
Roberto Tamagnini (Author) (2012) Thermodynamics of unsaturated soils in Unsaturated Soils 2011: Theory and Practice

publication icon
Tamagnini R (2010) Unsaturated Soils

 
Description Lime treatment of clay soils is a widely used technique in civil engineering (particularly road construction) to improve clay properties so that the likelihood of excessive settlements or heave movements is reduced. A material like London Clay (as well as other UK clays) has been much researched over the years by the Soil Mechanics community and an extensive database of information about the relevant properties (stiffness, strength) now exitsts. In contrast there is little information about the properties of lime treated UK clays (in particular when the soil is partially saturated) and this research has addressed this issue.

In this research a series of comparisons were made between the mechanical behaviour of untreated and lime-treated London Clay by a large number of drained (mostly) and undrained triaxial tests. All specimens were prepared by static compaction and both saturated and partially saturated specimens were tested. The triaxial tests showed that all saturated and partially staturated lime-treated specimens manifested a very pronounced peak in the strength (with strengths substantially higher than the untreated soils) within relatively low strain levels, followed by strain softening, with the stress decreasing dramatically after only 3-5% of axial strain. The behaviour was manifested irrespective of the overconsolidation ratio of the specimens.

The overall behaviour is consistent with the breakage of the cementation bonds. The results indicated that water curing and modest percentages of lime could be recommended as there was a sufficient increase in the stiffness and strength of the soil without resulting in very brittle behaviour and abrupt loss of strength upon strain softening within the range of strains involved in engineering design. Increasing suction resulted in an increase in the peak strength and stiffness of the lime-treated London Clay as is the case for untreated soil samples. The predominant factor controlling strength for the ranges of suctions used was found to be lime-treatment and consequent cementation bonding while this is not damaged (rather than suction).

Both types of soil tended to ultimate states (this was also testified by the volume change behaviour reported later). The lime-treated specimens tested under the same mean net stress appeared to reach the same ultimate strength irrespective of the suction. This is consistent with the observed behaviour of uncemented partially saturated soils. Specimens sheared under the same suction but different mean net stresses showed the peak strength to increase with mean net stress. Higher suctions appeared to induce a more gradual softening.

A non-linear relationship between peak shear strength and matric suction was found, whereas the ultimate shear strength was shown to be unaffected by the levels of suction considered in this study. Analysing the stress paths soils, it was found that lime did not appear to have modified the frictional properties of the soil. The critical state of the partially saturated lime-treated samples was initially expressed successfully using models suggested in the literature. A model along similar lines was also suggested during the grant to fit the peak values of the lime-treated specimens, and it showed a very good fit of the results.

A novel non-contacting system for the volume change measurement of unsaturated soils during triaxial testing was developed during the grant in collaboration with a commercial company, in an attempt to achieve higher accuracy in the measurements, simplicity of operation, and in particular, ensure continuous measurements for the partially saturated lime-treated soil with a brittle behaviour. This new volume measurement system successfully measured volumes and ensured continuous volume measurements up to large strain levels during testing. This included the post-peak region of the brittle, arti?cially cemented specimens (which cannot be reliably measured by local measurement techniques such as LVDTs and Hall effect sensors). The measurements showed that, as opposed to the untreated statically compacted London Clay specimens, which were continuously contracting, the volume change of the lime-treated specimens showed an initial contraction followed by some dilation. The amount of contraction decreased with increasing suction. On the other hand, for the same level of mean net stress, dilatancy increased with increasing suction, whereas for the same suction, higher net stresses increasingly suppressed the dilatancy of the specimens.With continuous shearing all specimens appeared to tend towards critical state, manifested by the recording of constant total volumetric strains, which was consistent with the stress-strain behaviour of the soil.

Based on suction-controlled oedometer and isotropic compression results the marked decrease in the swelling and compressibility of the lime-treated soil was quantified in comparison to that of the untreated clay; the wetting induced swelling behaviour of the material was found to be strongly affected by the mean net stress consistent with the behaviour of untreated soils. The increase in the vertical effective yield stress upon the addition of lime was also quantified for both saturated and partially saturated soils. During compression, the lime enabled the soil to maintain a more open structure with respect to the untreated soil throughout the compression process. Overall, the tests generally indicated that the compressibility of the lime-treated soil was largely controlled by the cementitious bonding (due to lime) rather than suction for the ranges of suction tested. Based on the data, a simplified relationship between suction and yield stress was suggested for the mathematical description of the isotropic compression data for different suctions.

The water retention capacity of the treated and lime-treated soil was also investigated for different compaction conditions with a number of techniques (filter paper, an in-house modified pressure plate apparatus and using the triaxial cell to assess the effect of different confinement pressures). The Soil Water Retention Curves of the lime-treated and untreated soil were determined for drying and wetting as long as subsequent re-drying cycles, which is relevant to changes in the water content of the soil in the field, due to seasonal variations. Two different curing conditions simulated shallow and deep treatment of the soil with lime, respectively. The filter paper tests showed that the shallow-mixed lime-treated soils are likely to be partially saturated in situ depending on the environmental conditions; thus, unsaturated soil mechanics concepts are applicable for the description of their behaviour.

