Systematic Performance Evaluation of Reinforcement Learning Algorithms Applied to Wastewater Treatment Control Optimization.

Treatment of wastewater using activated sludge relies on several complex, nonlinear processes. While activated sludge systems can provide high levels of treatment, including nutrient removal, operating these systems is often challenging and energy intensive. Significant research investment has been made in recent years into improving control optimization of such systems, through both domain knowledge and, more recently, machine learning. This study leverages a novel interface between a common process modeling software and a Python reinforcement learning environment to evaluate four common reinforcement learning algorithms for their ability to minimize treatment energy use while maintaining effluent compliance within the Benchmark Simulation Model No. 1 (BSM1) simulation. Three of the algorithms tested, deep Q-learning, proximal policy optimization, and synchronous advantage actor critic, generally performed poorly over the scenarios tested in this study. In contrast, the twin delayed deep deterministic policy gradient (TD3) algorithm consistently produced a high level of control optimization while maintaining the treatment requirements. Under the best selection of state observation features, TD3 control optimization reduced aeration and pumping energy requirements by 14.3% compared to the BSM1 benchmark control, outperforming the advanced domain-based control strategy of ammonia-based aeration control, although future work is necessary to improve robustness of RL implementation.

Modeling benzene permeation through drinking water high density polyethylene (HDPE) pipes.

Organic compounds such as benzene, toluene, ethyl benzene and o-, m-, and p-xylene from contaminated soil and groundwater may permeate through thermoplastic pipes which are used for the conveyance of drinking water in water distribution systems. In this study, permeation parameters of benzene in 25 mm (1 inch) standard inside dimension ratio (SIDR) 9 high density polyethylene (HDPE) pipes were estimated by fitting the measured data to a permeation model based on a combination of equilibrium partitioning and Fick's diffusion. For bulk concentrations between 6.0 and 67.5 mg/L in soil pore water, the concentration-dependent diffusion coefficients of benzene were found to range from 2.0×10(-9) to 2.8×10(-9) cm2/s while the solubility coefficient was determined to be 23.7. The simulated permeation curves of benzene for SIDR 9 and SIDR 7 series of HDPE pipes indicated that small diameter pipes were more vulnerable to permeation of benzene than large diameter pipes, and the breakthrough of benzene into the HDPE pipe was retarded and the corresponding permeation flux decreased with an increase of the pipe thickness. HDPE pipes exposed to an instantaneous plume exhibited distinguishable permeation characteristics from those exposed to a continuous source with a constant input. The properties of aquifer such as dispersion coefficients (DL) also influenced the permeation behavior of benzene through HDPE pipes.

Organic semiconductor wastewater treatment using a four-stage Bardenpho with membrane system.

Electronic wastewater from a semiconductor plant was treated with a pilot-scale four-stage Bardenpho process with membrane system. The system was operated over a 14-month period with an overall hydraulic retention time (HRT) ranging from 9.5 to 30 h. With a few exceptions, the pilot plant consistently treated the electronic wastewater with an average removal efficiency of chemical oxygen demand (COD) and total nitrogen of 97% and 93%, respectively, and achieving effluent quality of COD<15 mg/L, turbidity<1, and silt density index<1. Based on removal efficiencies of the pilot plant, it is possible to lower the HRT to less than 9.5 h to achieve comparable removal efficiencies. An energy-saving configuration where an internal recycle line was omitted and the biomass recycle was rerouted to the pre-anoxic tank, can reduce energy consumption by 8.6% and gave removal efficiencies that were similar to the Bardenpho process. The system achieved pre-anoxic and post-anoxic specific denitrification rate values with a 95% confidence interval of 0.091 ± 0.011 g NO3-N/g MLVSS d and 0.087 ± 0.016 g NO3-N/g MLVSS d, respectively. The effluent from the four-stage Bardenpho with membrane system can be paired with a reverse osmosis system to provide further treatment for reuse purposes.

Opportunity for a greener recovery of dysprosium(III) from secondary sources by a novel Mannich reaction-modified phosphorylated chitosan hydrogel.

