BEHAVIOUR OF SAND - FIBER MIXED GRANULAR PILES

Abstract

here are many situations wherein the soil is extremely weak or soft. Due to decrease in availability of good construction sites, now-a-days constructions are being carried out on otherwise poor sites. Special types of foundation or large foundation involving heavy expenditure are needed to construct structure on these soft soils. To utilize these poor sites economically, various in-situ ground improvement techniques have been developed during last four decades. Many sophisticated ground improvement techniques are available and may be employed depending upon the soil conditions. Stone column / granular pile has been recognized as a useful technique to strengthen soft cohesive soils and loose silty sand deposits. The main concept of stone column / granular pile reinforcement involves the replacement of some percentage of weak soil with coarse granular material such as stone, sand and stone chips with sand in the form of column and applicable to those structures which can permit higher settlements (100 to 300 mm). Due to installation of stone column / GP, the overall stiffness of the ground increases. As a result, the load carrying capacity of the ground increases and also considerable reduction in settlement takes place. From literature review, it has been established that bulging is one of the most common failure mechanisms in long granular piles (granular pile length > three times of granular pile diameter). Due to this bulging failure, load carrying capacity decreases and settlement increases. Load carrying capacity of granular pile can be increased and bulging failure of granular pile can be reduced by using concrete plug or cement grout, placing horizontal reinforcement in the upper portion of the granular pile, using circumferential reinforcement like geo-grid casing, skirting the granular pile or piles group and improving the quality of the granular material like mixing fiber with sand and using it as granular pile. Experimental investigations on granular pile have been reported by various investigators among them remarkable work has been done by Williams (1969), Greenwood (1970), Hughes and Withers (1974), Hughes et al. (1975), Madhav (1982), Bergado et al. (1984, 1988 1990, 1991), Bergado and Lam (1987), Madhave & Thiruselvam (1988), Rao et al. (1992), Wood et al. (1996, 2000), Rao et al. (1997), Sharma et al. (2004), Sivakumar et al. (2004), Black et al. (2007, 2007) and Ambily and Gandhi (2007). It has been established, from the literature, that randomly oriented fiber in sand (ply soil) considerably improves the load carrying capacity of shallow foundations. The performance of random fiber in granular pile is totally different from ply soil which is yet to be investigated in details. In this case, load is applied on composite ground including soft soil and granular pile. Load from structure is transferred to deeper depths through GP by skin friction. There is very scanty research work available on random fiber mixed granular pile. In 1992, Al-Refeai performed triaxial shear tests to study the behavior of soft soils reinforced with fiber-strengthened sand columns. It has been found that the column replacement ratio and depth of fiber reinforced sand layer inside column have a significant effect on load carrying capacity and settlement response of the reinforcement soft soil. The study has shown that a fiber content of 0.2% by weight can be used effectively to increase resistance of sand without any reduction in the soil density and permeability. Ranjan et al. (1999) carried out experimental study on soft clay reinforced with sand-fiber core. They conducted a series of triaxial compression tests on soft clay sample (50 mm diameter and 100 mm height) with central sand-fiber core. Till now, no study incorporating model or prototype testing on random fiber mixed GP is reported in literature. None of the studies investigated bulging of GP, effect of sand pad on load carrying capacity and settlement of GP at various depths. Also no study on group behavior of GP with random fiber has been carried out. The load carrying capacity of GP depends on passive resistance offered by surrounding soft soil. In actual field situation, the passive resistance is very low at the top of GP and increases with depth. So, bulging failure takes place in the GP within shallow depths from top. But in the case of triaxial simulation, the circumferential stress remains constant throughout the depth of soft soil. That means no variation in passive resistance. On the other hand, sample size is very small to study exact behavior of random fiber mixed granular pile. Keeping all these parameters in mind, it has been decided to perform model tests on soft clay reinforced by sand fiber mixed granular pile. In order to achieve above objectives, model tests have been conducted in laboratory with varying fiber content, fiber length, fiber depth, footing size and sand pad thickness. A detailed parametric study with random fiber has been carried out to observe the effect of fiber content, fiber length and fiber depth in GP and sand pad thickness on load carrying capacity, elastic rebound, bulging behavior and settlement profile of GP. From literature, it has also been found that no studies have been carried out on end bearing GP with varying length of circumferential reinforcement (geo-grid tube). So model tests with circumferential reinforcement have also been conducted. Effects of sand pad, settlement of GP and group behavior with circumferentially reinforced GP have also been studied. To study drainage properties of GP with random fiber, a series of falling head permeability test has been conducted with varying fiber content and fiber length. Similarly, a series of direct shear test has been conducted with varying fiber content and fiber length to investigate shear strength parameters of sand fiber mixture. Short term load tests have been conducted as per procedure mentioned in Federal Highway Administration (U.S. Department of Transportation, Report No FHWA/RD - 83/026). The results of the experimental work, analysis and conclusions have been presented in this thesis. It has been found that permeability, strength parameters, load carrying capacity, settlement pattern, bulging pattern etc. of granular piles are influenced by the inclusion of fibers in granular pile material. Also the circumferential reinforcement also has a definite effect on the behavior of granular piles. The results of the model tests have been found to compare fairly well with the earlier research findings. Conclusively, the results of the study can be used as a basis for further numerical and field studies. The new material being stiffer than the sand alone, it can be used effectively where no other alternative is available. The equations developed may be used as design aids in future. It is recommended that for optimum conditions, 0.5% fiber content (for facilitating proper mixing), 0.6d fiber length and 3.0d fiber depth is the best combination for randomly fiber mixed granular pile to improve its load carrying capacity, drainage property and to accelerate the consolidation process. Fiber content may be kept up to 2.0% in the granular pile without reducing the permeability of sand. In case of circumferentially reinforced (geo-grid tube) granular pile, the length of the geogrid tube may be chosen as 4.0d. For both the cases of reinforcement, a sand pad thickness T = 0.5d may be chosen for enhancing the load carrying capacity.