مجله آب و خاک (Feb 2017)
Transport Modeling of Modified Magnetite Nanoparticles with Sodium Dodecyl Sulfate in a Saturated Sandy Soil
Abstract
Introduction: Nanoparticles due to their large specific area and reactivity recently have been used in several environmental remediation applications such as degradation of organic compounds and pesticides and adsorption of heavy metals and inorganic anions. Because of concern over potential threats of nanoparticle releases into the soil–water environment, a number of studies have been carried out to investigate the transport, retention and deposition of nanoparticles in saturated porous media. Many of these studies are based on measurements of transport in columns packed with idealized porous media consisting of spherical glass beads or sand. The nanoparticles are usually introduced into the column and breakthrough curve concentrations are measured at the column outlet. To examine the effect of various parameters on the transport of nanoparticles in porous medium, for convenience, all the parameters considered the same in the experiments, and only one parameter in the experiments is changed and investigated. Materials and Methods: The objective of this research is quantitative study of modified magnetite nanoparticles transport in saturated sand-repacked columns. The modified magnetite nanoparticles with Sodium dodecyl sulfate were synthesized following the protocol described by Si et al. (2004). The experimental setup included a suspension reservoir, Teflon tubing, a HPLC pump, and a glass column (2.5 cm i.d. and 20 cm height). Therefore, breakthrough curves of modified magnetite nanoparticles with Sodium dodecyl sulfate and chloride were determined under saturated conditions and influence of nanoparticles concentration (0.1 and 0.5 g.L-1) and pore velocity (pressure head of 2 and 10 cm) on nanoparticles transport were investigated. For each medium bed, the background solution were first pumped through the column in the up-flow mode to obtain a steady flow state. Then, a tracer test was conducted by introducing CaCl2 solution into the column. The response curve was followed by analyzing the concentration history of Cl-1 in the effluent. Then, the influent was switched back to the background solution to thoroughly elute the tracer. Following the tracer test, a modified magnetite nanoparticles with sodium dodecyl sulfate was introduced into the column and the nanoparticle breakthrough curves were obtained by measuring the concentration history of total Fe in the effluent. Total iron concentration was analyzed with a flame atomic-absorption spectrophotometer. One site and two site kinetic attachment-detachment models in HYDRUS-1D software were used to predict the nanoparticles transport. Also parameters of model efficiency coefficient (E), root mean square error (RMSE), geometric mean error ratio (GMER), and geometric standard deviation of error ratio (GSDER) were used to determine the accuracy of the models. Results and Discussion: SEM measurements demonstrated that the particle size of nanoparticles was about 40-60 nm. The hydrodynamic dispersion coefficient (D) for each medium was obtained by fitting the classic 1-D convection–dispersion equation (CDE) to the experimental breakthrough data using the CXTFIT code (STANMOD software, USDA). The relative concentration of nanoparticles in comparison with chloride in the collected effluent from soil columns were much lower indicating a strong retention of nanoparticles in studied porous media, thereby attachment, deposition and possibly straining of nanoparticles. Modeling results showed that in all sites of both models (one site and two-site kinetic attachment-detachment models), attachment was rapid and detachment was slow. These attachment kinetic sites may be because of consistent charges of minerals with attachment. Therefore, considering to same attachment-detachment behavior in two sites of two-site kinetic model, it is concluded that the one site kinetic model had eligible estimation of nanoparticles breakthrough curve in the studied sandy soil columns lonely. Efficiency of one site and two-site models varied from 0.761 to 0.851 and 0.760 to 0.846 respectively that indicated both models had good estimation of nanoparticles transport in the sandy soil. Also, logarithmic form of nanoparticles breakthrough curve showed that both models had good estimation of all ranges of breakthrough curve containing its tail. Conclusion: Investigation of transport modeling of modified magnetite nanoparticles with Sodium dodecyl sulfate in a saturated sandy soil showed that decreasing the nanoparticles concentration would enhanced the mobility of modified magnetite nanoparticles, but increasing of pressure head had no effect on nanoparticles mobility. The results of models evaluation showed that both one site and two-site models had eligible estimation of nanoparticles transport in the studied sandy soil columns.
Keywords
