Journal of Environmental Engineering

A methodology for evaluating the variability of risk resulting from the use of uniform, concentration‐based soil‐quality criteria is developed and applied to polychlorinated‐biphenyl‐(PCB‐) contaminated soil as an example. A modified version of a soil risk model developed by the U.S. Environmental Protection Agency is used to assess risk variability for different levels of PCBs in soil, representing different cleanup goals. Continuous distributions representing the variability in site conditions across the United States are assigned to parameters in the fate and transport model to examine the effect that such variability has on individual risk when uniform cleanup levels for PCBs in soil are used. It is found that for a fixed cleanup goal applied uniformly to all sites, the risk varies little for the case where the hypothetical exposure occurs on‐site, but by as much as two orders of magnitude for the case where the hypothetical maximally exposed individual resides off‐site. Depending on the risk goal of the remediation program, some risk variability may be acceptable and a uniform cleanup goal may be warranted to expedite the remediation process.

The majority of hazardous waste is generated industrially, and the importance of industrial pollution prevention to economic and environmental sustainability is now recognized. Sustainability can be evaluated using benefit‐risk analysis, globally and locally by public agencies and firms. To evaluate pollution prevention seriously, using benefit‐risk analysis or other methods, the benefits of hazardous materials use in industry must be recognized. Pollution‐prevention technologies in a broad range of industries were reviewed as a part of research to develop benefit‐risk methods. Ten basic functions served by hazardous materials in industry were identified as binding, pigmentation, reactants, reaction inhibition, catalysis, bleaching, mass deposition, mass removal, by‐products, and end products. Pollution‐prevention technologies were associated with each function. The classification of technologies offers a basis for development of general pollution‐prevention approaches, and promotes transfer of technologies among industries in which hazardous materials perform similar functions. Hazardous materials used for mass transfer, and those generated as by‐products, appear to represent areas of opportunity for pollution‐prevention research and development.

Two test methods were developed to detect preferential flow along the grout/clay interface in a cylindrical clay sample with a cementious grout core. One test detects preferential flow along the interface by taking ratios of the specific discharge in different regions of the specimen as measured with a triple‐ring permeameter. The other test detects preferential flow along the interface by taking ratios of the velocity in different regions of the sample as measured by timing voltages changes as a sodium chloride solution passes through the sample. The results of both tests indicate that preferential flow does occur in the short term along the grout/clay interface. Comparing the velocity ratios to the specific discharge ratios, it is shown that small volumes of liquid travel rapidly along the interface. While the tests lend insight into the mechanisms of preferential flow, caution must be used in applying the results of such laboratory scale specimens to the field situations. Changes to the equipment and procedures for future testing are suggested to improve reproducibility.

The effects of temperature and pH on aluminum concentrations in alkaline‐treated waters were examined using laboratory and field data. Residual aluminum concentrations were measured in the effluent from the Van de Water treatment plant, Tonawanda, N.Y., over a one‐year period. Also, data from seven other treatment plants in the United States and Canada were compiled and analyzed. Raw water pH for the plants ranged from 6.7 to 8.4. Results showed that key factors influencing residual aluminum concentrations for the conventional treatment plants were pH and temperature. Trends in dissolved aluminum with temperature measured at the Van de Water plant were described reasonably well based on theoretical inorganic aluminum speciation with fitted values of ${\mathrm{pK}}_{\mathrm{s0}}=32.3$ (infinite dilution) and $\Delta H^0=15.9\mathrm{kJ}/\mathrm{mole}$ for $\mathrm{Al}\left(\mathrm{OH}{)}_3\right(s)$. Similar results were observed for three other conventional water treatment plants. Operationally, the effects of pH and temperature on residual aluminum can be described conveniently by pOH, which varies linearly with residual aluminum and is nearly independent of temperature.

