Journal of Environmental Engineering

Because the presence of hydrogen sulfide $\left(H_{2}S\right)$ and methyl mercaptan (MeSH) inhibits the removal of some organic reduced sulfur compounds (e.g., dimethyl sulfide, ${\mathrm{Me}}_{2}S)$ in biofilters, a two-stage biofilter may be an appropriate method to treat a mixture of reduced sulfur compounds. This work studied the treatment of odorous air emissions [a mixture of $H_{2}S\text{,}$ MeSH, ${\mathrm{Me}}_{2}S\text{,}$ and dimethyl disulfide $\left({\mathrm{Me}}_2{S}_2\right)]$ using a two-stage biofilter. $H_{2}S$ had the highest overall removal efficiency (96%) followed by MeSH and ${\mathrm{Me}}_{2}S$ (90 and 91% removal, respectively) and ${\mathrm{Me}}_2{S}_2$ (81%). Most of the removal of $H_{2}S\text{,}$ MeSH, and ${\mathrm{Me}}_2{S}_2$ occurred in the primary biofilter (72% $H_{2}S\text{,}$ 66% MeSH, and 52% ${\mathrm{Me}}_2{S}_2)\text{,}$ while most of the ${\mathrm{Me}}_{2}S$ removal occurred in the secondary biofilter (64% ${\mathrm{Me}}_{2}S)\text{.}$ A dynamic model that describes biofiltration was calibrated and validated to $H_{2}S$ and MeSH field data. This is the first time a model was evaluated with organic odor-causing sulfur compound data obtained from a biofilter packed with compost and wood chips. Model simulations showed that $H_{2}S$ removal in a lava rock packed biofilter would be better than in a similar biofilter packed with compost and wood chips.

The objectives of this work were to evaluate the effects of different parameters (Pb concentration, solution pH, Ca/Mg/Na/K salt concentrations) on Pb uptake by cocoa shells and to study the mechanisms of Pb removal in very acidic conditions. Sorption tests were conducted in shaken flasks with synthetic Pb solutions and 15 g/L cocoa shells. A lead uptake value of 161 mmol/kg was measured during the assay with $[\mathrm{Pb}{]}_i=3.66 \mathrm{mmol}/L$ in solution at ${\mathrm{pH}}_i=2.0$ and $T=22°C\text{.}$ Results show that Pb uptake is very similar (14.5–16.0 mmol/kg) for an initial pH between 2.0 and 4.0, but a moderate decrease (10.8 mmol/kg) occurred when the initial pH=1.5 and the $[\mathrm{Pb}{]}_i=0.25 \mathrm{mmol}/L\text{.}$ High Ca and Mg concentrations (2.35 mol/L) in solution induced a significant decrease in Pb removal by cocoa shells, whereas high K and Na concentrations did not affect Pb uptake by this sorbent. This suggests that Pb uptake by cocoa shells is controlled by ion-exchange reactions with Ca/Mg ions and protons. Finally, chemical tests have shown that carboxyl and amine functional groups play a dominant role in Pb uptake by cocoa shells.

An optimal prepolymerized inorganic coagulant, poly-alumino–iron sulphate (PAFS), was developed and examined for drinking water treatment. PAFS is a new type of prehydrolyzed metal–ion coagulant. An investigation of the conditions for preparing PAFS has been conducted using ${\mathrm{Al}}_2({\mathrm{SO}}_4{)}_3$ with ${\mathrm{Fe}}_2({\mathrm{SO}}_4{)}_3$ as primary raw materials. Optimization of the preparation conditions of the PAFS involved variation of the total metal–ion concentration, oxidation temperature, the molar ratio of the oxidant to metal–ion, and the oxidation period. An evaluation of the coagulation performance with two types of test waters showed that selected PAFS coagulants achieved either a greater or similar removal in terms of percentages of color, ${\mathrm{UV}}_{\left(254\mathrm{nm}\right)}$ absorbance and dissolved organic carbon in comparison with another polymeric coagulant, polyferric sulphate (PFS). However, both of these (PAFS and PFS) had a superior performance to ferric sulphate (FS) and aluminum sulphate (AS). Of particular significance was that PAFS achieved the lowest residual metal–ion concentrations, Fe and Al, in comparison with PFS, FS, and AS, respectively.

The adsorption of reactive dyes on sepiolite has been investigated in a series of batch adsorption experiments. Three reactive dyes (Everzol Black B, Everzol Red 3BS, Everzol Yellow 3RS H/C) were used in bottle adsorption studies. While no significant adsorption capacity was obtained for natural sepiolite, high-adsorptive capacities were observed upon using sepiolite modified with quaternary amines. The electrokinetic measurements along with calculations using the cross-sectional area reveal that the quaternary amines adsorb close to bilayer coverage. A mechanism involving electrostatic attraction of the anionic groups of dye molecules onto oppositely charged amine-modified sepiolite surface is proposed to be responsible for the uptake of dyes. The adsorption data were fitted to the Langmuir isotherm. It is found that the modified sepiolite yields adsorption capacities $\left(q_e\right)$ of 169, 120, and 108 mg/g for Yellow, Black, and Red, respectively. These results are comparable to the adsorption capacity of the same dyes onto activated carbon.

