MULTILAYERED SYSTEMS


General Considerations

Earth scientists and geotechnical engineers are increasingly challenged to solve environmental problems related to waste disposal facilities and remediations of polluted sites by pollutants containment systems. The design of such structures typically involves some form of "barrier" that separate the contaminants from the general sub-soil and groundwater systems. The primary objective of a barrier is to minimize as much as possible the advective and dispersive transport rate of the pollutants in solution. The knowledge of the parameters that characterize the different types of mineral barriers versus different types of pollutants is of outstanding importance. While their assessment in laboratory tests doesn't pose any particular problem from a practical point of view, the interpretation of the results of the experiments is difficult, because the model that describes correctly the transport properties of the contaminant in porous media can be solved analytically only in very special cases. The validity of these analytical solutions is usually extrapolated to explain the results of the experiments, very often with poor agreement.
The most important factor causing these discrepancies is that the analytical solutions available in the literature are obtained by applying right at the top of the sample two kinds of boundary conditions, given by continuity of concentration (so called first type) and continuity of flux (so called third type). In addition, a finite or a semi-infinite system can be considered, resulting in four different situations. In fact, as suggested in (1) and (2), the appropriate model should be chosen by taking into account the presence of all other elements in the system, including those in direct contact with the soil sample, such as very thin and long connecting lines.

Results


In papers (1) and (2), we propose a numerical solution of the contaminant transport equations in porous media for a multi-layer system with general geometry, where every layer represents an element of the apparatus used for the assessment of the seepage velocity, the diffusion-dispersion coefficient and the sorption capacity. For this multi-layer system it is possible, by using simple mathematical expressions, to introduce a transformation that reduces the general and variable system geometry to an equivalent and more simple geometry where the seepage velocity becomes constant in each layer. If the seepage velocity vis constant, the advection-diffusion equation can by simplified into a purely diffusive equation by considering a Galilean transformation into a reference system moving with velocity v with respect to the laboratory system. In addition, it is simpler to take into account the boundary conditions imposed on the system in an appropriate manner. For the solution of the diffusion equation and the definition of the concentration profile in the equivalent system, we propose an algorithm based on a simple implicit finite difference (IFD) scheme in which each layer is discretized separately and the boundary conditions are derived from the continuity of flux and concentration. We calculate the concentration profiles, as functions of space and time, for some particular systems and we investigate the validity of the various analytical solutions used in the literature and of the boundary conditions adopted for their derivation. To this purpose, we perform a parametric study by varying the diffusion coefficients of the layers of a chosen equivalent system and study the behaviour of the numerical solution in some relevant cases.
 
  1. L. Demeio, D. Sani and M. Manassero, A numerical method for the solution of the diffusion equation in multilayered systems, Internal Report N. 5/1998, Dipartimento di Matematica "V. Volterra", Universita' degli Studi di Ancona.
  2. L. Demeio, M. Manassero e D. Sani, Modelli matematici degli esperimenti di trasporto di sostanze inquinanti: studio numerico dell'equazione di diffusione, Proc. IV Congresso Nazionale della SIMAI, Giardini Naxos (ME), 1-5 Giugno 1998, vol. 2, p. 320.

Back to Research Page