Toxic hexavalent chromium, Cr(VI), was reduced by an Escherichia coli strain, ATCC 33456, under both aerobic and anaerobic conditions by using a variety of electron donors including glucose, acetate, propionate, glycerol and glycine. In addition to electron‐donor type, the initial cell concentration, Cr(VI) concentration, pH, and temperature had profound effects on the rate of Cr(VI) reduction. Cr(VI) reduction was not significantly affected by the presence of moderate level of phenolic compounds. The concentration levels that caused a 50% reduction in the rate of Cr(VI) reduction $\left({\mathrm{IC}}_{50}\right)$ were 986, 1,526, and 5,263 mg/L for p‐cresol, 2‐chlorophenol, and phenol, respectively, in anaerobic cultures with higher ${\mathrm{IC}}_{50}$ values noted in aerobic cultures. Cr(VI) reduction by this strain was more susceptible to heavy metals (Zn²⁺ and Cu²⁺) than the phenolic compounds. The reduction of Cr(VI) by E. coli ATCC 33456 was enzymatic, and was not affected by redox potential of the culture media and other possible electron acceptors such as sulfate and nitrate.

Capture and recycle of off‐gas from a complete‐mix activated‐sludge reactor was evaluated as an alternative VOC‐control strategy to reduce the volume and cost of off‐gas treatment. A mathematical model was developed to predict the fraction of individual influent VOC compounds that are biodegraded, stripped, and discharged in the liquid and gaseous effluents of an activated‐sludge reactor with off‐gas recycle. The model utilizes fundamental physical and biokinetic growth and substrate‐removal constants for each VOC. The results indicated that for a biodegradable VOC off‐gas recycle increased biodegradation and reduced stripping of the compound. The fate of a VOC depended on the relative magnitude of $H_C$, ${\mu }_{\mathrm{MAX}}$, and $K_S$ rather than their absolute values. The fate of VOCs having certain combinations of these parameters can be controlled by process and operational parameters of the activated sludge and off‐gas recycle system. These benefits are obtained without additional energy utilization.

The need for the control of volatile organic contaminants (VOCs) has led to the development of a computer‐based suite of mechanistic models collectively known as TOXCHEM, which predicts the fate of VOCs in wastewater treatment plants. TOXCHEM includes both steady‐state and dynamic models, and will predict the location and mass of VOC emissions from several locations within a wastewater treatment installation. Derivations are shown for expressions describing important removal rate coefficients and VOC mass balances for each unit process of a secondary treatment plant. Calibration of the model at pilot scale was conducted, and data from other full‐scale investigations were used to verify the model. It was found that the TOXCHEM models predicted reasonably well and would improve as the true rate coefficients are established. This model has the potential to be an extremely useful tool for wastewater treatment plant engineers and administrators of sewer use regulations.

This paper describes the testing of a wastewater treatment optimization model for the developing world (described by Ellis and Tang in 1991) in applications to four existing wastewater treatment units (in Puchong, Malaysia; Shatin, Hong Kong; Min Shen, Taiwan; and Pattaya, Thailand). In each case, a reciprocal matrix of a suitable number of wastewater‐treatment alternatives (decision variables) was developed for each of 20 technical and socio‐economic parameters. These 20 parameters are flow capacity, influent and effluent characteristics, size of treatment site, nature of site, land cost for site, local/foreign money for construction, local skill for construction, community support, power source, availability of local material, cost of operation and maintenance, professional/technical skill available for operation and maintenance, administration set up, training, professional ethics climate, waterborne diseases, and endemic vector‐borne diseases. The matrices were then integrated by a computer program to obtain a final ranking of the original alternatives using the analytical hierarchy process mathematical technique. The model was also tested for its sensitivity in predicting changes to an appropriate technology selected as socioeconomic parameters change with time.

The calibration of wastewater treatment models suffers from various ill‐definition problems, one of them being the difficulty in the identifiability associated with Monod‐type equations. The present paper presents a methodology for the calibration of such systems that, although not solving the identifiability problem, aims at locating it and finding a trajectory for circumventing its limitations. The procedure is a combination of the classical error‐function minimization with Monte Carlo simulations. The Monte Carlo technique assists in the selection of the optimal ranges of the parameters, and the optimization method focuses on the derivation of the optimal values of the parameters. The structure of the algorithm is such that it can also be applied in an autocalibration mode. The procedure was tested with data deterministically generated by the same model, and was able to converge to the “true” parameter values, which had been established a priori. The methodology was also utilized for the calibration of a larger model (11 parameters) with field data and the results obtained were superior to those from an error‐minimization algorithm conventionally applied.

