Assessment and Remediation of Contaminated Sediments (ARCS) Program
Table of Content
- Chapter 1
- Chapter 2
- Chapter 3
- Chapter 4
- Chapter 5
- Chapter 6
- Chapter 7
- Chapter 8
- Chapter 9
- Chapter 10
- Chapter 11
- List of Figures
- List of Tables
Remediation Guidance Document
US Environmental Protection Agency. 1994. ARCS Remediation Guidance Document. EPA 905-B94-003. Chicago, Ill.: Great Lakes National Program Office.
Table of ContentsPRETREATMENT TECHNOLOGIES
- DESCRIPTIONS OF TECHNOLOGIES
- SELECTION FACTORS
- ESTIMATING COSTS
- ESTIMATING CONTAMINANT LOSSES
Pretreatment is a component of a remedial alternative in which sediments are modified or conditioned prior to final treatment or disposal. This definition is somewhat artificial, because some of the pretreatment technologies do "treat" the sediments, and if conducted alone, could logically be called a treatment component.
There are two primary reasons for pretreating contaminated sediments. The first reason is to condition the material such that it meets the requirements of the treatment and/or disposal components of the remedial alternative. Most treatment technologies require that the feed material (e.g., sediment) be relatively homogeneous and that its physical characteristics (e.g., solids content, particle size) be within a narrow range for efficient processing. Pretreatment technologies may be employed to modify the physical characteristics of the feed material to meet subsequent processing needs. Examples of the feed requirements for selected treatment technologies are shown in Table 6-1. Sediment treatment technologies that use a continuous feed system generally have more stringent requirements for pretreatment than those using a batch feed system. For example, oversized material can cause blockage or ruptures in conveyance systems. In addition, excessive fluctuations in the solids content can alter the process conditions, thereby reducing treatment efficiencies. Pretreatment requirements for sediment disposal are generally less stringent than those for treatment.
The second reason for pretreating contaminated sediments is to reduce the volume and/or weight of sediments that require transport, treatment, or restricted disposal. Some physical separation technologies can separate fractions of sediments that may be suitable for unrestricted disposal or beneficial use, and concentrate the contaminants in a smaller fraction of the sediments.
Most of the design and operating experience with the pretreatment technologies discussed in this chapter was developed from applications involving municipal and industrial sludges and mining and mineral processing. These applications are generally of a larger scale than that expected for most sediment remediation projects and are usually part of a permanent process operation, whereas most sediment remediation projects will be of shorter duration. These differences should be considered when applying guidance developed for processing municipal and industrial sludges and mining materials to contaminated sediment sites.
The applicability of pretreatment technologies to dredged material was examined by the Corps as part of a pilot program to investigate alternative disposal methods for dredged material from the Great Lakes (USACE Buffalo District 1969) and as part of the Dredged Material Research Program (Mallory and Nawrocki 1974). A detailed literature review of pretreatment technologies is provided by Averett et al. (in prep.).
This chapter provides descriptions of two types of pretreatment technologies--dewatering and physical separation. Discussions of the factors for selecting the appropriate technology and techniques for estimating costs and contaminant losses are also provided.
Dewatering technologies are used in sediment remedial alternatives to reduce the amount of water in sediments or residues and to prepare the sediments for further treatment or disposal. The need for dewatering is determined by the water requirements or limitations of the treatment or disposal technologies and the solids content of the sediments following removal and transport.
Mechanically dredged sediments typically have a solids content comparable to that of in situ sediments (about 50 percent by weight for most fine-grained sediments). Hydraulically dredged sediments are in a slurry with a solids content typically in the range of 10-20 percent. Some hydraulic dredge pumps are able to move slurries with higher solids content, but the average solids content in an extended dredging operation is rarely greater than 20 percent. To prepare dredged sediments for most treatment or disposal technologies, water must be removed and/or the solids content of the sediments must be made more uniform. Dewatering will be required for most sediment remedial alternatives that involve hydraulic dredging or transport. If the sediments are mechanically dredged and transported, the dewatering requirements may be greatly reduced or eliminated.
