Journal of Environmental Engineering

This work explores the rate-limited dissolution of entrapped nonaqueous phase liquids (NAPLs) in a fractured “permeable” formation through the use of mathematical models, developed on the basis of a simple conceptual framework. The feasibility of pump-and-treat operations in such formations is assessed through characterization of dimensionless system parameters governing NAPL dissolution and solute transport. These dimensionless parameters include the ratio of permeable block to fracture flow (mobility number), the system dimensionless mass transfer coefficient, and the number of contaminated fracture sections. A closed form analytical solution is developed for the simple case of a single contaminated fracture section. For domains with a large number of contaminated fracture sections, a numerical approach is presented that combines a finite difference scheme for simulation of solute transport in the fracture network together with an analytical solution for solute transport in the permeable block flow. The most favorable conditions for pump-and-treat systems are shown to be systems with large mobility numbers and large mass transfer coefficients. In systems with intermediate and small values of these parameters, effluent solute concentrations may be significantly below equilibrium values, reducing the effectiveness of pump-and-treat remediation. It is also shown that an equivalent continuum approach can be used to model rate-limited NAPL dissolution and transport in the simplified fractured permeable formation. The effective mass transfer coefficient for the representative continuum, however, is nonlinearly related to the local mass transfer coefficient.

A bench-scale study of seven Pb-contaminated soils from actual sites was undertaken to determine the feasibility of soil washing to meet cleanup goals. In addition, soil characteristics and operational factors affecting washing performance were identified. Soils were characterized by a sequential extraction method and scanning electron microscopy (SEM). Following a treatment process that consisted of particle size and density separation, and batch soil washing, soil Pb treatment goals were met for five of the seven sandy soils. Toxicity characteristic leaching procedure (TCLP) limits were met for all soils. In soil washing, the most important factor affecting removal of Pb was pH, with greater removals observed at a lower pH. Ethylenediaminetetraacetic acid (EDTA) addition significantly improved soil-washing performance at pH 3, and the temperature and liquid-to-solid (L/S) ratio had a minimal effect for the range of conditions studied. Washing results were in qualitative agreement with expectations extrapolated from the soil-characterization data. However, definitive identification and distribution of the Pb-binding mechanisms in these real waste-site soils could not be ascertained from the sequential extraction or SEM data.

Release profiles of p,p′-DDT, p,p′-DDD, and p,p′-DDE from sediment collected from Indian Creek, Alabama, were measured with a gas purge-induced desorption technique and compared to a model developed in this work. DDT entered this sediment via effluent discharged upstream from the collection site from 1947 to 1970 by a DDT manufacturing facility. The purge technique used vessels constructed with 70–100 μm fritted glass bottoms, through which air enters, distributing gas bubbles evenly to a sediment-water suspension. Purging compound from the water phase with the gas bubbles induced desorption of compound from the sediment particles. The purged chemicals were captured on tenax traps. Purge experiments were performed with sediment masses ranging from 0.37 to 3.7 g in 200 mL water at an air flow rate of 1 L/min. The total percentage removal of each compound after 46 days averaged 22, 58, and 75% for DDT, DDD, and DDE, respectively, indicating the extreme resistance of DDT to desorption from this sediment. The time to reach 25% removal was approximately 50 days for DDT and 4 days for DDD. Results are compared to a simple Fickian diffusion model in which diffusion is assumed to occur in one direction from a fixed depth to a plane surface. The concentration at the exposed surface is assumed to be in equilibrium with the aqueous concentration, which in turn is depleted by transfer to the gas phase through first-order decay. Simulations indicate that DDD transfer across the liquid-gas interface is limiting at high sediment concentrations.

An evaluation was performed of the uncertainty in predicted carcinogenic risk resulting from uncertainty in site properties and fate and transport predictions for a simple contaminated soil site. SoilRisk, a risk model for organic contaminants in soil, was applied to a case-study site consisting of sandy soil and a thin, near surface, unconfined aquifer. The site-related parameters found to affect predicted risk most significantly were the soil-water volumetric flux rate (J_w) in the unsaturated zone, the longitudinal dispersivity (α_L) and the Darcy velocity (V_d) in the saturated zone, and the soil organic carbon fraction in both zones (foc, foc_sat). Model runs using probability distribution functions (PDFs) for these input parameters yielded cumulative distribution functions (CDFs) for the total risk estimates, the shape and location of which depended on the chemical and exposure scenario. In general, uncertainty in risk at the case-study site was found to be greater for the more mobile and less degradable of the chemicals studied [trichloroethylene (TCE) and chlordane] than for benzene, which is highly degradable, and benzo[a]pyrene (BaP), which is very immobile in the subsurface.

