Cesium removal from a water system using a polysulfone carrier containing nitric acid-treated bamboo charcoal.

Laboratory scale sorption and desorption experiments were performed to investigate the cesium (Cs) removal efficiency of a bead-shaped polysulfone carrier containing HNO3-treated bamboo charcoal (BC). The average Cs removal efficiency of BC only and of polysulfone carrier without BC after 1 h sorption reaction was 53 and 18%, respectively. However, the Cs removal efficiency for the polysulfone carrier with 5% HNO3-treated BC (P-5N-BC) after 1 h and 24 h reaction was 66 and 98%, respectively. The Cs removal efficiency after 24 h reaction remained >85% over a wide range of pH and temperature conditions, suggesting that using P-5N-BC as the Cs adsorbent is feasible in a variety of aquatic environments. The maximum Cs sorption capacity (qm) of P-5N-BC, as calculated from a Langmuir isotherm model, was 60.9 mg/g, which is much higher than those of other adsorbents from previous studies for 1 h of sorption time. The Cs desorption rate of P-5N-BC for 24 h desorption time was <17%, showing that the Cs was stably enough attached to the HNO3-treated BC for long-term use. The results of continuous column experiments showed that the total amount of treated water from the column packed with P-5N-BC increased more than nine times when compared with that from the only BC-granule-packed column. The P-5N-BC maintained more than 68% Cs removal efficiency after 90 pore volumes of flushing, suggesting that only 15 g of P-5N-BC (with only 0.75 g of HNO3-treated BC) could clean 5 L of Cs-contaminated water (initial Cs concentration: 1.0 mg/L; effluent concentration: < 0.09 mg/L). The present results demonstrate that P-5N-BC has remarkable potential for removal of Cs from diverse water systems.

Diffusion of solutes from depleting sources into and out of finite low-permeability zones.

Two important factors that affect groundwater contaminant persistence are the temporal pattern of contaminant source depletion and solute diffusion into and out of aquitards. This study provides a framework to evaluate the relative importance of these effects on contaminant persistence, with emphasis on the importance of thin aquitards. We developed one-dimensional (1D) analytical solutions for forward and back diffusion in a finite domain with a no flux boundary using the method of images and demonstrated their applicability to measured data from three well-controlled laboratory diffusion experiments with exponentially depleting sources. We used both in situ aquitard solute concentrations and aquifer breakthrough curves for sorbing and non-sorbing solutes. The finite-domain no flux boundary solutions showed better agreement with measured data than was available with semi-infinite approaches, with increasing discrepancy for dimensionless relative diffusion length scale beyond a critical threshold value (Zd > 0.7). We also used a mass balance to demonstrate that the temporal pattern of contaminant source depletion controls the duration of solute mass accumulation in the aquitard, as well as the total solute mass release back into the aquifer. Lower rates of source depletion result in a longer period of mass accumulation in the aquitard and later back diffusion initiation time. The amount of solute mass stored in the aquitard increases with longer loading duration, thereby contributing to overall longer contaminant persistence in aquifers. This study entails widespread implications for anthropogenic waste and contamination sites, which are all dependent on efficient and cost-effective contaminant management strategies.

Field-scale forward and back diffusion through low-permeability zones.

Understanding the effects of back diffusion of groundwater contaminants from low-permeability zones to aquifers is critical to making site management decisions related to remedial actions. Here, we combine aquifer and aquitard data to develop recommended site characterization strategies using a three-stage classification of plume life cycle based on the solute origins: aquifer source zone dissolution, source zone dissolution combined with back diffusion from an aquitard, and only back diffusion. We use measured aquitard concentration profile data from three field sites to identify signature shapes that are characteristic of these three stages. We find good fits to the measured data with analytical solutions that include the effects of advection and forward and back diffusion through low-permeability zones, and linearly and exponentially decreasing flux resulting from source dissolution in the aquifer. Aquifer contaminant time series data at monitoring wells from a mature site were well described using analytical solutions representing the combined case of source zone and back diffusion, while data from a site where the source had been isolated were well described solely by back diffusion. The modeling approach presented in this study is designed to enable site managers to implement appropriate remediation technologies at a proper timing for high- and low-permeability zones, considering estimated plume life cycle.