Another function performed by dewatering is the reduction of the volume and weight of the sediments, which decreases the subsequent costs of handling, transport, and treatment and/or disposal of the solids. Dewatering will reduce the weight of a sediment load, but the effects of dewatering on the volume of a sediment load are more complex. When a sediment slurry is dewatered, the removal of free water will directly reduce the volume of material remaining in a nearly one-to-one relationship. Sediments that have been partially dewatered or mechanically dredged will lose additional water, but the volume will not always be reduced because the water driven from the voids between sediment particles is replaced by air. Some dewatering processes may even increase the volume of the sediments. The water removed during dewatering may be contaminated and require further treatment, as discussed in Chapter 9, Residue Management.
Three general types of dewatering technologies are discussed below:
- Passive dewatering technologies
- Mechanical dewatering technologies
- Active evaporative technologies
Passive Dewatering Technologies
In this document, the term "passive dewatering" refers to those dewatering techniques that rely on natural evaporation and drainage to remove moisture. Drainage may occur by gravity or may be assisted (e.g., using vacuum pumps). Some mechanical movement of the sediments, such as the construction of trenches, may also take place.
Dewatering of dredged material has traditionally been accomplished in CDFs, which rely on primary settling, surface drainage, consolidation, and evaporation. Subsurface drainage and wick (vertical strip) drains have also been demonstrated or used at CDFs to promote dewatering and consolidation. These technologies require significant amounts of land and are most effective if the sediments can be spread out in thin layers or "lifts."
Sediments can also be dewatered in temporary holding/rehandling facilities, tanks, and lagoons using the same design principles developed for CDFs. CDFs are discussed in more detail in Chapter 8, Disposal Technologies. Specific aspects of dewatering within a CDF or CDF-like structure are described below.
Surface Drainage--Drainage of surface water can be accomplished through a number of mechanisms. Most existing in-water CDFs on the Great Lakes have dikes constructed of stone and granular material that remain permeable as they become filled. Water drains through the permeable sections, and suspended sediments become entrapped by the dike material (Miller 1990). At upland facilities, and at in-water CDFs that have filled above the water table, surface water is drained to the discharge point(s), which may include overflow weirs, filter cells, or pump control structures. Drainage water from a CDF includes both the water in the sediment slurry and rainfall runoff. Progressive trenching is a method employed to aid the drainage of water in CDFs and hasten evaporative drying.
Evaporative Drying--The desiccation of dredged material by evaporative drying results in the formation of a crust at the sediment surface. This method of drying is a two-stage process. The first stage of drying occurs until all free-standing water has been decanted from the dredged material surface. The corresponding void ratio at this point is termed the initial void ratio (eoo) and has been determined to occur at a water content of approximately 2.5 times the Atterberg liquid limit of the material. The second stage of drying occurs until the material reaches a void ratio called the desiccation limit (edl). At this point, evaporation of any additional water from the dredged material will effectively cease. The edl corresponds to a water content of 1.2 times the Atterberg plastic limit (USACE 1987b). The thickness of the crust and rate of evaporative drying and consolidation are dependent on local conditions and sediment properties, and can be estimated using the Primary Consolidation and Desiccation of Dredged Fill (PCDDF) module of the Automated Dredging and Disposal Alternatives Management System (ADDAMS) model (Schroeder and Palermo 1990).
Subsurface Drainage--A subsurface drainage system can be used at a CDF for dewatering of dredged material and/or leachate collection. One approach is the placement of perforated pipes under or around the perimeter of a CDF that drain into a series of sumps from which water is withdrawn. The pipes can be placed in a thin layer or trenches of drainage material, typically sand or gravel. The feasibility of subsurface drainage as a sediment dewatering technology may be limited where several layers of fine-grained sediments are to be disposed because they may clog the drainage materials.