A dynamic model has been developed to describe upflow anaerobic sludge blanket (UASB) reactors from several aspects including reactor hydraulics, biological reaction kinetics, and mass transfer within anaerobic granules. A flow model of a nonideal continuously stirred tank reactor (CSTR) followed by a dispersion plug flow reactor (PFR) was used to simulate the reactor hydraulics as observed from a LiCl tracer study. The dynamic model based on this flow model was then evaluated by a set of acetate impulse data and verified with a data set from a two-step acetate feed increase experiment from a bench-scale UASB reactor. The model describes UASB reactor performance well. Simulation results indicate significant effects of reactor nonideal flow, diffusional resistance, as well as degradation kinetics on overall substrate utilization rate. Sensitivity analyses on model parameters K_s, k_m, K_L, D, R, and nonideal flow factors revealed granule size has a strong impact on the reactor performance. The effect of K_L is not significant. Reactor mixing was improved by an increase in biogas production.

Bound water content is generally taken as an index for differentiating the moisture content in a sludge. However, some recent works have demonstrated that such an index is an operationally defined value differing markedly when different techniques are employed. In this study, the continuous moisture distribution in waste activated sludge is constructed on the basis of the combined thermal gravimetry analysis (TGA) and differential thermal analysis (DTA). The water-solid bond strength is considered here to be an adequate classification index. The bound water contents measured via filtration, drying, or expression are interpreted as the portion of moisture displaying a bond strength higher than the threshold energy that could be provided by filtration, drying, or expression tests. Employing one of these bound water measurement methods is thereby equivalent to setting a threshold bond strength to divide the continuous moisture distribution into bound water and free water. Moreover, the effects of a freeze/thaw treatment on the continuous moisture distribution are discussed.

The flow and density fields in final settling tanks of biological treatment plants with a rectangular plan are accounted for by extensive observations in a prototype basin. The time-averaged flow and density fields can be summarized as follows: for an inlet structure located at the bottom, a density current of variable concentration develops. Because its flow is supercritical behind the inlet wall and due to energy dissipation, a density hydraulic jump is generated that is responsible for macroturbulence and local resuspension. The hydrodynamics of the density current, the basin end zone, and the surface effluent layer are described and a generalized approach is presented. The flow in a final settling tank may be improved with a number of modifications that can be added to existing designs or included easily in future designs. Modifications refer mainly to the inlet structure, the density current, and the outlet channels. A preliminary design procedure is presented and examples outline the computational procedure.

Combined-sewer overflow (CSO) must be controlled by a storage-treatment system because storm flow in the combined sewerage system is intermittent and highly variable in both pollutant concentration and flow rate. A treatment facility operating without the benefit of upstream storage would need to be very large and costly in order to handle the relatively high volume and flowrate of a CSO. Similarly, if storage is used without treatment, the storage volume required would be very large and also expensive. This paper describes a strategy to optimize CSO control system. This optimized system maximizes the use of the existing system before new construction and sizes the storage volume in concert with the wastewater treatment plant capacity to obtain the lowest-cost storage and treatment system.

The effects of initial moisture content and compressive loads on the compactability and air permeability of two compost mixes, biosolids (primary) + bark + sawdust + recycled compost, and cow manure are described. Biosolids compost in the moisture range of 42–57% was subjected to compressive stresses of 0–43.2 kPa and the degree of compaction and pressure drops in the flow range of 0.05–0.20 m/s were measured. The relationship between air permeability and total air-filled porosity at different moisture levels was described using the Kozeny-Carman flow model. Biosolids showed significant compressive behavior and lower permeabilities with increasing moisture content (42–57%). Cow manure had a high moisture-retaining capacity and the differences in compaction behavior and air permeabilities were small when moisture levels were varied (57–73%). The model described the relationship between air permeability and effective free air space with a high level of accuracy (R²> 0.95) at all moisture levels and compacted state. It was shown that in beds of moisture levels greater than 60% (wet basis), free air space less than 30% can be reached, and fan power requirements can increase greatly with increasing depth.