The depleting supply of natural sources of rare earth elements (REE) is a concern to many nations as demand for advanced technology is becoming vital for national security. In this communication, the recovery of dysprosium(III) from aqueous systems was exemplified by a modified phosphorylated chitosan (PCs/MB) prepared by the C-Mannich reaction of phosphorylated chitosan, glutaraldehyde, and 4-hydroxycoumarin in ethanolic solution. Batch adsorption studies achieved a maximum adsorption capacity (qmax) of 34 mg/g at 25 °C and pH = 5.4 for 2 h. Fourier Transform-Infrared Spectroscopy, elemental mapping, and quantitative analyses revealed ion-exchange mechanism with C6-phosphate and a synergistic complexation with the amino group between two hexose units of the chitosan chain confirming the correlation provided by the pseudo-second order kinetics (R2 = 0.9996), extrapolated mean free energy of adsorption (Eads) of 12.9 kJ/mol from the corrected Dubinin-Radushkevich isotherm, and the extrapolated enthalpy of adsorption (ΔH0ads) of -42.4 kJ/mol from the linearized Van't Hoff plot. Competitive adsorption with iron(II), cerium(III), and neodymium(III) demonstrated preferential removal of dysprosium(III) and complete exclusion of iron(II), which illustrates potential application in the separation of REE from electronic wastes.

Unified control of diverse actions in a wastewater treatment activated sludge system using reinforcement learning for multi-objective optimization.

The operation of modern wastewater treatment facilities is a balancing act in which a multitude of variables are controlled to achieve a wide range of objectives, many of which are conflicting. This is especially true within secondary activated sludge systems, where significant research and industry effort has been devoted to advance control optimization strategies, both domain-driven and data-driven. Among data-driven control strategies, reinforcement learning (RL) stands out for its ability to achieve better than human performance in complex environments. While RL has been applied to activated sludge process optimization in existing literature, these applications are typically limited in scope, and never for the control of more than three actions. Expanding the scope of RL control has the potential to increase the optimization potential while concurrently reducing the number of control systems that must be tuned and maintained by operations staff. This study examined several facets of the implementation of multi-action, multi-objective RL agents, namely how many actions a single agent could successfully control and what extent of environment data was necessary to train such agents. This study observed improved control optimization with increasing action scope, though control of waste activated sludge remains a challenge. Furthermore, agents were able to maintain a high level of performance under decreased observation scope, up to a point. When compared to baseline control of the Benchmark Simulation Model No. 1 (BSM1), an RL agent controlling seven individual actions improved the average BSM1 performance metric by 8.3 %, equivalent to an annual cost savings of $40,200 after accounting for the cost of additional sensors.

Degradation of trichloroethylene using iron, bimetals and trimetals.

A cold, electrodeless method was used to prepare bimetals (Fe/Cu, Fe/Ni) and trimetals (Fe/Cu/Ni) for the treatment of trichloroethylene (TCE). With Fe/Cu, the degradation of TCE was observed to increase with increasing copper content up to 9.26 % (w/w) with a first-order degradation rate constant approximately 10 times faster than that of zero-valent iron (ZVI) alone. For copper content greater than 9.26 %, the TCE degradation rate decreased. Dechlorinated compounds were initially observed but they were transitory and accounted for no more than 9 % of initial TCE mass on a carbon molar basis. Ethylene was the primary end product of TCE reduction. Similarly for Fe/Ni, increasing rates of degradation were observed with increasing amounts of nickel with a maximum degradation rate constant of about 30 times higher than that of ZVI alone. However, the amount of nickel needed to reach the maximum rate was only 0.25 %. When copper and nickel were plated onto iron, the maximum reaction rate constant was approximately 50 times higher than that of ZVI. The maximum degradation of TCE was observed for a copper and nickel content of 4.17 % and 0.40 %, respectively. The experimental results indicated that TCE degradation was enhanced by more than one order of magnitude when copper and/or nickel was plated onto the zero-valent iron. However, copper or nickel plated onto iron by the elctrodeless process was found to leach out during the reaction which may, in turn, impact the contaminated water.

Biological phosphorus removal and its microbial community in a modified full-scale activated sludge system under dry and wet weather dynamics.

