Study on cement and lime stabilised chittagong coastal soils for use in road construction

Abstract

In the present study, cement stabilisation of two selected soils (collected from Anwara and Banshkhali) and lime stabilisation of a soil (collected from Banshkhali) of Chittagong coastal region were carried out in order to assess their suitability for use in road construction. The soils from Anwara and Basnkhali were respectively a clayey silt of low plasticity (LL = 30, PI = 7) and a silty clay of medium plasticity (LL = 44, PI = 19). As additives, ordinary Portland cement was used in percentages of 1, 3 and 5 while slaked lime was used in 3%, 5% and 7%. Analytical works using CIRCLY computer program were also undertaken in order to determine the thickness of cement stabilised bases and lime stabilised sub-bases of rural roads in the coastal region. Index tests indicated that compared with the untreated samples, plasticity index and linear shrinkage of the cement and lime stabilised samples of the soils reduced. Shrinkage limit, however, reduced for cement-treated samples while it increased for lime-treated samples. For the cement and lime stabilised samples, maximum dry density increased and reduced respectively, while optimum moisture content reduced and increased for cement and lime stabilised samples respectively with the increase in additive content. For samples of both the coastal soils, compared with the untreated samples, unconfined compressive strength (q,) of cement and lime treated samples increased significantly, depending on the additive content and curing age. It was found that compressive strength of samples treated with 3 % and 5 % cement and cured for 14 and 28 days satisfied the requirements of PCA (1956) for the compressive strength of soil-cement mix and that for all cement contents and all curing ages, compressive strength of the stabilised samples fulfilled the requirements of soil-cement mix for use in road sub-base and base subjected to light traffic, as proposed by Ingles and Metcalf (1972). It was also found that the compressive strength of samples treated with 5 % and 7 % lime met the requirements for upgrading heavy clays to sub-base material quality type, as proposed by Ingles and Metcalf (1972). In an attempt to investigate the effect of moulding water content on q" it appeared that in order to achieve maximum compressive strength, the cement and lime stabilised samples should be compacted at their optimum and wet side of optimum moisture content respectively. Compared with the untreated samples, CBR of the cement and lime stabilised samples increased considerably. It was found that CBR-values of cement and lime stabilised samples increased up to about 5 times and 3 times respectively. CBR-values of samples of both the soils, treated with 3% and 5% cement, fulfilled the requirements of soil-<:ement road sub-base and base for light traffic while CBR of sample stabilised with 7% lime did not satisfy the criteria of the minimum CBR for soil-lime mix for improvement of base material in road construction, as proposed by Ingles and Metcalf (1972). The flexural stress versus deflection curves have been found to be approximately linear for both cement and lime stabilised samples. Compared with the untreated samples, flexural strength and flexural modulus of the cement and lime stabilised samples increased considerably, depending on the additive content. Compared with the untreated sample, the flexural strength and flexural modulus of cement-treated samples increased up to about 6 times and 4.3 times while for lime-treated samples the respective increases were about 2 times and 2.25 times. The loss in soil-cement of cement-treated samples reduced with the increase in cement content. Although, the cement-treated samples did not meet the PCA (1956) durability requirements, the samples treated with 3 % and 5 % cement, however, fulfilled the requirements as suggested by Compendium 8 (1979). It was found from comparisons that the values of q" CBR, flexural strength and flexural modulus of the cement-treated samples of Banshkhali were significantly higher than those of the lime-treated samples. Moreover, it is expected that compared with soil-lime mix, soil-cement mix would be much more durable in the weather conditions of tropical regions. It could be concluded that cement stabilisation of the coastal soils studied would be more suitable than lime stabilisation for their use in road construction. Analytical results indicated that at a particular modulus and CBR, the thicknesses of soil-cement base (for both unpaved and paved road) and soil-lime sub-base increased with increasing allowable number of load repetitions to fatigue (N) and that for any particular value of N, the values of thicknesses of soil-cement base (for both unpaved and paved road) and soil-lime sub-base reduced as modulus of soil-cement base and soillime sub-base increased. Design charts have been developed for the three types of roads analysed and it has been found that, in general, at any CBR and N, the thickness of cemented layer reduced as modulus of the cemented layer increased.