Experimental Simulation Of Solute Transport In Porous Media

The dispersion is an important parameter to quantify solute transport in ground water of porous media. In the traditional advection-dispersion equation, dispersion is often treated as a constant. In recent years, many studies showed that variation in the dispersion has small effect that it can be approximately treated as constant in large scale model, and in a small scale model, dispersion parameters have scale dependence, and the mechanism is not yet clear. On the basis of summarizing and analyzing groundwater solute transport of domestic and overseas studies, we established the unidimensional physical experimental model of solute transport through porous media by perspex sheet, used inert tracer (blue tracer) to trace the solute transport under the condition of Darcy flow. In order to avoid the influence of the sampling method on experimental results, this paper adopts the method of taking pictures to improve the accuracy of the experiment, treats dispersion as a function, which is a linear relation with transport distance, establishes the mathematics model of LAF, the breakthrough curves of solute transport have been simulated by the mathematical model of ADE and LAF, and the fitting results are analyzed, summarized the mechanism and regularity of solute transport in porous media. The experimental results of this paper are summarized, and the following conclusions are drawn:(1) During the processes of homogeneous and heterogeneous solute transport, the phenomenon of arrive early in the first half of the breakthrough curves is found for the existence of dominant flow. Due to adsorption and desorption, resulting in the tailing phenomenon in latter part of the breakthrough curves.(2) when compared the simulation effect between the model of ADE and LAP in the breakthrough curves of solute transport, it can be concluded that:when we simulate the homogeneous porous medium by a homogeneous glass beads packed column, the evaluation parameter of LAP simulation effect determines the coefficient R2 and the root mean square error RMSE. When the R2 are more than 0.95, the values of RMSE are very small, less than 0.1, indicating that the accuracy of the LAF model is very high for simulating unidimensional solute transport in porous media. The evaluation parameters of ADE simulation effect determines the coefficient R2 and the root mean square error RMSE. when the R2 are less than 0.85, the values of RMSE are more than 0.05, and the R2 of the LAF model is larger than the R2 of the ADE model, and the RMSE of the LAF model is less than the RMSE of the ADE model. The results show that the simulating effect of LAF model is better than that of ADE model.(3) In the fitting of LAF mathematical model, we emphatically discussed the value of critical point Lo, the value of Lo in 1-2 mm homogeneous quartz sand packed column is 1.8. The value of Lo is 7.4 in 0.5mm-1.0mm white sand,1.0mm-2.0mm quartz sand and 3.0mm glass beads in heterogeneous medium. The value of Lo is 5.7 in the 3.0mm glass beads,1.0mm-2.0mm quartz sand, and 0.5mm-1.0mm white sand in the heterogeneous medium. From this we can find that the value of Lo in homogeneous medium is obviously smaller than that of Lo in non-homogeneous medium. The value of Lo under the condition of the porosity gradually get larger is greater than the porosity gradually get smaller.