The hydraulic behavior of free flowing granular drains
Granular drains are among the most effective tools for controlling the migration of contaminants. For long-term decommissioning in the mining industry, these drains might be subjected to clogging caused by fine particles of the surrounding soils. Fine particles become trapped in the granular drains, lowering permeability and reducing effectiveness. Therefore, it is necessary to minimize the effect of clogging by allowing particles to move more freely within the drains. The primary objectives of this research were to: i) Develop an understanding about the nature of clogging in sand drain materials subjected to infiltration of fines; ii) Measure the impact of a wide range of fines on clogging of typical sand drain materials; and iii) Modify an existing clogging model to more accurately predict permeability reduction. This study is a continuation of Bergerman’s (2011) study. The laboratory test program was built based on that study with some modifications to the test apparatus to gain better control over the output data. In addition, this study examined a wider range of variables, including particle size, clay activity and swelling capacity. The laboratory test program was conducted using three types of granular drainage soils: Coarse sand, Uniform sand and French drain sand, all under low gradient conditions. In order to simulate the transient effects of different fines on drainage sands, various suspensions of kaolinite, Battleford till and bentonite fines were used to infiltrate the sands. The test equipment used was similar to that of Bergerman’s (2011) study but with bigger inlet and outlet tubes and a fixed inflow elevation. These modifications were made with the goal of achieving a condition in which all head losses occurred only in the sand. Test results were recorded in terms of total heads versus time and were interpreted to determine the changes in head due to clogging of fine particles in the inflow. A normalization method was employed in the interpretation process to scale all test results to the same boundary conditions, which is important when comparing results from one test to another. The result of the test program showed that Coarse sand and Uniform sand were able to maintain flow longer than French drain sand due to their coarse nature. The positions of fines captured in these sand samples were also different. Fines were captured mostly at the effluent end in the Coarse sand column, at the middle of the Uniform sand column, and at the influent end of the French Drain sand column. Among the fines chosen, Bentonite was the most active clay with a plasticity index of 473%, activity of 6.5 and modified free swelling index of 58.5. The presence of fully water-absorbed bentonite particles and flocculation caused fast clogging in the sand samples. A model, developed by Reddi et al. (2000), was chosen to simulate the test results in this study. The proposed modelling equations were applied with modifications in critical velocity of flow and the characteristic pore length of drain sand. These modifications were part of the calibration process, and they reflected the properties of materials used in this study. Most modelling results were in good agreement with the test results after calibration, except for Coarse and Uniform sands with bentonite. The unpredictable behavior of bentonite could be due to its great swelling potential and ability to create big flocculated particles. This model is therefore best used to simulate the behavior of clay suspensions with low-to-medium swelling capability. High-swelling clays would require additional effort in adjusting input parameters and perhaps modifying the model itself.
Granular drains, Clogging, infiltration, drainage soil, head loss, hydraulic head, normalized head.
Master of Science (M.Sc.)
Civil and Geological Engineering