Several variations of subsurface drainage systems can be used, including the gravity underdrain, vacuum-assisted underdrain, vacuum-assisted drying beds, and electro-osmosis. The gravity underdrain system provides free drainage at the base of the dredged material by the gravity-induced downward flow of water. The vacuum-assisted underdrain is the same as the gravity-fed system, but uses an induced partial vacuum in the underdrainage layer. The latter system improves dewatering by 50 percent (Haliburton 1978), but requires considerable maintenance and supervision.
Wick Drains--Wick drains or "wicks" are polymeric vertical strips that provide a conduit for upward transport of pore water, which is under pressure from the overlying weight of the material. By placing the vertical strips on 5-ft (1.5-m) centers to depths of 40 ft (12 m), both radial and vertical drainage are promoted. Wick drains can reduce consolidation time by a factor of 10 compared to natural consolidation (Koerner et al. 1986).
Mechanical Dewatering Technologies
Mechanical dewatering systems have been extensively used for conditioning municipal and industrial sludges and slurries, as well as mineral processing applications. These systems require the input of energy to squeeze, press, or draw water from the feed material. Generally, mechanical dewatering technologies can increase the solids content up to 70 percent by weight. The features and requirements of six mechanical dewatering processes are summarized in Table 6-2 [part i] [part ii] [part iii].
The performance of a mechanical dewatering system is measured by a number of parameters, including:
- Chemical conditioning dosage, measured as the mass of conditioner per mass of dry solids
- Solids capture, defined as the dry mass of dewatered solids per dry mass of solids fed into the process
- Solids content of the dewatered material
With sewage sludges, the dosage of organic (polymer) conditioners in mechanical dewatering systems is generally <0.1 percent by weight, while the dosage of inorganic conditioners is substantially higher. For example, lime and ferric chloride may be used in dosages as high as 20 percent (Dick 1972).
A high solids capture is desirable, because solids lost from the process (i.e., in the filtrate or centrate) represent a route for contaminant loss. Some particulate loss during mechanical dewatering is inevitable; therefore, the effluent stream must be treated using treatment technologies described in Chapter 9.
Most mechanical dewatering processes increase the solids content of the feed material to a level comparable to that of the in situ sediment deposits (about 50-percent solids). These dewatering processes work best with homogeneous waste streams at a constant flow rate. Because hydraulic dredging produces highly variable flow rates and solids concentrations, direct dewatering of hydraulically dredged slurries would be inappropriate. Temporary storage in a tank, lagoon, or CDF would be necessary to equalize flows and concentrations prior to further dewatering by one of the mechanical processes.
Mechanical dewatering has been tested with dredged sediments on a limited scale (Averett et al., in prep.). A vacuum filtration unit was tested on sediments from Toledo Harbor, Ohio (Long and Grana 1978). The solids content prior to conditioning with lime ranged from 15 to 23 percent. The post-treatment solids content was consistently above 43 percent. An 2.5-m belt filter press was demonstrated on sediment from the Ashtabula River in Ohio at a rate of 23 tonnes/hour. Solids were increased to 50-60 percent by weight, with solids losses of 2-5 percent (Rexnord, Inc. 1986).
A substantial amount of design and operating guidance on mechanical dewatering systems has been developed for municipal and industrial wastewater applications (USEPA 1987b) and mineral processing applications (Weiss 1986). There are some fundamental differences between sediments and sludges that need to be considered when using this guidance, including:
- Sediments are usually less compressible, less gelatinous, and lower in organic content than wastewater sludges, and thus are generally easier to dewater
- The solids content of feed material, typically 3-6 percent in a wastewater treatment plant, will be considerably higher for sediments (15-25 percent)
- Sediments can contain rocks and large particles that can interfere with or damage dewatering equipment, necessitating pretreatment by screening
- Municipal sludges are generated on a continuous basis, whereas dredging produces sediments over a relatively short time scale
- The disposal of wastewater and filtrate is a relatively minor concern for municipal sludges because this water can be easily returned to the treatment process; however, wastewater from the dewatering of contaminated sediments is a significant concern, and separate treatment for this water may need to be employed
There are numerous manufacturers of mechanical dewatering equipment. Vendor contacts are listed in USEPA (1987b) and may be obtained through wastewater treatment and mining/mineral processing trade journals.
