Journal of Environmental Engineering

The United States produces significant quantities of waste biomass from wastewater treatment, food production, food services, and landscape and wood debris. This waste contains essential resources, including water, carbon, and nutrients. Conversion of carbon to energy and recovery of nutrients and water have the potential to reduce the use of scarce resources, protect the environment, and save funds. Several established, demonstrated technologies that convert wastes to resources are available and are reviewed in this paper. Included are anaerobic digestion, direct combustion, biodiesel production, ethanol production, and particulate phosphorus and organic nitrogen separation using ultrafiltration, ammonia stripping, and reverse osmosis. Thermochemical technologies include pyrolysis, torrefaction, and gasification. Unlike solar, wind, hydrological (dams and pumped storage), and closed-loop geothermal renewable energy technologies, the transportation of biomass can be energy intensive. Therefore, a holistic evaluation of complexity, costs, and benefits of converting wastes to resources is mandatory to ensure a net-positive energy value and environmental protection. Several tools described in this paper are available to aid stakeholders when considering biomass waste-to-resource projects. Many of these tools have long been in practice and have been extensively covered in technical and popular literature. As interest in renewable energy resources increases, new tools are being developed to address the challenges posed by biomass projects. This paper discusses tools that combine waste biomass inventories with a geographic information system mapping platform and cost-benefit analysis that return environmental, economic, and financial assessments of selected feedstocks that can be used in planning, siting, and financing a biomass-to-energy project.

Most current commercial applications of ballasted flocculation use silica sand to increase floc size and density. Other ballast media with different specific gravity may offer advantages such as increased applicable superficial velocity or increased particulate matter removal. This study assessed the relative effect of five ballast media on ballasted flocculation/settling performance: anthracite, recycled crushed glass, conventional silica sand, garnet sand, and magnetite sand, with a common $d_{50}$ of 150 μm but variable specific densities of 1.45, 2.58, 2.62, 3.93, and 5.08, respectively. Based on microscopic observations and assuming discrete particle removal in an ideal settler, mean superficial media settling velocities were respectively calculated as 35, 73, 74, 122, and $137\mathrm{m}/\mathrm{h}$. These values do not account for the effect of lamellae or other specific geometries of different patented clarifiers (e.g., CoMag, Densadeg, Sirofloc, and Actiflo). Although the use of magnetite sand allows the total suspended solids load to increase by more than twofold compared to silica sand, the residual turbidity increased after settling as the mixing intensity needed to maintain denser media in suspension was augmented. Consequently, the lowest residual turbidity (0.78 NTU for surface water and 1.38 NTU for wastewater) was observed when anthracite was used as the ballast medium. The ballast media geometry did not significantly affect turbidity removal and settling velocity. Hence, recycled crushed glass was identified as a potential alternative to conventional silica sand despite its higher angularity.

Graphene oxide (GO) has emerged as a promising material for next-generation water treatment membranes. In this study, three types of GO membranes were produced and tested. Both GO and graphene oxide plus polyvinyl alcohol (GO/PVA) membranes were produced using the vacuum-assisted self-assembly (VASA) method, and a chitosan and graphene oxide composite membrane (CSGO) was assembled using GO as a nanoscale filler via evaporation under reduced pressure. Scanning electron micrographs (SEMs) confirmed that GO and GO/PVA membranes were particularly well ordered, whereas CSGO exhibited reduced ordering attributed to structural and composition differences of GO and CS. The fabricated membranes were tested in a dead-end flow system indicating that each membrane containing GO produced a higher flux at 413 kPa pressure relative to conventional polyamide reverse osmosis (RO) membranes used for benchmark comparison, but was lower in flux to a nanofiltration (NF) membrane. Challenging the membranes with $7.5\mathrm{mg}/\mathrm{L}$ methylene blue solution resulted in $\ge \hspace{0.17em}97.5\%$ removal of the dye by all three of the membranes.

In the Hong Kong Harbour Area Treatment Scheme (HATS), a concentrated 10% chlorine solution (in sodium hypochlorite with a specific gravity of 1.2) is used directly to disinfect sewage that receives a chemically enhanced primary treatment (CEPT). The dense chlorine solution is injected into the treated sewage flow in the form of multiple dense jets. Field observations have revealed large variations of chlorine consumption in the treatment system. In view of the importance of the disinfection system to the environmental impact and operational cost of HATS, the mixing of a chlorine jet with the CEPT effluent and its associated chlorine consumption are studied for the first time using a $1:2$-scale sectional physical model located inside the sewage treatment plant, using sewage and a chlorine solution obtained on-site. The total residual chlorine (TRC) concentration distribution is measured at different cross sections of the chlorine mixing chamber. The overall disinfection efficiency is assessed by detailed measurement of the water quality of the outflow from the chamber, which receives only a few seconds of chlorine exposure. At a targeted dosing rate of $10–20\mathrm{mg}/\mathrm{L}$, only approximately 60–80% of the sewage flow is exposed to chlorine upon leaving the chamber, and approximately 70–80% of the chlorine mass flux is consumed within a very short distance (0.5–1 m, or a matter of several seconds) from the chlorine dosing unit. Overall, there is a 1-log Escherichia coli (E. coli) kill within the chlorine mixing chamber, with effective E. coli kill associated with a TRC concentration above $15\mathrm{mg}/\mathrm{L}$. The detailed field-scale model tests help to unravel the causes of the observed large TRC fluctuations in the plant and provide a basis for optimization of the chlorine disinfection operations (e.g., use of lower source chlorine concentration).

A concentrated 10% chlorine solution (in sodium hypochlorite) with a specific gravity of 1.2 is used for the disinfection of treated primary effluent in the Stonecutters Island Sewage Treatment Works (SCISTW) in Hong Kong. The chlorine solution is injected in the form of coflowing dense jets into a sewage flow of approximately $1.4\mathrm{million}{\mathrm{m}}^3/\mathrm{day}$, with a target dosing rate of $10–20\mathrm{mg}/\mathrm{L}$ (or an initial dilution of 5,000 to 10,000). The large sewage flow to chlorine dosing flow ratio has resulted in insufficient mixing and high chlorine demand as a result of the fast reaction of chlorine with organic and inorganic substances in the sewage. An integral jet model is developed to predict the jet mixing in the initial contact region of chlorine and sewage, accounting for chemical kinetics of the chlorine–sewage reaction. The model is calibrated using field data obtained from experiments using real sewage and chlorine solution in a field-scale physical model inside the sewage treatment works. The mathematical model is used to guide experimentation of various alternatives to reduce the chlorine demand. It is found that under the same chlorine dosage, diluting the concentrated chlorine solution by four times (i.e., containing 2.5% available chlorine) and doubling the number of injection jets can result in the reduction of chlorine demand by 10–15%. The chlorine dosage optimization was supported by extensive field measurements and three-dimensional (3D) computational fluid dynamics (CFD) simulations.