Retention effect and mode of capillary zone on the migration process of LNAPL pollutants based on experimental exploration.

Two-dimensional sand tank experiments were designed to investigate the retention process of the capillary zone during the migration of light non-aqueous phase liquid (LNAPL) pollutants. The fine sand and silt media experiments simulated the LNAPL migration process given a shallow point source leakage scenario. The results indicate that the LNAPL was retained in the capillary zone. A retention factor, based on the ratio of the change in the vertical migration velocity of the LNAPL front with time, was proposed to quantitatively characterize the retention effect. The retention factor and time satisfied the function of σ=A×exp(-kt). And the retention factor increased gradually with time, indicating the enhanced retention effect of capillary zone on the vertical migration of LNAPL. The concentration change rate was then used to investigate the LANPL redistribution process, which had a relationship with time of νc=B×ln(t)+C. The capillary zone could be divided from top to bottom into a weak retention zone (B > 0, vc < 0), a strong retention zone (B < 0), and a barrier zone (B > 0, vc > 0). The retention effect of capillary zone on LNAPL migration gradually strengthened during the vertical migration of LNAPL. In addition, the coefficient B had a relationship with the environmental factors (i.e., EC, pH, and ORP) of B=a×sin(b×α×ß×γ)c and the fitting coefficient R2 of the function was above 0.913 for both media, indicating a strong correlation between the LNAPL redistribution process and the key environmental factors.

Dispersion of silica-encapsulated DNA magnetic particles in a homogeneous sand tank.

In this study, we focused on the 3D dispersion of colloids. To our knowledge, we were the first to do so. Thereto, we injected silica encapsulated DNA tagged superparamagnetic particles (SiDNAmag) in a homogeneous coarse grain sand tank. At four downstream locations, SiDNAmag concentrations were determined as a function of time. Longitudinal and transverse dispersivity values and associated uncertainties of SiDNAmag were determined using Monte Carlo modelling approach. The parameter associated uncertainties of hydraulic conductivity as well as of the effective porosity estimated from SiDNAmag breakthrough curves were statistically similar to those estimated from salt tracer breakthrough curves. Further, the SiDNAmag dispersivity uncertainty ranges were then statistically compared with the salt tracer (NaCl, and fluorescein) dispersivities. Our results indicated that time to rise, time of peak concentration and shape of the breakthrough curves of SiDNAmag were similar to those of the salt tracer breakthrough curves. Despite the size difference between the salt tracer molecules and SiDNAmag, size exclusion did not occur, probably due to the large pore throat diameter to SiDNAmag diameter ratio. The median longitudinal dispersivity (αL) of salt tracer and SiDNAmag were 4.9 and 5.8 × 10-4 m, respectively. The median ratio of horizontal and vertical transverse dispersivities to αL, (αTH /αL and αTV /αL, respectively), for salt tracer and SiDNAmag ranged between 0.52 and 0.56. Through the statistical tests, we concluded that the longitudinal and traverse dispersivities of SiDNAmag were not statistically significantly different from salt tracer in 3 dimensions and could be used to characterize the dispersive properties of the medium we used. Our work contributes to a better understanding of 3D dispersion of SiDNAmag in saturated porous media.

Evaluation of scale-up effect on cold-active enzyme production and biodegradation tests using pilot-scale bioreactors and a 3D soil tank.

Despite recent attention being paid to the biodegradation of petroleum hydrocarbons in cold environments, scale-up studies of biodegradation are lacking. Herein, the effect of scale-up on the enzymatic biodegradation of highly contaminated soil at low temperatures was studied. A novel cold-adapted bacteria (Arthrobacter sp. S2TR-06) was isolated that could produce cold-active degradative enzymes (xylene monooxygenase (XMO) and catechol 2,3-dioxygenase (C2,3D)). Enzyme production was investigated on 4 different scales (lab to pilot scale). The results showed a shorter fermentation time, and the highest production of enzymes and biomass (107 g/L for biomass, 109 U/mL, and 203 U/mL for XMO and C2,3D after 24 h) was achieved in the 150-L bioreactor due to enhanced oxygenation. Multi-pulse injection of p-xylene into the production medium was needed every 6 h. The stability of membrane-bound enzymes can be increased up to 3-fold by adding FeSO4 at 0.1% (w/v) before extraction. Soil tests also showed that biodegradation is scale-dependent. The maximum biodegradation rate decreased from 100% at lab-scale to 36% in the 300-L sand tank tests due to limited access of enzymes to trapped p-xylene in soil pores, low dissolved oxygen in the water-saturated zone, soil heterogeneity, and the presence of the free phase of p-xylene. The result demonstrated that formulation of enzyme mixture with FeSO4 and direct injection of enzyme mixture (third scenario) can increase the efficiency of bioremediation in heterogeneous soil. In this study, it was demonstrated that cold-active degradative enzyme production can be scaled up to an industrial scale and enzymatic treatment can be used to effectively bioremediate p-xylene contaminated sites. This study could provide key scale-up guidance for the enzymatic bioremediation of mono-aromatic pollutants in water-saturated soil under cold conditions.

