This paper investigated the effect of transport layer in the diversion capacity of natural capillary barrier system using laboratory slope model. The slope model was constructed with stainless steel and 5 mm thick acrylic sheets. Grade VI and grade V soils classified as sandy silt of very high plasticity (MVS) and silty gravel of high plasticity (MHG), respectively were arranged in the slope model with sand and gravel transport layers sandwiched at their interface. The model was subjected to different rainfall intensities using rainfall simulator to determine the behaviour of water flow and suction distribution in the slope model. The results obtained show a modification in the suction distribution behaviour and the natural capillary barrier effect was sustained. It was also observed that a transport layer formed with gravel material was more effective in diverting the infiltrating water compared to that of gravelly sand. This occurred because the upper grade VI layer possessed capillary forces due to its finer pore structures and relatively large air entry value, thus, it retained the infiltrating water, and the gravel transport layer possessed relatively larger pore structures compared to the grade VI layer and hence it possessed higher hydraulic conductivity values and small water entry value. This arrangement, provide a capillary break and allowed the infiltrating water to flow above the interface. Therefore, the inclusion of transport layer provides a definite path through which the infiltrating water flows and diverted laterally. Thus, improving the performance of natural capillary barrier effect.

No abstract

This study is an investigation into the possibility of using rice husk ash blended with cement (C-RHA) in the modification of the basic engineering properties of a marginal lateritic soil with the aim of qualifying the soil for use in the construction of roads. In the study, the properties of the natural soil sample such as bearing strength (measured in California Bearing Ratio: CBR), compaction behaviour as well as other consistency parameters were evaluated. The same properties were also evaluated for the soil samples modified with C-RHA. The findings of the study show that an increase in values of the C-RHA results a decrease in plasticity. An improvement in the CBR was observed with increasing value of the RHA at specified cement contents with peak values at 5% cement and 10% RHA. This indicates the potential of using 7.5-10% RHA admixed with 5% cement contents for laterite soil stabilization for use as a sub-grade soil for road construction.

Particle size distribution and void ratio of a soil are considered as the direct information that can be used in a relatively easy manner for hydraulic conductivity estimation. Estimation of hydraulic conductivity from particle size distribution can be used to check permeability values obtained from other methods. Therefore, this paper attempts to relate the particle size distribution to hydraulic conductivity. The study was carried out on 24 soil samples which were collected from the embankment of an earth reservoir and subgrade at the toe. The investigation was carried out in accordance with the standard procedure given in BS1377. A series of hydraulic conductivity tests were carried out on optimum moisture content (OMC) compacted soil samples using the falling head method. The mean sizes of particles in each sample from particle size distribution curves were determined and used to generate regression models for the flow. Linear, exponential, polynomial and logarithmic models were used to test the validity; however, the best was adopted for this study. The findings of this study show that there is variability in the particle sizes of the soil material which in turn results to variation in the hydraulic conductivity. The hydraulic conductivity was found to increase with an increase in mean particle size. The relationship between mean particle size and hydraulic conductivity yielded coefficients of determination (R 2 ) of stronger correlation when the plastic and non-plastic samples were separately analyzed. However, all values of R 2 (0.9949, 0.9968 and 0.8918 for samples 1 to 16, 17 to 24 and 1 to 24 respectively) can be considered satisfactory. In addition, generalized models for the flow were generated for the plastic, non-plastic and the combined samples. The generated models can be used to predict the hydraulic conductivity of different soil samples.

No abstract