Resilient modulus of black cotton soil

Resilient modulus (MR) values of pavement layers are the basic input parameters for the design of pavements with multiple layers in the current mechanistic empirical pavement design guidelines. As the laboratory determination of resilient modulus is costly, time consuming and cumbersome, several empirical models are developed for the prediction of resilient modulus for different regions of the world based on the database of resilient modulus values of local soils. For use of these relationships there is a need to verify the suitability of these models for local conditions. Expansive clay called black cotton soil (BC soil) is found in several parts of India and is characterized by low strength and high compressibility. This soil shows swell - shrink behaviour upon wetting and drying and are problematic. The BC soil shows collapse behaviour on soaking and therefore the strength of the soil needs to be improved. Additive stabilization is found to be very effective in stabilizing black cotton soils and generally lime is used to improve the strength and durability of the black cotton soil. In this paper, the results of repeated load tests on black cotton soil samples for the determination of MR under soaked and unsoaked conditions at a relative compaction levels of 100% and 95% of both standard and modified proctor conditions are reported. The results indicate that the black cotton soil fails to meet the density requirement of the subgrade soil and shows collapse behaviour under soaked condition. To overcome this, lime is added as an additive to improve the strength of black cotton soil and repeated load tests were performed as per AASHTO T 307 - 99 for MR determination. The results have shown that the samples are stable under modified proctor condition with MR values ranging from 36 MPa to 388 MPa for a lime content of 2.5% and curing period ranging from 7 to 28 days. Also, it is observed that, the CBR based resilient modulus is not in agreement with the laboratory based MR value and therefore it is necessary to use the laboratory determined MR or MR predicted from an appropriate prediction equation for pavement design. An attempt is also made to develop an empirical model for the prediction of MR of lime stabilized black cotton soil. There is a good agreement between the predicted and laboratory MR values with an error of 12%.