Solute source depletion control of forward and back diffusion through low-permeability zones.

Solute diffusive exchange between low-permeability aquitards and high-permeability aquifers acts as a significant mediator of long-term contaminant fate. Aquifer contaminants diffuse into aquitards, but as contaminant sources are depleted, aquifer concentrations decline, triggering back diffusion from aquitards. The dynamics of the contaminant source depletion, or the source strength function, controls the timing of the transition of aquitards from sinks to sources. Here, we experimentally evaluate three archetypical transient source depletion models (step-change, linear, and exponential), and we use novel analytical solutions to accurately account for dynamic aquitard-aquifer diffusive transfer. Laboratory diffusion experiments were conducted using a well-controlled flow chamber to assess solute exchange between sand aquifer and kaolinite aquitard layers. Solute concentration profiles in the aquitard were measured in situ using electrical conductivity. Back diffusion was shown to begin earlier and produce larger mass flux for rapidly depleting sources. The analytical models showed very good correspondence with measured aquifer breakthrough curves and aquitard concentration profiles. The modeling approach links source dissolution and back diffusion, enabling assessment of human exposure risk and calculation of the back diffusion initiation time, as well as the resulting plume persistence.

Uranium and cesium accumulation in bean (Phaseolus vulgaris L. var. vulgaris) and its potential for uranium rhizofiltration.

Laboratory scale rhizofiltration experiments were performed to investigate uranium and cesium accumulation in bean (Phaseolus vulgaris L. var. vulgaris) and its potential for treatment of uranium contaminated groundwater. During 72 h of rhizofiltration, the roots of the bean accumulated uranium and cesium to concentrations 317-1019 times above the initial concentrations, which ranged from 100 to 700 µg l(-1) in artificially contaminated solutions. When the pH of the solution was adjusted to 3, the ability to accumulate uranium was 1.6 times higher than it was for solutions of pH 7 and pH 9. With an initial uranium concentration of 240 µg l(-1) in genuine groundwater at pH 5, the bean reduced the uranium concentration by 90.2% (to 23.6 µg l(-1)) within 12 h and by 98.9% (to 2.8 µg l(-1)) within 72 h. A laboratory scale continuous clean-up system reduced uranium concentrations from 240 µg l(-1) to below 10 µg l(-1) in 56 h; the whole uranium concentration in the bean roots during system operation was more than 2600 µg g(-1) on a dry weight basis. Using SEM and EDS analyses, the uranium removal in solution at pH 7 was determined based on adsorption and precipitation on the root surface in the form of insoluble uranium compounds. The present results demonstrate that the rhizofiltration technique using beans efficiently removes uranium and cesium from groundwater as an eco-friendly and cost-effective method.

Back diffusion from thin low permeability zones.

Aquitards can serve as long-term contaminant sources to aquifers when contaminant mass diffuses from the aquitard following aquifer source mass depletion. This study describes analytical and experimental approaches to understand reactive and nonreactive solute transport in a thin aquitard bounded by an adjacent aquifer. A series of well-controlled laboratory experiments were conducted in a two-dimensional flow chamber to quantify solute diffusion from a high-permeability sand into and subsequently out of kaolinite clay layers of vertical thickness 15 mm, 20 mm, and 60 mm. One-dimensional analytical solutions were developed for diffusion in a finite aquitard with mass exchange with an adjacent aquifer using the method of images. The analytical solutions showed very good agreement with measured breakthrough curves and aquitard concentration distributions measured in situ by light reflection visualization. Solutes with low retardation accumulated more stored mass with greater penetration distance in the aquitard compared to high-retardation solutes. However, because the duration of aquitard mass release was much longer, high-retardation solutes have a greater long-term back diffusion risk. The error associated with applying a semi-infinite domain analytical solution to a finite diffusion domain increases as a function of the system relative diffusion length scale, suggesting that the solutions using image sources should be applied in cases with rapid solute diffusion and/or thin clay layers. The solutions presented here can be extended to multilayer aquifer/low-permeability systems to assess the significance of back diffusion from thin layers.