All water retention tests showed that, for two different lime percentages the effect of the lime on the water retention capacity of the London Clay soil was not overall much different compared to the respective untreated sample, despite the fact that lime has changed the soil structure (confirmed also by SEM tests), and although there is a change in the nature of the soil after treatment. As with the untreated London Clay soil, the water retention curves of the lime treated specimens were found to be related to the compaction water content, compaction dry density and the drying/wetting paths followed. As for untreated soils, the lime-treated sample compacted dry of optimum, showed a lower water retention capacity compared to the respective sample compacted dry of optimum due to the more open structure.

The curing time and method (air versus water curing and confined versus unconfined water curing) were found to affect the results only slightly (in particular the curing time effect was manifested only after long curing times). The deformability of the lime-treated samples compacted wet of optimum upon suction changes was higher which was consistent with the behaviour of the untreated soil. However the lime treated samples showed overall much lower volumetric strains with respect to the untreated soil, especially for the higher lime content as a result of cementation. Overall it was clearly noted that bonding induced by lime considerably affected the strain related quantities (the void ratio and the volumetric strain) linked to the increased stiffness of the lime treated soil. The results of these tests were consistent with independently conducted swelling and shrinkage tests during the grant (the former tests also showed the considerable decrease in the swelling pressure of soil upon lime treatment).

Overall, from an engineering point of view, it was concluded that lime had generally a favourable effect on the soil for the range of strains and suctions relevant for common engineering design situations in the UK.

In addition, a constitutive model for the cemented, partially saturated soil was developed extending a modified Cam-Clay model to the description of this type of soil, proven theoretically to be consistent with the thermodynamics principles. The model was implemented into the commercial Finite Element software Abaqus using an implicit integration formulation. The model and its numerical implementation was intended as an engineering tool for the analysis of the lime-treated soil in sophisticated situations. The model was verified based on a number of experimental data giving a good fit and was applied to two prototype structures (shallow foundation and embankment) to investigate the effect of lime treatment in terms of parametric studies. Some potential practical consequences regarding the relative stiffness of the treated soil versus the untreated foundation soil were identified as a result of the numerical analyses.
Exploitation Route The findings are relevant to the civil engineering industry regarding the design and analysis of lime-treated earthworks. From the academic point of view they are of interest for the development of new tools for the experimental testing of cemented unsaturated soils and give a considerable amount of quality experimental data regarding the properties and behaviour of these soils. Modelling tools extending familiar to engineers and researchers models to describe the behaviour of this type of soil have also been developed
Sectors Construction,Environment

 
Description Italian geotechnical company linked to railway infrastrcture showed interest in the research and in particular constitutive modelling aspect
First Year Of Impact 2010
Sector Construction
 
Description Internal PhD scholarship
Amount £12,000 (GBP)
Organisation London South Bank University 
Sector Academic/University
Country United Kingdom
Start 02/2010 
End 02/2014
 
Title Volume measurement system for unsaturated soil testing 
Description We developed a novel volume change measurement system for the triaxial testing of unsaturated soils. The main impetus for this research was the development of a reliable, high-accuracy volume measurement technique for arti?cially cemented (lime-treated soils, i.e. the topic of the EPSRC grant) unsaturated soils, whose failure mode was expected to be brittle. For this type of soil, direct measurements based on local strain measuring devices would not be reliable at strains correspondingto post-peak strengths. The system was also introduced to over come overall system complexity, equipment saturation, and transducer issues, linked to other systems used for volume measurements. The proposed technique uses a high accuracy linearized laser sensor to measures unsaturated soil volume changes indirectly an inner cell inside the triaxial system and an hydraulically connected outer cell. The operating principle of the system relies on the level of water in an open inner cell changing due to the total volume change of the tested specimen. As the water level inside the inner cell changes due to specimen deformation, so does the water level in the external measurement cell. This change is captured by the laser sensor, placed above the external measurement cell. The system was validated, calibrated and proved for the testing of unsaturated lime treated soils; the development of this tool, and results have been published in a number of conference and journal papers. 
Type Of Material Improvements to research infrastructure 
Provided To Others? No  
Impact Collaboration with major geothechnical research equipment development company (GDS Ltd) who showed interest in the system and potential commercialisation. Publications are co-authored by GDS executive (Mr Jerry Sutton) and Geotechnical Consultant Group (GCG) Senior partner (Dr Cabarkapa) 
 