The solid waste management plan of 1989 stipulates that landfilling is the chief waste disposal method in Hong Kong. Although currently no charge is involved in solid waste disposal, a landfill charge, covering the capital, operation, and maintenance costs but not the external environmental costs, will soon be introduced. This paper aims to identify the minimum size of the external costs involved with landfilling and incineration in Hong Kong with respect to global and local air pollution, water pollution, noise, and pollution from waste transport. It is found that landfilling results in a slightly lower external cost than incineration. However, factors not included in this paper, such as road casualties that arise from hauling waste and disamenities (a term coined to embrace negative impacts such as eyesores and loss of amenity values) may affect the preference of landfilling over incineration. The present study also shows that the external costs of landfilling are much greater than its capital, operation, and maintenance costs. The considerable size of the estimated external costs of traditional waste management methods demonstrates the need for in-depth and comprehensive research to develop local data.

The biological treatability of assimilable organic carbon (AOC) in drinking water was investigated under oligotrophic conditions. AOC of the simulated feed was provided with model compounds such as glucose, acetate, or glycerol at 1 mg$\dot{s}$C/L. The bioreactors were initially inoculated with Pseudomonas fluorescens strain P17, an indigenous organism isolated from water-distribution systems, to initiate biological activity. Cellimmobilization techniques investigated consisted of entrapment in calcium (Ca) alginate beads and carrier-binding with Type-Z carrier. It was found that reactors packed with either alginate beads or Type-Z carrier were operational at stable conditions over a period of two years without any loss of metabolic functionality. Excellent removal of influent substrates was obtained by all reactor types at hydraulic retention times ranging from 6 to 30 min. Type-Z carrier appears to be the preferred choice of biocarrier because of its inorganic structure and simplicity of immobilization. The steady-state, average effluent AOC and total organic carbon (TOC) from bioreactors packed with Type-Z carrier ranged from 9 to 28 μg$\dot{s}$C/L and from 0.23 to 0.32 mg$\dot{s}$C/L, respectively, when the operating hydraulic retention times varied from 20 to 30 min. Maintaining a low level of AOC in finished water helps minimize microbial regrowth in a water-distribution system.

This paper considers the troubled diffusers of four major U.S. outfalls, one in Hawaii and three on the West Coast. All four of these facilities had damaged and/or blocked outflow devices, and contained deposits that were responsible for flow obstruction. The difficult and expensive rehabilitation measures are described on a case by case basis. In one instance, a pipe in deep water was simply flushed following port-clearing operations. In other instances, nonhazardous materials were vacuumed into shallow local waters, and risers were returned to full service. In the fourth case, the hazardous deposits were carefully pumped from deep water to a barge and treated; the liquid and solid fractions were disposed of separately. This paper concludes with a pair of sections: a summary of the important lessons learned from the case studies, and a list of vital considerations for the whole outfall inspection-operation-maintenance-repair process. Full realization of the potential pitfalls should help engineers and coastal wastewater agencies prevent their recurrence.

The performance of sequencing batch reactors (SBRs) for the treatment of effluent from a conventional coke-plant wastewater biological treatment system has been evaluated. The results showed that an operating mode that incorporated two anoxic portions, one ahead of the aeration period and one after the aeration, was superior to the other modes (prearranged denitrification and postarranged denitrification) for nitrogen removal. The 4-h anoxic fill probably allowed some substrates in the influent to be stored in the biomass and led to subsequent storage-induced denitrification in the second anoxic portion during the reaction period. Under these conditions, the ammonia nitrogen (NH₃-N) and chemical oxygen demand (COD) reductions were 82.5 and 65.2%, respectively. During the fill period of an SBR, the organics concentration was reduced, probably due to the adsorption of organics onto activated sludge and dilution of the mixed liquor. Some readily biodegradable organics in the influent, such as phenol and o-cresol, were utilized by denitrifiers as a carbon source. A 16-h aeration significantly reduced the concentrations of o-cresol, 3,4-dimethyl phenol, and 2-quinoline alcohol, while reductions in quinoline, isoquinoline, indol, and methyl quinoline were limited.