Experimental and Numerical Investigations of Using Nanoparticles in Groundwater Remediation

The efficiency of using zeolite nano-clay and silica sand in removing iron from groundwater is investigated experimentally as an application of in situ remediation technique using permeable reactive barrier (PRB). In the first stage of the study batch experiments were conducted on coarse silica sand and fine silica gravel to determine their iron removal efficiency from contaminated water. The results showed that the removal efficiency was about 51.90% for coarse silica sand while fine silica gravel failed to reduce the iron concentration in water. A mix of zeolite nano-clay and coarse silica sand with a ratio of 1:30 was tested. It was found that iron removal efficiency significantly increases to about 99.70%. The linear adsorption isotherm was found to be the most representative for the adsorption of iron onto coarse silica sand and the mix of nano-clay and coarse silica sand with distribution coefficients equal 0.0009 and 0.001, respectively. In the second stage a glass sand tank with coarse silica sand and a mix of zeolite nano-clay and coarse silica PRBs was constructed. The sand tank was used to investigate the effect of sampling time, head difference, iron concentration, nano-clay dosage, and thickness of permeable reactive barrier on the iron removal efficiency. The results indicated that increasing iron concentration and head difference decreases iron removal efficiency. The mix of zeolite nano-clay and coarse silica sand improves iron removal efficiency. Furthermore, increasing nano-clay dosage slightly increases the removal efficiency. However, increasing the thickness of filter layer (i.e., PRB) significantly improves the removal efficiency. Finally, the experimental model data were used to test the capability of a numerical contaminant transport model to predict the removal efficiency. The MT3DMS numerical model included within the Groundwater Modeling System (GMS) was used along with the different experimental data to obtain reaction rate values as calibration parameters for linear isotherm. Reaction rates that yield modeling results matching experimental data were obtained.