Active Evaporative Technologies
Active evaporative technologies are different from the evaporative drying techniques used at CDFs in that artificial energy sources are used to heat the sediments, as opposed to solar radiation. Evaporation is the most expensive dewatering technology, but has been effectively used to prepare municipal sludges for incineration or for sale as fertilizer (Dick 1972). Nearly all of the water is removed, resulting in a solids content of about 90 percent. Technologies applied to sludges that may be applicable to fine-grained sediments include:
- Flash dryers
- Rotary dryers
- Modified multiple hearth furnaces
- Heated auger dryers
The most common conventional evaporation process used for waste recycling is agitated thin-film evaporation (Averett et al., in prep.). This process is capable of handling high-solids content slurries and viscous liquids. It may also be possible to use conventional evaporation equipment commonly found in the chemical-and food-processing industries. These technologies remove water in the form of steam and may also remove volatile contaminants.
Evaporative dewatering technologies have not been demonstrated with sediments on any scale. Most of the design and operating experience and guidance on these technologies are from municipal and industrial wastewater applications (USEPA 1987b).
Physical separation technologies are used in sediment remedial alternatives to remove oversized material and debris in order to produce an acceptable feed material for subsequent handling and treatment. These technologies are also used to separate the sediments into two or more fractions based on physical properties or characteristics to reduce the quantity of material requiring treatment or confined disposal.
The following types of physical separation technologies may be applicable to contaminated sediments:
- Debris removal
- Screens and classifiers
- Gravity separation
- Froth flotation
- Magnetic separation
The general features of these technology types are summarized in Table 6-3 [part i] [part ii] [part iii] and discussed in the following paragraphs. Many of the physical separation technologies discussed below are mineral processing technologies, which have been widely used in the mining industry to recover valuable minerals or metals from ores. Methods such as size classification, magnetic separation, gravity separation, or froth flotation, collectively known as mineral processing, can be applied in some cases to separate contaminated sediment fractions from the bulk sediments.
Debris Removal Technologies
Dredged material often has significant quantities of debris and oversized materials. Examples of debris commonly encountered during dredging include: cobbles, bricks, large rocks, tires, cables, bicycles, shopping carts, steel drums, timbers, pilings, and automobiles.
Pockets of bulk materials, such as coal or gravel, may be encountered near docks and loading areas. The amount of debris is generally greatest in sediments along riverbanks and at bridge crossings, especially where there is unrestricted public access to the waterway.
Debris can be a significant problem for a dredging operation because it can clog hydraulic cutterheads and cause bucket dredges to be raised without full closure, resulting in increased sediment resuspension. Debris can also complicate the transport of dredged sediments, possibly requiring separate handling. Large debris must be separated and removed prior to any other pretreatment or treatment process. The size requirements of feed materials for various treatment technologies are shown in Table 6-1.
Debris may be separated during removal (dredging) or as part of material handling activities in between other components. For example, debris might be separated while sediments are being removed from a barge and transferred into truck trailers for transport, or while sediments are being removed from a disposal/storage area and fed into a pretreatment process. The technologies available for debris removal are relatively simple, such as a drag-line, grapple bucket, mechanical removal, and screens (discussed in later sections of this chapter).
A drag-line is a grappling hook or rake that is dragged along the river bottom with a steel cable from a boat or from a land-based winch. A grapple bucket is a specialized crane-operated bucket, commonly used for placement of large stones, that can be used to remove debris from a waterway. Large debris can be cleared from the sediments prior to dredging. This method may also be used to clear debris from a CDF prior to excavating sediments for treatment.
Mechanical removal is the separation of large debris using mechanical dredging or construction equipment. During a dredging operation employing a clamshell dredge or backhoe, large debris can be separated from the bulk of the dredged material. This requires a skilled operator and a place to store the debris. For a land-based operation, the debris might be separated and placed in a bin or dumpster for storage and transport. During marine operations, a clamshell dredge is often placed on a large floating platform, which may provide sufficient space for storing debris. Conventional earthmoving equipment that may be used for handling and rehandling of sediments between other components could also be used for separating large debris. Large plants may require grinding to ease rehandling and disposal.