A series of bench‐scale, continuous‐flow experiments, using primary sludge from a local municipal wastewater‐treatment plant, has been employed to investigate the effect of hydraulic retention time (HRT) on the acid‐phase anaerobic digestion process. Results show that both volatile fatty acid (VFA) production and chemical oxygen demand (COD) solubilization increase significantly with increasing HRT up to 12 h, but drop moderately at a longer HRT. Acetic acid and propionic acid are the main VFAs formed, averaging 46% and 32% of the total, respectively. Variation in HRT has a profound effect on organic substrate degradation as well. Regardless of the prevailing experimental conditions, lipids and carbohydrates are converted at higher percentages than proteins. Furthermore, protein utilization appears to be independent of reactor configuration, but carbohydrate and lipid degradation patterns are a function of the reactor regime.

Volatilization of selected organic compounds from water without agitation has been studied. Mathematical models, incorporating Fick's diffusion law, are developed for volatilization under quiescent condition. A theoretical model is obtained by solving a partial differential equation and is further approximated to a simplified model for practical applications. The volatilization‐rate constant from this model is found to be inversely proportional to the square of the water depth instead of the thickness of liquid film multiplied by water depth used for turbulent condition. The experiments on volatilization of benzene, toluene, trichloroethylene (TCE), and tetrachloroethylene (PCE) were conducted in the laboratory. Both the theoretical model and the simplified model were applied to the experimental data, and correlation coefficients of better than 0.95 were obtained for all tested cases. However, for benzene, the dependency of volatilization‐rate constant on water depth was found to be inversely proportional to the 1.8 power.

Limiting flux theory has become accepted in the design of secondary clarifiers, and has proven useful in the operation of existing activated sludge facilities. In this paper, the governing equations used to find the recycle and waste flow rates which correspond to limiting flux operation for a given plant condition are reviewed. A numerical technique is derived and applied in order to obtain an approximation to this equation set using a specific example. This approach is useful for incorporating limiting flux theory directly into design optimization models.

Settling column test data are used in the design of settling tanks. The traditional method of analyzing test data graphically is sensitive to inaccuracies in individual data points. Curve fitting can be used instead, in which the entire data set is characterized analytically. If the analytic form adequately represents the trends in the data, errors due to inaccuracies in a few experimental data points are minimized. In this study, two simple equations are developed to express percentage concentration as a function of settling time and depth, and published data are used to test the accuracy of the equations. Removal efficiency equations are obtained by integrating the fitted equations. The removal efficiencies for ideal settling tanks are easily calculated using the method given in this study and are in good agreement with those obtained by graphical methods. The critical time beyond which efficiency does not increase substantially for a given depth is defined, which can be useful in the design of settling tanks.

The present study was conducted to examine petroleum hydrocarbons in the sediments of a storm‐water retention basin in central New Jersey. A dry storm‐water detention basin, which services a 0.405‐km² (100‐acre) condominium development, was used for the investigation. Soil samples were collected following the general pattern of water progression through the basin and analyzed for total petroleum hydrocarbons. The study found petroleum hydrocarbons, being hydrophobic in nature, sorb to particulates that settle and reside in the sediment. Results comparing two successive summer samplings show that levels of petroleum hydrocarbons are maintained in the basin sediments. General trends showed that the hydrocarbon concentrations are a function of depth and proximity to the influent culverts. One soil core was extended to a depth of 15.2 cm (6 in.). The results show hydrocarbon concentrations decreasing to background levels at 15.2 cm (6 in.). At present, however, it is not possible to determine whether or not the hydrocarbons are degraded or pose a migration threat.

The problem of solute-mass transport in natural streams is tackled via a governing equation that employs the cumulative discharge, instead of the transverse distance, as an independent variable thus eleminating the need to know the transverse-velocity field. The problem is solved by a numerical scheme, which utilizes exponential interpolation between adjacent nodes. The scheme can regress to a linear-interpolation scheme for Peclet number equal to zero or to an upwind scheme for very large values of the Peclet number. Comparison with field data showed that the problem can be solved satisfactorily based on the knowledge of cross-sectional-geometric data and total-river discharge only, provided that the riverine cross sections are not highly irregular. The computed transverse-dispersion coefficients agree well with values reported in the literature. An independent verification of the transverse-dispersion-coefficient values was obtained by application of a modified method of moments on the same field data.