Polychlorobiphenyls (PCBs) are one of the most environmentally problematic organic contaminants present in soils and sediments due to their long-life in direct disposal or spills in soil. It is known that surfactant-aided soil washing followed by a photodechlorination process has the ability to separate PCBs from the soil and decompose them all in one step. However, the influence of humic substances on the removal of PCBs in surfactant-aided ultraviolet systems is of special interest because of the involvement of humic materials in the soil-washing process. A PCB congener, 4,4^′-dichlorobiphenyl (DCB), was used as a probe in this study. At low humic acid (HA) concentrations, humic substances promote the photochemical reaction, but a high dose of HA leads to retardation of the DCB decay. A mechanism for the amphoteric effect of HA on the photodecay of DCB is proposed in this study, and two mathematical models are developed to successfully rationalize the amphoteric effect of HA in the photolysis process.

A common failure mode for landfills is clogging of the leachate-collection system. The reduction in hydraulic conductivity associated with clogging causes a buildup of leachate head on the underlying liner, potentially increasing advective contaminant transport from the landfill and contaminating adjacent groundwater. In this paper, the biogeochemical model CCBATCH is used to link a primary cause of leachate collection system failure—${\mathrm{CaCO}}_{3\left(s\right)}$ precipitation—to anaerobic degradation of volatile fatty acids (VFAs) in column reactors used to study the clogging phenomena. One key to applying CCBATCH correctly was dividing the VFA conversion into two steps: conversion of propionate to acetate, carbonic acid, and methane; and acetate conversion to methane and carbonic acid. The primary driver for ${\mathrm{CaCO}}_{3\left(s\right)}$ precipitation in the columns was acetate fermentation to ${\mathrm{CH}}_4$ and $H_2{\mathrm{CO}}_3\text{,}$ which increased the total carbonate concentration in the leachate and shifted the acid/base control to a weaker acid system, which caused an increase in solution pH. A second key to proper modeling was adding ${\mathrm{CO}}_{2\left(g\right)}$ gas transfer to CCBATCH. The modeling results indicate that the kinetics of ${\mathrm{CO}}_{2\left(g\right)}$ gas transfer was a key control over leachate chemistry once acetate fermentation was nearly complete. These results suggest that the best approach for the long-term control of ${\mathrm{CaCO}}_{3\left(s\right)}$ clogging may be to enhance ${\mathrm{CO}}_{2\left(g\right)}$ gas transfer from the leachate while buffering the leachate pH to near neutral. Taken together, these actions should decrease the yield of ${\mathrm{CaCO}}_{3\left(s\right)}$ precipitated per mass of acetate removed.

A mathematical model was developed to describe nitrogen transport in duckweed-covered static ponds for nutrient recovery from swine lagoon water. A finite difference technique was used to solve the partial differential equations describing the ammonia transport and concentration in the pond. The key parameters in the model include the diffusion coefficient of ammonium in the medium $\left(D\right)$ and kinetic constant of nitrogen uptake by duckweed $\left(k\right)\text{.}$ Using one order of magnitude parameter variations, the simulations showed that the model was clearly much more sensitive to $D$ than to $k\text{,}$ indicating the process of nitrogen removal in a static pond by duckweed is diffusion limited. Laboratory testing was conducted with Spirodela punctata 7776, a duckweed strain, to calibrate the model. The calibration of the model with experimental data yielded a new ammonium transport coefficient $\left(T\right)$ that is 85 times of $D$ value. Model results showed good agreement with depth-wise experimental ammonium concentration and the model also demonstrates that intermittent mixing every 3 h can enhance ammonium uptake. Additionally, an apparent drop in pH near the duckweed mat at the surface was observed that may explain low rates of ammonia emission from duckweed ponds.

Five fluorescent dyes were evaluated as partitioning tracers for detection and quantification of tetrachloroethylene (PCE): fluorescein, rhodamine WT (RWT), sulforhodamine B (SRB), eosine, and pyranine. Results obtained from batch tests were (a) sorption partitioning coefficients $\left(K_P\right)$ $0.088\pm 0.025$ and 0.091 ${\mathrm{cm}}^3/g\text{,}$ for RWT and SRB, respectively (eosine and pyranine increased fluorescence after soil/dye batch tests; thus, $K_P$ could not be obtained for these tracers) and (b) nonaqueous phase liquid/water partitioning coefficients $\left(K_{\mathrm{NW}}\right)$ for RWT, SRB, and eosine were 0.1703, 0.0603, and 0.049, respectively (pyranine also increased fluorescence in PCE/dye tests). Retardation factors $\left(R_f\right)$ measured in column tests for fluorescein, RWT, SRB, eosine, and pyranine were (a) $1.08\pm 0.07\text{,}$ $1.64\pm 0.38\text{,}$ 1.23, 1.45, and 2.97, in the absence of PCE; and (b) 1.31, 2.3, 1.33, 1.79, and 3.51, in the presence of PCE, respectively. The results of this study suggested that the partitioning behavior of the tracers tested is different from simple hydrophobic partitioning and that their fluorescence is complex and dependent upon several variables. Column tests indicated that RWT, SRB, and eosine are possibly suitable as partitioning tracers.

