Technical Papers and Presentations

The presence of anthropogenic or natural pollutants in groundwater system poses a threat to water resources. Several remediation techniques have been explored and developed to decrease the concentration of contaminants. One such remediation technique, injection of nano-particles in the contaminated aquifers, has become promising owing to its high retention capacity (very large surface area). However, the efficiency of these nano-particles for capturing water contamination greatly depends not just on the particle characteristics, but also on the geochemistry of the aquifer. The non-linear relationship between these processes makes it difficult to understand the behavior of contaminant and nano-particles under the subsurface. Thus, to develop an optimized remediation strategy and to evaluate the efficiency of particles for contaminants treatment, a predictive tool based on mathematical models becomes significantly useful. To make such mathematically backed remediation strategy, an integrated model approach is developed in this study which addresses the different stages of remediation process.

From the numerical solution point of view, the solution is obtained by Galerkin-based finite element analysis with COMSOL Multiphysics^® simulation software. The modeling work consists of four stages involving multiphase flow, contaminant transport, nano-particle transport and kinetic degradation of contaminant. To incorporate all physics in a single multiphysics modeling framework, a coupled formulation including pre-defined interfaces and general Coefficient Form PDE interfaces were used. To validate the formulation of individual physics, two-phase flow and nano-particle transport models were developed as stand-alone models. Their results were matched with the benchmark models from literature. A 2-dimensional domain was selected to represent an aquifer at Skovlunde Byvej (Copenhagen region, Denmark) in a small scale. The model was assigned with the properties of the contaminant and nano-particle obtained from laboratory experiments. A free triangular mesh having a total number of elements of 45869 was used.

The first stage of Dense Non-aqueous Phase Liquid (DNAPL) flow simulation predicts that the DNAPL would take nearly 15 days to reach the bedrock beneath the aquifer. However, a significant part of it is predicted to become immobile in clay lenses which will act as secondary source of contamination. In the next stage of the simulation, the miscibility of DNAPL is simulated using kinetic mass transfer of contaminants from secondary source zone. The results show the persistence of steady state contaminant plume in the aquifer. In the third stage, the injection of nano-particles was modeled. With the nano-particle injection of concentration 0.5 kg/m³, the radius of influence (RI) formed by deposited particle was predicted to be 1.6 m laterally and 3 m in depth. In the final stage, the contaminant and particle interaction are modeled using pseudo-first order kinetic rate. The result shows that the contaminant within the reach of the RI is degraded while contaminant beyond it remain undegraded. From all these results as shown in Figure 1, it can be concluded that this multi-stage model has been proved to be an effective tool to anticipate the distribution of contaminant as well as efficiency of remediating particles and can be used for the optimization processes.