Debris that has been separated is generally covered with contaminated sediments and may need to be decontaminated. Possible reasons for decontaminating debris include:
- The cost of disposal of the decontaminated debris is lower than the cost of disposal along with contaminated sediments
- The disposal facility for sediments will not accept the contaminated debris
- Transport of the contaminated debris is not allowable
- The decontaminated debris has a salvage value
Contaminated debris should be stored in a secure place or container until disposed or decontaminated. Decontamination may involve washing with water or steam. Wash water must be collected and treated as necessary.
Screens and Classifiers
While hydrocyclones are the most popular separation devices, grizzlies, trommels, vibrating screens, and mechanical classifiers are all widely used in mineral processing applications. Screens and classifiers may be the first units in a complex separation process or the only units in a simple process. A trommel and vibrating screen were used in the ARCS Program demonstration at Saginaw, Michigan (USACE Detroit District 1994). A grizzly, vibrating screen, and screw classifier were used at a sediment remediation demonstration conducted at Welland, Ontario (Acres International Ltd. 1993).
A hydrocyclone is a high-throughput, particle-size classifier that can accurately separate sediments into coarse- and fine-grained portions. The typical hydrocyclone (Wills 1988) is a cone-shaped vessel with a cylindrical section containing a tangential feed entry port and axial overflow port on top and an open apex at the bottom (the underflow). A slurry of the particles to be separated enters at high velocity and pressure through the feed port and swirls downward toward the apex. Near the apex the flow reverses into an upward direction and leaves the hydrocyclone through the overflow. Coarse particles settle rapidly toward the walls and exit at the apex through a nozzle. Fine particles are carried with the fluid flow to the axial overflow port.
The particle size at which separation occurs is primarily determined by the diameter of the hydrocyclone. Hydrocyclones from 0.4-50 in. (0-125 cm) in diameter make separations from 1 to 500 um. The common practice is to employ several identical cyclones from a central manifold to achieve the desired capacity. Most manufacturers provide detailed manuals for selecting and sizing hydrocyclones (Arterburn 1976; Mular and Jull 1980).
The feasibility of using hydrocyclones for processing dredged material was investigated by the USACE Buffalo District (1969) and Mallory and Nawrocki (1974). A 12-in. (30-cm) hydrocyclone was tested using sediments from the Rouge River in Michigan. The physical separation was considered good, but the coarse fraction contained a large amount of volatile solids, determined to be detritus and light organic matter (USACE Buffalo District 1969).
Hydrocyclones were the major process unit used in a pilot-scale demonstration of particle size separation technologies conducted at Saginaw, Michigan, by the ARCS Program (USACE Detroit District 1994) and at a similar demonstration in Toronto, Ontario (Toronto Harbour Commission 1993). At the Saginaw demonstration about 75 percent of the sediments were successfully separated into a sand fraction, reducing the concentrations of PCBs from 1.2 ppm in the feed material to 0.2 ppm in the sand fraction.
Gravity separators separate particles based on differences in their density. Organic contamination in sediments is often associated with solid organic material or detritus, which have much lower densities than the natural mineral particles of the sediment. Particles with high concentrations of heavy metals would be significantly more dense than the natural mineral particles. A dense media separator was used at the ARCS Program demonstration at Saginaw, Michigan (USACE Detroit District 1994), and at the demonstration conducted in Toronto, Ontario (Toronto Harbour Commission 1993).
Froth flotation is used in the mining industry to process millions of tonnes of ore per day. Copper, iron, phosphates, coal, and potash are a few of the materials that can be economically concentrated using this process. The process is based on manipulating the surface properties of minerals with reagents so that the mineral of interest has a hydrophobic surface (i.e., lacks affinity for water) such as wax. The minerals to be rejected have, or are made to have, a hydrophilic surface (i.e., a strong affinity for water). When air bubbles are introduced, the hydrophobic minerals attach themselves to the bubbles and are carried to the surface and skimmed away.