Experimental investigations on scale-dependent dispersivity in three-dimensional heterogeneous porous media.

The relationship between the scale-dependent dispersivity and heterogeneous sedimentary structures is investigated through conducting non-reactive tracer experiments in a three-dimensional heterogeneous sand tank. The heterogeneous porous media consists of three sedimentary facies of silty, fine, and medium sands collected from the west of the Songnen Plain, China. Moreover, several corresponding individual facies soil columns were constructed for comparison. A conservative tracer was continuously injected from an upstream source. The effective parameters were estimated by inverse modeling of a one-dimensional transport model. The results show that the scale dependence of the estimated dispersivities was discovered in the individual facies column (with relatively weaker effect) and the heterogeneous porous media (with more significant effect). With increasing transport distances, the dispersivities of the individual facies tend to reach an asymptotic value, while those of the heterogeneous media increase continuously. Furthermore, the results show that a power function can describe the relationship between effective dispersivities and transport distances. The exponent of the function is greater than one for the heterogeneous media, but less than one for the individual facies. The results also indicate that the dispersion plume is macroscopically dominated by the distribution of facies. The heterogeneity of hydraulic conductivity causes the variations of flow velocity, which further enhances the scale dependence of dispersivities. The tracer experiment in heterogeneous media provides the fundamental insight into the understanding of contaminant transport processes.

Experimental and numerical assessment of saltwater recession in coastal aquifers by constructing check dams.

Artificial freshwater recharge has been considered as a feasible and effective procedure to mitigate seawater intrusion in coastal regions. The efficiency of freshwater infiltration through a check dam reservoir on saltwater recession (SWR) is investigated using two physical models. The results demonstrate the apparent tendency of recharge freshwater to move horizontally toward the boundaries rather than flowing downward to influence saltwater wedge toe. Thereby, it would affect the saltwater wedge tip instead of its toe due to the new establishment of a positive hydraulic gradient from a dam reservoir to the boundaries. Moreover, numerical dispersive simulations have been carried out on a large-scale aquifer to find the optimum location of the dam as well as the aquifer characteristics impacts on SWR efficiency. The results show that the best location to construct a check dam is immediately above the saltwater wedge toe. It is found that when saltwater head declines, the steeper hydraulic gradient between boundaries is established and the efficiency of recharge performance will be improved. Moreover, the reduction of hydraulic conductivity in vertical direction improves SWR, while higher hydraulic conductivity in the homogeneous cases only accelerates the infiltration rate but has no meaningful effect in the long term. The considered recharge method also works better in scenarios with higher dispersivity. However, the construction of check dams on floodways might be a practical and low-cost solution but it can be concluded that as the dominant direction of the recharged freshwater is toward boundaries, it cannot promptly retreat saltwater around toe position.

Remediation of BTEX plume in a continuous flow model using zeolite-PRB.

Adsorption is a well-known phenomenon that causes the remediation of BTEX (Benzene, Toluene, Ethylbenzene, and Xylene). Zeolite is typically useful for the removal of BTEX from groundwater. In this study, the migration of the BTEX plume was investigated in a bench-scale tank model as a shallow aquifer. The objective of this research was to analyze the performance of a natural zeolite in-situ PRB remediation technique. Natural zeolite was applied as a physical permeable reactive barrier. In the first part of the experiment, 40 ml of BTEX as a contaminant was injected at the injection point (BI) into the sand tank. Samples were taken periodically via 14 boreholes for BTEX test for 23 days and analyzed using a GC-FID instrument. The results indicated high removal rates of BTEX by passing through the zeolite barrier. Zeolite barrier reduced the BTEX concentration up to 90% of the initial value. However, the barrier efficiency started to decrease after 132 h since pollution injection reached a minimum amount (%53 of the initial value) due to occupying the free space and grain pore where BTEX was adsorbed onto the surface of zeolite, thereby decreasing the barrier efficiency.