Transport of surface-modified iron nanoparticles through model subsurface porous media

The overall objective of this research is to evaluate significant mechanisms for deposition of surface-modified NZVI in granular subsurface media during transport. Although surface-modified NZVI have been shown to transport more easily than bare NZVI, there is a lack of knowledge of how different parameters such NZVI particle concentration, NZVI size, aqueous-phase flow velocity, and sand particle size influence nanoparticle transport. To investigate the effects of these parameters on transport, a number of laboratory experiments were conducted with NZVI synthesized from ferrous sulfate in the presence of polymers that were effective in colloidal stabilization of the particles. The bare and surface modified-NZVI was characterized for size and surface chemistry by a wide array of analytical instruments. The polymer-stabilized NZVI were employed in three different studies to identify parameters that influence deposition of NZVI in model, granular, subsurface media. In the first study, the breakthrough patterns of carboxymethyl cellulose (CMC)- and polyacrylic acid (PAA)- stabilized NZVI eluted from packed sand columns under a range of pore water velocities and NZVI influent concentrations were investigated. The NZVI effluent relative concentrations of both types of particles decreased with decreasing flow velocities and increasing particle concentrations. PAA-NZVI exhibited slower elution from the columns than CMC-NZVI under identical experimental conditions, and this is attributed to more rapid aggregation kinetics of PAA-NZVI. The second study focused on the quantitative evaluation of aggregation kinetics and the possible effects of aggregation on NZVI deposition. Aggregation of CMC-NZVI particles resulted in a change in particle size distribution (PSD) with time, and the changes in particle size were evaluated by nanoparticle tracking analyzer (NTA). The effects of particle concentrations on the transport in porous media were evaluated by comparing the time profiles of elution of CMC-NZVI from packed sand columns. Changes in PSD over time were responsible for a gradual increase in effluent concentration between 1 and 4 pore volumes, and beyond 4 pore volumes particle detachment contributed to non-steady state effluent concentrations. The NZVI elution profiles had a good fit with aggregation kinetics equations coupled to colloid transport equations that account for particle deposition and detachment. The third study focused on assessing the significance of straining of CMC-NZVI particles during transport in model subsurface porous media. Laboratory experiment were conducted to assess the transport of CMC-NZVI in columns packed with four different sized sands and with three different concentrations. . Breakthrough curves (BTC) and retention profiles of CMC-NZVI along the column length were analyzed to characterize CMC-NZVI transport. The breakthrough curves suggest that with decrease in mean sand diameter, the effluent concentrations decrease. Very high CMC-NZVI particle retention towards the inlet, particularly for the finer sands was observed. These observations are consistent with particle retention in porous media due to straining and/or wedging. Two colloid transport models considering 1) particle deposition by attachment only, and 2) particle retention by straining along with particle deposition by attachment were fitted to the experimental data. Comparison of experimental data and the model calculations suggest that in addition to deposition on collector surface, CMC-NZVI particles are removed from the solution by straining in packed sand beds, with straining rate coefficients that decrease with increase in sand diameter.