When using flotation to remove oily contaminants from sediments, a surfactant is used in a manner that resembles a detergent. Most organic contaminants are naturally hydrophobic, and the objective in using a surfactant is to reduce the hydrophobicity of the oil phase to the point where it will be wetted by the water phase and detach itself from solid surfaces. Surfactants are able to accomplish this because such molecules have a lipophilic (fat-soluble) head, which is absorbed into the oil phase, and a hydrophilic tail, which extends into the water phase. The result of this is that the overall hydrophobicity of the oil phase is decreased. The strength of a surfactant's attachment to an oil phase is approximated by the hydrophile-lipophile balance of the surfactant. Once freed of the solid surface, an oil droplet is assisted to the surface by air bubbles and skimmed away.
Magnetic separations are classified as two types depending on the intensity of the magnetic field employed (or the magnetic susceptibility of the minerals to be separated). Low-intensity separations usually employ permanent magnets, and are most often used for material coarser than about 75 um with high magnetic susceptibility, such as iron ore. High-intensity separations that employ electromagnets are much more versatile and capable of recovering iron-stained or rusted silicate minerals from other purer, nonmagnetic phases.
Wet, high-intensity magnetic separation (WHIMS) appears to be most applicable to sediment remediation, with separations possible down to 5 um, although at very low capacity. The WHIMS unit is essentially a large solenoid. Magnetic material is trapped on magnetized media in the chamber of the device, then flushed free in a rinse cycle when the feeding of sediment and magnetic current are stopped. Thus, the WHIMS is not technically a continuous throughput device, but operates in separate loading and rinsing cycles (Bronkala 1980).
Magnetic separation was used during part of the dredging and treatment demonstration conducted with sediments from the Welland River, Ontario (Acres International Ltd. 1993).
Not all remedial alternatives will require a pretreatment component, while others may require several process options for pretreatment. The need for pretreatment is generally driven by the treatment and/or disposal components selected for a remedial alternative and the physical characteristics of the sediments. A treatment technology with restrictive feed requirements may necessitate a multiunit pretreatment system, as illustrated in Figure 6-1.
The design of a pretreatment system must be compatible with other remedial components. Sufficient lands must be available at the treatment or disposal sites to operate pretreatment units and accommodate residues. Water extracted from dewatering technologies and process water from separation technologies may require a separate treatment system from that used for disposal site effluent or leachate. Some of the pretreatment water may be reusable within the process system.
The selection of a dewatering technology usually involves choosing between a passive and a mechanical approach. Active evaporative technologies would only be employed where subsequent processes (e.g., thermal desorption or incineration) require extremely dry materials. The advantages and disadvantages of passive and mechanical dewatering are listed in Table 6-4.
If a permanent or temporary confined (diked) facility is a part of the remedial alternative, passive dewatering can be conducted within this facility. Facility design might accommodate a number of functions, including settling, dewatering, storage, rehandling, and disposal. Other pretreatment and treatment equipment might be stationed within or adjacent to the facility to minimize transport distances. Separate cells might be constructed in the facility to accommodate different functions. The design of CDFs is discussed in Chapter 8, Disposal Technologies.
Haliburton (1978) and the Corps' engineering and design manual, Confined Disposal of Dredged Material (USACE 1987b), provide detailed guidance on the use of CDFs for dewatering and consolidating sediments. The Corps developed computer software for evaluating the primary consolidation and desiccation of dredged material as part of ADDAMS (Stark 1991).
Mechanical dewatering is most suitable where land is not available for a temporary or permanent diked facility. Selection of a specific type of mechanical dewatering equipment depends on the requirements of the treatment or disposal components to follow. Maximum solids content is generally achieved using a recessed plate or diaphragm plate filter. However, if lower solids content is acceptable (e.g., for transport to a landfill), less costly processes such as centrifugation or belt filter presses may be more appropriate. A summary of selection factors is provided in Table 6-5.