Enhanced biological phosphorus removal (EBPR) performance and microbial community dynamics during dry and wet-weather conditions of a full-scale treatment plant was evaluated by converting a section of activated sludge basins using low-cost operational modifications into an anoxic/anaerobic zone to promote EBPR. Two trains of the activated sludge system at the Des Moines, Iowa Metropolitan Wastewater Reclamation Facility were used for the study with one train modified for EBPR, and the other remained as nitrification-only for comparison. In addition to measuring the modification effectiveness for phosphorus removal, performance was compared during dry and wet weather conditions over the course of two summer seasons to improve understanding of wet and dry weather dynamics for EBPR. DNA sequencing and qPCR tests were conducted to develop an understanding of microbial population changes between control and modified basins and wet and dry weather conditions. Basin hydraulic retention times varied from 2.6 to 12.7 hours with an average of 8.9 hours. EBPR activity was successfully established in the modified basins with average phosphorus content of the return activated sludge 0.032 ± 0.002 compared to 0.016 ± 0.001 mg TP/mg TSS (95% confidence) in the control basins. Phosphorus removal was significantly decreased by prolonged wet weather conditions, particularly in year two of the study, however the modified basin maximum removal of 96% and average of 43.7 ± 5.3% remained significantly higher than the maximum of 46% and average 12.6 ± 2.4% removal in the control basins. DNA sequencing showed a significant increase in relative abundance of phyla Chloroflexi, Nitrospirae, and Verrucomicrobia in the modified basins, but no correlation to EBPR performance. qPCR indicated significant increase in relative quantity of Accumulibacter, but not for Actinetobacter-like phosphorus accumulating organisms (PAOs), which includes the PAO Tetrasphaera. Significant abundance of some Accumulibacter clades found within the modified basins was contrary to previous literature which focused on small-scale and batch studies. A higher than expected dominance of clade I and increased relative quantities of clades IIB and IIC during extended wet weather was observed which may have contributed to rapid recovery of phosphorus removal when dry weather resumed. The abundance of PAOs did not significantly correlate with changes in phosphorous removal performance, contrary to reports from previous small-scale and batch studies.

Effect of pH on transport of Pb2+, Mn2+, Zn2+ and Ni2+ through lateritic soil: column experiments and transport modeling.

This study investigated the effects of pH on the transport of Pb2+, Mn2+, Zn2+ and Ni2+ through lateritic soil columns. Model results by fitting the symmetric breakthrough curves (BTCs) of bromide (Br-) with CXTFIT model suggested that physical non-equilibrium processes were absent in the columns. The heavy metal BTCs were, however, asymmetrical and exhibited a tailing phenomenon, indicating the presence of chemical non-equilibrium processes in the columns. The retardation factors of Pb2+ were the largest of the four metal ions at both pH 4.0 (33.3) and pH 5.0 (35.4). The use of Langmuir isotherm parameters from batch studies with HYDRUS-1D did not predict the BTCs well. Rather the two-site model (TSM) described the heavy metal BTCs better than the equilibrium linear/nonlinear Langmuir model. The fraction of instantaneous sorption sites (f) of all four metal ions on the lateritic soil was consistently about 30%-44% of the total sorption sites.

Temperature effects on nitrification in polishing biological aerated filters (BAFs).

The effects of temperature on nitrification in a polishing biological aerated filter (BAF) were investigated using a 75-mm diameter pilot-scale BAF with a gravel media size of 5 mm and a depth of 1.7 m. Influent soluble chemical oxygen demand (sCOD) and ammonia-nitrogen (NH3-N) concentrations were approximately 50 mg/L and 25 mg/L simulating the effluent from an aerated lagoon system. For an influent wastewater temperature of 6.5 degrees C, approximately 95% of NH3-N was nitrified at a hydraulic retention time (HRT) of 2 hours. By recirculating 200% of the effluent back into the BAF for a HRT of 1 hour and at 6.5 degrees C, NH3-N percentage removal improved from 54% to 92%. For NH3-N loading larger than 0.9 kg NH3-N/m3-day at 24 degrees C, the mass of NH3-N removed in kg NH3-N/ m3-day reached an asymptotic value of 0.63 kg NH3-N/m3-day. The NH3-N concentrations within the column at different temperatures were modelled using zero-order biotransformation rate kinetics. The results showed that gravel BAF operating at an HRT of 1 hour with 100% or 200% recirculation can be used as an add-on technology for nitrification for cold weather conditions.