Light reflection visualization to determine solute diffusion into clays.

Light reflection visualization (LRV) experiments were performed to investigate solute diffusion in low-permeability porous media using a well-controlled two-dimensional flow chamber with a domain composed of two layers (one sand and one clay). Two different dye tracers (Brilliant Blue FCF and Ponceau 4R) and clay domains (kaolinite and montmorillonite) were used. The images obtained through the LRV technique were processed to monitor two-dimensional concentration distributions in the low-permeability zone by applying calibration curves that related light intensity to equilibrium concentrations for each dye tracer in the clay. One dimensional experimentally-measured LRV concentration profiles in the clay were found to be in very good agreement with those predicted from a one-dimensional analytical solution, with coefficient of efficiency values that exceeded 0.97. The retardation factors (R) for both dyes were relatively large, leading to slow diffusive penetration into the clays. At a relative concentration C/C0=0.1, Brilliant Blue FCF in kaolinite (R=11) diffused approximately 10 mm after 21 days of source loading, and Ponceau 4R in montmorillonite (R=7) diffused approximately 12 mm after 23 days of source loading. The LRV experimentally-measured two-dimensional concentration profiles in the clay were also well described by a simple analytical solution. The results from this study demonstrate that the LRV approach is an attractive non-invasive tool to investigate the concentration distribution of dye tracers in clays in laboratory experiments.

Perchlorate in soybean sprouts (Glycine max L. Merr.), water dropwort (Oenanthe stolonifera DC.), and lotus (Nelumbo nucifera Gaertn.) root in South Korea.

The occurrence of perchlorate in soybean sprouts (Glycine max L. Merr), water dropwort (Oenanthe stolonifera DC.), and lotus (Nelumbo nucifera Gaertn.) root, which are commonly consumed by people in South Korea, was determined by using an ion chromatograph coupled with a tandem mass spectrometer. For soybean sprouts (11 samples), perchlorate was detected in most (91%) of the samples at various concentrations of up to 78.4 µg/kg dry weight (DW); the mean concentration was 35.2 µg/kg DW. For water dropwort, of the 13 samples examined, four showed concentrations that were above the limit of quantification (LOQ). The mean perchlorate concentration was 20.7 µg/kg DW, and the highest perchlorate value was 39.9 µg/kg DW. Of the six lotus root samples examined, only one exhibited a detectable perchlorate concentration (17.3 µg/kg DW). For the accumulation experiments with artificially contaminated solutions, the concentrations of perchlorate in soybean sprouts gradually increased with the increase of perchlorate concentration in the solution. However, there was a decrease in the bioconcentration factor as the perchlorate concentration in the solution increased.

Rhizofiltration using sunflower (Helianthus annuus L.) and bean (Phaseolus vulgaris L. var. vulgaris) to remediate uranium contaminated groundwater.

The uranium removal efficiencies of rhizofiltration in the remediation of groundwater were investigated in lab-scale experiments. Sunflower (Helianthus annuus L.) and bean (Phaseolus vulgaris L. var. vulgaris) were cultivated and an artificially uranium contaminated solution and three genuine groundwater samples were used in the experiments. More than 80% of the initial uranium in solution and genuine groundwater, respectively, was removed within 24h by using sunflower and the residual uranium concentration of the treated water was lower than 30 microg/L (USEPA drinking water limit). For bean, the uranium removal efficiency of the rhizofiltration was roughly 60-80%. The maximum uranium removal via rhizofiltration for the two plant cultivars occurred at pH 3-5 of solution and their uranium removal efficiencies exceeded 90%. The lab-scale continuous rhizofiltration clean-up system delivered over 99% uranium removal efficiency, and the results of SEM and EDS analyses indicated that most uranium accumulated in the roots of plants. The present results suggested that the uranium removal capacity of two plants evaluated in the clean-up system was about 25mg/kg of wet plant mass. Notably, the removal capacity of the root parts only was more than 500 mg/kg.