Soil vapor extraction is a popular soil remediation technology that is hampered by less than optimal performance in the field due to mass transfer limitations. Therefore, laboratory column venting experiments were completed to quantify mass transfer limitations for the removal of multicomponent nonaqueous phase liquid (NAPL) contaminants from a silt loam soil at three water contents. The observed mass transfer limitations were quantified using a four phase multicomponent, nonequilibrium contaminant transport model based on first-order mass transfer kinetics. The overall mass transfer coefficient $K_{g}a$ was treated as a variable and modeled as a linear function of the NAPL volumetric fraction using two adjustable parameters (m, the slope parameter and $K_g{a}_{\min }\text{,}$ the intercept). Both were back calculated from column venting data. The agreement between the calibrated model and experimental results were favorable for the removal of single and binary contaminants under conditions ranging from near equilibrium to severe mass transfer limitations and extended tailing. A strong dependency of $K_{g}a$ on water content was evident by the differences in $K_g{a}_{\min }$ and to a lesser extent, m, at the three water contents investigated. A single expression $K_{g}a$ captured the performance of both components in the binary mixture. For the quaternary venting experiments a single expression for $K_{g}a$ captured the performance of all four components well under air dry conditions. However, the agreement between the hexane model versus the experimental result deteriorated significantly as the water content increased. This difference is attributed to hexane’s lower affinity for the water phase relative to the other three components in the mixture.

Water diversion for hydroelectric power generation impacts the temperature of mountain streams. Such changes are estimated by using a coupled one-dimensional dead-zone heat balance model. In very steep river sections, the dissipation of kinetic energy is the dominant heat source. For such streams, water diversion has only a minor effect on water temperature, because dissipation-induced temperature changes are independent of discharge. In contrast, in river sections of gradual slope, the influence by solar radiation, long-wave radiation, and heat exchange with the streambed is stronger. In such cases, a discharge reduction can lead to significant temperature changes. For a small stream in the southern Swiss Alps, model results show that diversion increases temperature by about 3.7 (±0.9)°C in a 21 km long river section under high solar radiation during summer. During a cold winter episode, water temperature is estimated to be about 1.8 (±0.8)°C lower compared to natural conditions. This heat balance model can also be used to simulate the effect of different measures to reduce water temperature changes in affected streams.

Load estimation is an important component of watershed analysis and management, particularly in light of the Total Maximum Daily Load program set forth by the Clean Water Act. This paper addresses load estimation in tidally driven channels that are dominated by advection and well mixed over the channel cross section. Such systems are common in Southern California, and represent an important link between watersheds and the coastal ocean. The oscillating flow field of tidal channels calls for a time series of data (volumetric flow rate and scalar concentrations) sampled at a sufficiently high frequency to resolve the net load of each scalar. Results of an experiment conducted in the Talbert Channel in Huntington Beach, California, to quantify freshwater loads suggest that sampling should be conducted at a minimum frequency of 0.5/h. In this system, traces of freshwater were detected in the channel by salinity depressions that ranged from 1–7 ppt and lasted 4–6 h. The freshwater load was computed as the integral of the volumetric flow rate multiplied by the freshwater fraction of the water column. Based on the duration of these events and the 0.5/h sampling frequency, these results suggest that loads will be resolved to a reasonable level of accuracy when each event is captured by at least three or four data points. Strategies to estimate the flow rate and scalar concentrations are discussed.

Ultraviolet (UV) irradiation has emerged as a viable alternative for water/wastewater disinfection. Laboratory dose-response data from collimated-beam tests are commonly used as a basis for determining the necessary delivered UV dose for full-scale UV systems as measured by UV intensity and exposure time. While researchers often think that germicidal dose can be estimated confidently in a collimated beam system, the reported dose-response relationships vary considerably. Numerous factors may affect the test results to some extent. They include apparatus setup, column dimensions, UV lamp type and output, intensity measurement, shutter type and operation, petri dish specifications, sample volume and depth of the liquid, mixing condition, laboratory settings, microbial organism preparation and testing, and water quality. The methodology used to calculate the UV dose for collimated beam tests is also a critical factor. To ensure reproducibility or to have a meaningful comparison of results from different collimated beam tests, a standardized protocol for collimated beam testing and its dose calculation is necessary. The standardized collimated beam test can then serve as a tool to “calibrate” the dose calculation models for field-scale UV reactors and, consequently, to meaningfully compare the dose values derived from these models. A comprehensive standardized collimated beam testing protocol is warranted to advance applications of UV disinfection for water/wastewater. However, development of such a protocol is a significant research effort that will require input from researchers of various technical and disciplinary backgrounds. In addition, some technical issues still need to be clarified by further research. This paper presents important issues associated with collimated beam testing in the context of developing a standard protocol as well as future research needs. The main objective is to serve as a starting point for development of a standardized collimated beam testing protocol that is universally acceptable by researchers, industries, and regulatory agencies.