Biodegradation of 17alpha-methyltestosterone and isolation of MT-degrading bacterium from sediment of Nile tilapia masculinization pond.

The fast growing and highly tolerant fish Nile tilapia is one of the most commonly raised fish in the aquaculture industry. To produce an all-male population, a common practice is to feed the Nile tilapia fry with 17alpha-methyltestosterone (MT)-impregnated food. Uneaten fish food with MT may accumulate in the masculinization ponds and be released into the receiving waters. Not much is known about the fate of MT in the fish farms and in the receiving streams. The objective of this study is to investigate the biodegradation of MT under aerobic condition and to isolate responsible microorganisms. Aerobic biodegradation tests were conducted with MT concentrations of 0.3, 1.0, 5.0, 7.0, and 10.0 mg/L using sediment from the masculinization pond as microbial seed. The results suggested that MT is biodegradable. Lag phase was not observed in all cases. With initial concentrations of 0.3, 1.0, 5.0, 7.0, and 10.0 mg/l, the first-order degradation rates were 0.52, 0.23, 0.17, 0.13 and 0.10 day(-1), respectively. Degradation rates were found to decrease with an increase in the initial MT concentration. Analysis of 16S rRNA gene sequences of a strain isolated from the sediment indicated that the strain was highly similar to Pimelobacter simplex strain S151 (100%) which is in the genus Nocardioidaceae. Using this strain, MT is degraded with a first-order degradation rate of 0.044 h(-1) excluding the lag phase. This is the first work reporting biodegradation of MT and isolation of MT-degrading bacterium from environment.

Comparison of recirculation configurations for biological nutrient removal in a membrane bioreactor.

A 12-L lab-scale membrane bioreactor (MBR), consisting of an anaerobic and anoxic compartment followed by an oxic plate-frame membrane compartment, was evaluated for carbonaceous and nutrient removals by varying the recirculation of mixed liquor and permeate. The hydraulic retention times (HRTs) for the anaerobic, anoxic, and oxic compartments were 2, 2, and 8h, respectively. The solids residence time (SRT) for the oxic compartment was 25 days. Five different recirculation configurations were tested by recirculating mixed liquor and/or permeate recirculation equal to the influent flow rate (identified as 100%) into different locations of the anaerobic and anoxic compartments. Of the five configurations, the configuration with 100% mixed liquor recirculation to the anaerobic compartment and 100% permeate recirculation to the anoxic compartment gave the highest percentage removal with an average 92.3+/-0.5% soluble chemical oxygen demand (sCOD), 75.6+/-0.4% total nitrogen (TN), and 62.4+/-1.3% total phosphorus (TP) removal. When the mixed liquor and permeate recirculation rates were varied for the same configuration, the highest TP removal was obtained for 300% mixed liquor recirculation and 100% permeate recirculation (300%/100%) with a TP removal of 88.1+/-1.3% while the highest TN removal (90.3+/-0.3%) was obtained for 200%/300% recirculation. TN and TP concentrations as low as 4.2+/-0.1 and 1.4+/-0.2mg/L respectively were obtained. Mass loading rates were generally low in the range of 0.11-0.22kgCOD/kgMLSS/d due to high biomass concentrations within the oxic reactor (approx. 8000mg/L). The BioWin model was calibrated against one set of the experimental data and was found to predict the experimental data of effluent TN, TP, and NO(3)(-)-N but over-predicted sCOD and NH(3)-N for various recirculation rates. The anoxic heterotrophic yield for the calibrated model was 0.2kg biomass COD/kg COD utilized while the maximum growth rates were found to be 0.45day(-1) for mu(max-autotroph), 3.2day(-1) for mu(max-heterotroph), and 1.5day(-1) for mu(max-PAO).

Mass transfer of VOCs in laboratory-scale air sparging tank.