Title Constitutive model 
Description We introduced a Cam Clay type constitutive model extended for cemented/unsaturated soils (related to lime treatment). 
Type Of Material Computer model/algorithm 
Provided To Others? No  
Impact The model is based on critical state soil mechanics theory which is familiar to geotechnical engineers. It includes a number of parameters which are relatively easily identifiable from soils testing, hence providing a relatively easy to use practical tool for geotechnical engineers 
URL http://ebooks.iospress.nl/ISBN/978-1-60750-801-4
 
Description GCG 
Organisation Geotechnical Consulting Group
Country United Kingdom 
Sector Private 
PI Contribution Expertise in lime treated and unsaturated soils
Collaborator Contribution Expertise in geotechnical applications and advice on testing
Impact 8 co-authored publications (with Dr Z. Cabarkapa, GCG): Zhang, X. Mavroulidou, M., Sutton, J., Cabarkapa, Z. and Gunn, MJ (2012) "Demonstration of a novel volume change measurement technique for unsaturated soils", Geotechnical Testing Journal , 35(5), 673-677, doi: 10.1520/GTJ104613 Mavroulidou M, Zhang, X., Gunn M.J. and Cabarkapa, Z. (2013) "Water retention and compressibility of a lime-treated, high-plasticity clay" Geotechnical and Geological Engineering ; 31(4) 1171-1185; DOI 10.1007/s10706-013-9642-6; ISSN 0960-3182 Zhang, X. Mavroulidou, M., Gunn, MJ, Sutton J., Cabarkapa, Z., Kichou, Z. (2014) "Application of a novel laser sensor technique to the volume measurement of cemented and uncemented unsaturated soils during triaxial testing", Acta Geotechnica; 9 , 945-957; DOI 10.1007/s11440-013-0254-3; Zhang, X. Mavroulidou, M. Gunn MJ and Cabarkapa, Z. (2009) "Experiences gained using various axis translation technique apparatus to determine soil water retention curve", Experimental Studies in Unsaturated Soils and Expansive Soils, Proceedings of the 4th Asia Pacific Conference on Unsaturated Soils, Newcastle, Australia, 23-25 November 2009, Buzzi O., Fityus S., Sheng D., CRC Press ; 269-276. Mavroulidou, M, Zhang, X. Cabarkapa, Z. and Gunn MJ (2009) "A study of the laboratory measurement of the Soil Water Retention Curve", 11th International Conference on the Environmental Science and Technology (CEST2009), 3 - 5 September 2009, Chania, Greece, pp. A907-A915; ISBN:9789607475466 Zhang, X. Mavroulidou, M. Gunn MJ, Cabarkapa, Z. (2010) "A study of soil water retention characteristics of lime-treated and untreated London Clay", 5th International Conference on Unsaturated Soils UNSAT2010, Barcelona, 6-8 September 2010, pp. 751-763 Zhang, X. Mavroulidou, M. Gunn MJ, Cabarkapa, Z. and Sutton J. (2010) "A novel non-contact laser displacement transducer technique for volume change measurements of unsaturated soils", 5th International Conference on Unsaturated Soils UNSAT2010, Barcelona, 6-8 September 2010, pp. 519-524 Mavroulidou M, Zhang, X., Gunn MJ, Cabarkapa Z (2012) "An investigation of the effects of cementation and suction on lime treated London Clay" In Unsaturated Soils 2011: Theory and Practice, Proceedings of the 5th Asian-Pacific Conference on Unsaturated Soils, Papaya, Thailand , Jotisankasa et al., (eds), Kasetsart University Thailand, pp. 527-532; ISBN: 987-616-7522-77-7;
Start Year 2008
 
Description GDS 
Organisation GDS Instruments
Country United Kingdom 
Sector Private 
PI Contribution Suggested a novel apparatus design for the volume change measurement of unsaturated soils
Collaborator Contribution Built the actual system and gave advice for further refinements
Impact 3 co-authored publications (with J. Sutton, GDS Instruments Ltd) Zhang, X. Mavroulidou, M., Sutton, J., Cabarkapa, Z. and Gunn, MJ (2012) "Demonstration of a novel volume change measurement technique for unsaturated soils", Geotechnical Testing Journal , 35(5), 673-677, doi: 10.1520/GTJ104613 Zhang, X. Mavroulidou, M., Gunn, MJ, Sutton J., Cabarkapa, Z., Kichou, Z. (2014) "Application of a novel laser sensor technique to the volume measurement of cemented and uncemented unsaturated soils during triaxial testing", Acta Geotechnica; 9 , 945-957; DOI 10.1007/s11440-013-0254-3; Zhang, X. Mavroulidou, M. Gunn MJ, Cabarkapa, Z. and Sutton J. (2010) "A novel non-contact laser displacement transducer technique for volume change measurements of unsaturated soils", 5th International Conference on Unsaturated Soils UNSAT2010, Barcelona, 6-8 September 2010, pp. 519-524
Start Year 2008