Volatilization of VOCs was investigated using a 55-gal laboratory-scale model in which air sparging experiments were conducted with a vertical air injection well. In addition, X-ray imaging of an air sparging sand box showed air flows were in the form of air bubbles or channels depending on the size of the porous media. Air-water mass transfer was quantified using the air-water mass transfer coefficient which was determined by fitting the experimental data to a two-zone model. The two-zone model is a one-dimensional lumped model that accounts for the effects of air flow type and diffusion of VOCs in the aqueous phase. The experimental air-water mass transfer coefficients, KGa, obtained from this study ranged from 10(-2) to 10(-3)1/min. From a correlation analysis, the air-water mass transfer coefficient was found to be directly proportional to the air flow rate and the mean particle size of soil but inversely proportional to Henry's constant. The correlation results implied that the air-water mass transfer coefficient was strongly affected by the size of porous media and the air flow rates.

Modification of a full-scale sequencing batch reactor operational mode for biological nutrient removal.

Two biological nutrient removal modes, consisting of anaerobic, anoxic, and oxic sequences, were tested in a full-scale sequencing batch reactor. The modes, identified as BNR-S1 and BNR-S2, had average total nitrogen removals of 84 and 89%, respectively, for the months of August to October. Over the same period, total phosphorus removals for BNR-S1 and BNR-S2 were 88 and 87%, respectively. In contrast, total nitrogen and total phosphorus removals for the regular aerobic mode were 54.7 and 44.7%, respectively. When the wastewater temperature changed from approximately 20 to 15 degrees C in the winter months, total nitrogen and total phosphorus removals for BNR-S2 were reduced to 81 and 70%, respectively. Total nitrogen effluent concentrations were between 2.5 and 4 mg-N/L (at approximately 20 degrees C), while the effluent total phosphorus concentrations were between 1 and 2 mg/L. The BNR-S2 mode was found to require less energy per kilogram of soluble chemical oxygen demand removed than the regular and BNR-S1 modes.

Mineralization of PAHs in coal-tar impacted aquifer sediments and associated microbial community structure investigated with FISH.

The microbial community structure and mineralization of polycyclic aromatic hydrocarbons (PAHs) in a coal-tar contaminated aquifer were investigated spatially using fluorescence in situ hybridization (FISH) and in laboratory-scale incubations of the aquifer sediments. DAPI-detected microbial populations in the contaminated sediments were three orders of magnitude greater than nearby uncontaminated sediments, suggesting growth on coal-tar constituents in situ. Actinobacteria, beta- and gamma-Proteobacteria, and Flavobacteria dominated the in situ aerobic (>1 mg l(-1) dissolved oxygen) microbial community, whereas sulfate-reducing bacteria comprised 37% of the microbial community in the sulfidogenic region of the aquifer. Rapid mineralization of naphthalene and phenanthrene were observed in aerobic laboratory microcosms and resulted in significant enrichment of beta- and gamma-Proteobacteria potentially explaining their elevated presence in situ. Firmicutes, Flavobacteria, alpha-Proteobacteria, and Actinobacteria were also enriched in the mineralization assays, but to a lesser degree. Nitrate- and sulfate-limited mineralization of naphthalene in laboratory microcosms occurred to a small degree in aquifer sediments from locations where groundwater chemistry indicated nitrate- and sulfate-reduction, respectively. Some iron-limited mineralization of naphthalene and phenanthrene was also observed in sediments originating near groundwater measurements of elevated ferrous iron. The results of this study suggest that FISH may be a useful tool for providing a much needed link between laboratory microcosms and groundwater measurements made in situ necessary to better demonstrate the potential for natural attenuation at complex PAH contaminated sites.

Assessment of intrinsic bioremediation of a coal-tar-affected aquifer using Two-dimensional reactive transport and Biogeochemical mass balance approaches.

Expedited site characterization and groundwater monitoring using direct-push technology and conventional monitoring wells were conducted at a former manufactured gas plant site. Biogeochemical data and heterotrophic plate counts support the presence of microbially mediated remediation. By superimposing solutions of a two-dimensional reactive transport analytical model, first-order degradation rate coefficients ((day-1) ) of various compounds for the dissolved-phase plume were estimated (i.e., benzene [0.0084], naphthalene [0.0058], and acenaphthene [0.0011]). The total mass transformed by aerobic respiration, nitrate reduction, and sulfate reduction around the free-phase coal-tar dense-nonaqueous-phase-liquid region and in the plume was estimated to be approximately 4.5 kg/y using a biogeochemical mass-balance approach. The total mass transformed using the degradation rate coefficients was estimated to be approximately 3.6 kg/y. Results showed that a simple two-dimensional analytical model and a biochemical mass balance with geochemical data from expedited site characterization can be useful for rapid estimation of mass-transformation rates.

Phosphorus (P) recovery coupled with increasing influent ammonium facilitated intracellular carbon source storage and simultaneous aerobic phosphorus & nitrogen removal.

Under decreasing C/N (from 8.8 to 3.5) conditions, an alternating anaerobic/aerobic biofilter (AABF) was used to remove nitrogen and accumulate/recover phosphorus (P) from synthetic wastewater. The AABF was periodically (every 10 days) fed with an additional carbon source (10 L, chemical oxygen demand (COD) = 900 mg L-1 sodium acetate (NaAC) solution) in the anaerobic phase to induce the release of P sequestered in the biofilm. An increase in PHA storage in the biofilm was observed and characterized with TEM and a GC-MS method. The accumulation of P and removal of total nitrogen occurred primarily in the aerobic phase. As the NH4+-N loading rate increased from 0.095 to 0.238 kg m-3 d-1 at a total empty bed retention time (EBRT) of 4.6 h, the TN removal in AABF was reduced from 91.2% to 43.4%, while the P removal or recovery rate remained unaffected. The high-throughput community sequencing analysis indicated that the relative abundance of Candidatus Competibacter, Nitrospira and Arcobacter increased while the Accumulibacter phosphatis decreased with an increase of ammonium loading rate within a short operational period (30 days). A putative N and P removal pattern via simultaneous nitrification and PHA-based denitrification, as well as P accumulation in the biofilm was proposed. The research demonstrated that an efficient N removal and P recovery process, i.e., simultaneous nitrification and denitrification, P accumulation and carbon source-regulated P recovery can be achieved by the symbiotic functional groups in a single biofilm reactor.

Enhanced phosphorus recovery and biofilm microbial community changes in an alternating anaerobic/aerobic biofilter.

The operation of an alternating anaerobic/aerobic biofilter (AABF), treating synthetic wastewater, was modified to enhance recovery of phosphorus (P). The AABF was periodically fed with an additional carbon source during the anaerobic phase to force the release of biofilm-sequestered P which was then harvested and recovered. A maximum of 48% of the total influent P was found to be released in the solution for recovery. Upon implementation of periodic P bio-sequestering and P harvesting, the predominant bacterial communities changed from ß-Proteobacteria to γ-Proteobacteria groups. The genus Pseudomonas of γ-Proteobacteria was found to enrich greatly with 98% dominance. Dense intracellular poly-P granules were found within the cells of the biofilm, confirming the presence of P accumulating organisms (PAOs). Periodic addition of a carbon source to the AABF coupled with intracellular P reduction during the anaerobic phase most probably exerted environmental stress in the selection of Pseudomonas PAOs over PAOs of other phylogenic types. Results of the study provided operational information on the selection of certain microbial communities for P removal and recovery. This information can be used to further advance P recovery in biofilm systems such as the AABFs.

Sorption of tylosin onto swine manure.

Sorption of tylosin was conducted on manure solids (<2 mm) and colloidal materials (<1.2 microm) collected from open (OL) and covered (CL) anaerobic swine manure lagoons. The aqueous concentration of tylosin in the sorption studies bracket the levels expected in lagoons, between 1 mgl(-1) and 30 mgl(-1). Sorption isotherms were found to be slightly non-linear for 2 mm solids, with Freundlich distribution coefficients (K(f)) of 39.4 with n=1.32 for CL slurry and 99.5 with n=1.02 for OL. These values are comparable to those reported for loam soils, but higher than those reported for sandy or clay soils and lower than those reported for fresh manure. Normalization of K(d) to the organic carbon content of the solids gave K(oc) values of 570 lkg(-1) and 818 lkg(-1), for CL and OL solids, respectively. The K(d) and K(f) values were not significantly different between colloids and 2 mm solids in OL slurry, but were significantly different in CL due to the non-linearity of the colloid isotherm. Based on the K(d) values obtained and comparing the K(d) values of other antibiotics, tylosin is strongly sorbed to manure, and would be more mobile than tetracyclines, but less mobile than sulfonamides, olaquindox, and chloramphenicol. However, tylosin mobility may be facilitated through transport with colloidal manure materials.

Effects of kaolinite colloids on Cd²âº transport through saturated sand under varying ionic strength conditions: Column experiments and modeling approaches.

Column experiments were performed under various ionic strengths (0.0-0.9 mM) using 10 mg L(-1) of Cd(2+) without kaolinite colloids and 10 mg L(-1) Cd(2+) mixed with 100 mg L(-1) kaolinite colloids. The nonequilibrium two-site model (TSM) described the behavior of both Cd(2+) transport and Cd(2+) co-transported with kaolinite colloids better than the equilibrium model (CD(eq)) (R(2)=0.978-0.996). The results showed that an increase in ionic strength negatively impacted the retardation factors (R) of both Cd(2+) and Cd(2+) mixed with kaolinite colloids. The presence of kaolinite colloids increased the retardation factors of Cd(2+) from 7.23 to 7.89, 6.76 to 6.61 and 3.79 to 6.99 for ionic strengths of 0.225, 0.45 and 0.9 mM, respectively. On the other hand, the presence of kaolinite colloids decreased the retardation factor of Cd(2+) from 8.13 to 7.83 for ionic strength of 0.0 mM. The fraction of instantaneous sorption sites (f) parameters, kinetic constant for sorption sites (α) and Freundlich constant (K(f)) were estimated from HYDRUS-1D of TSM for Cd(2+) transport. The fraction of instantaneous sorption sites was found to increase for an increase in ionic strength. K(f) values of Cd(2+) transport without kaolinite colloids for 0.0, 0.225 and 0.45 mM were found to be higher than those of Cd(2+) transport with kaolinite colloids, except for ionic strength of 0.9 mM. Hence, the presence of kaolinite colloids probably retarded the mobility of Cd(2+) in porous media for higher ionic strengths. Furthermore, retardation factors and K(f) values of both Cd(2+) transport and Cd(2+) co-transport were shown to decrease when ionic strength increased. Interestingly, according to TSM, the fraction of instantaneous sorption sites tends to increase for an increase in ionic strength, which imply that the mechanism of Cd(2+) sorption onto quartz sand can be better described using equilibrium sorption rather than nonequilibrium sorption for an increase in ionic strength.

Factors affecting EDTA extraction of lead from lead-contaminated soils.

The effects of solution:soil ratio, major cations present in soils, and the ethylenediaminetetraacetic acid (EDTA):lead stoichiometric ratio on the extraction of lead using EDTA were studied for three different Superfund site soils, one rifle range soil, and one artificially lead-contaminated soil. Extraction of lead from the lead-contaminated soils was not affected by a solution:soil ratio as low as 3:1 but instead was dependent on the quantity of EDTA present. Results of the experiments showed that the extraction efficiencies were different for each soil. If sufficiently large amount of EDTA was applied (EDTA-Pb stoichiometric ratio greater than 10), most of the lead were extracted for all soils tested except for a Superfund site soil from a lead mining area. The differences in extraction efficiencies may be due to the major cations present in soils which may compete with lead for active sites on EDTA. For example, iron ions most probably competed strongly with lead for EDTA ligand sites for pH less than 6. In addition, copper and zinc may potentially compete with lead for EDTA ligand sites. Experimental results showed that addition of EDTA to the soil resulted in a very large increase in metals solubility. The total molar concentrations of major cations extracted were as much as 20 times the added molar concentration of EDTA. For some of the soils tested, lead may have been occluded in the iron oxides present in the soil which may affect lead extraction. While major cations present in the soil may be one of the factors affecting lead extraction efficiency, the type of lead species present also play a role.