Laboratory methods are available for predicting the performance of some mechanical dewatering systems. Prediction of vacuum and pressure filtration performance and capacity can be done with a filter-leaf test, which involves filtration on a filter medium disc of known area (Dahlstrom and Silverblatt 1980). Laboratory methods are also available to predict the performance of gravity thickening. The method of Coe and Clevenger (1916) is standard for simple gravity thickening, while the method of Kynch (1952) is more useful for coagulated or flocculated solids. For some mechanical dewatering systems, bench-scale or pilot-scale applications may be needed to fully assess equipment performance and operating conditions, and to select conditioning agents.
Evaporative (drying) technologies, which are by far the most expensive form of dewatering, would usually not be employed for sediments. In certain cases, such as when sediments are to be processed in a thermal treatment system, the removal of water is a primary consideration in reducing the cost of treatment. In these cases, thermal treatment systems may provide a source of waste heat that could be used for evaporation. The primary concern regarding use of this technology is volatile emissions. Because sediments are heated, volatile and semivolatile contaminants are released. Contaminants of concern for this process include low molecular weight PAHs, PCBs, and mercury. Subsequent treatment of off-gases would probably be required and could add significant costs to the process.
The factors for selecting a physical separation technology will depend on the objective of pretreatment. If the objective is to remove materials from the sediments that may interfere with subsequent handling, treatment, or disposal, selection factors would be related to the feed requirements of these subsequent components and the physical character of the sediments delivered by front-end components. If the objective is to separate the sediments into two or more fractions with differing treatment and disposal requirements, the selection factors would be related to the distribution of contaminants within the sediment matrix and their separability based on physical characteristics.
The selection of equipment for removing oversized material from a process stream is fairly straightforward. Each process unit will have a maximum feed size (above which the unit might be damaged) and a target particle size separation, as summarized in Table 6-6. Most of the equipment is available in different screen sizes or diameters to accommodate a range of particle size separations. Equipment selection must consider the characteristics of the incoming sediments and the feed requirements of subsequent components with the operation and performance specifications of the pretreatment unit.
Aside from removing oversized materials that might disrupt subsequent pretreatment or treatment processes, physical separation processes may reduce the quantity of materials requiring expensive treatment or disposal. Virtually any sediment can be separated into two or more fractions based on one or more physical properties (i.e., particle size, mineralogy, density, magnetic, and particle surface properties). With some sediments, contaminants can be separated into specific fractions by mineral processing technologies that use these same physical properties.
Best results will be obtained when the pretreatment system is chosen based on a detailed knowledge of the physical and chemical characteristics of the sediment. Mineral processing unit operations appropriate to the physical characteristics of the sediment can then be arranged into an integrated system. Detailed characterization of the physical properties of the sediment, including the analyses shown in Table 6-7, and chemical analysis of separable fractions are needed to determine the selection of a mineral processing method or methods.
Other testing that is helpful is sediment mineralogy, or identification of chemical phases, using scanning electron microscopy with energy-dispersive techniques and possibly x-ray diffraction. Equally important is knowledge of the history of the contaminated site, which could provide information about the nature of the contaminant-bearing phases.
If discrete sediment phases containing contamination have been identified, then an appropriate mineral processing method can be selected. Mineral processing methods are selected to separate sediments based on the known physical properties of the phases found to be present. For example, if most of the contamination is found to be associated with fine silt or clay particles, size classification techniques may be appropriate. The distribution of PCBs in relation to particle size in sediments from the Saginaw River is shown in Figure 6-2. As illustrated, most of the PCBs were associated with a relatively small particle size fraction of the sediments. Particle size separation of the Saginaw River sediments during a pilot-scale demonstration yielded a small fraction (20 percent of original material) of silt and clay containing most of the PCBs, and a large fraction (80 percent of original material) of sand with reduced concentrations of PCBs (USACE Detroit District 1994). Toxicity testing of the sand fraction showed a slight decrease in comparison to the untreated sediments, indicating that these materials may be suitable for unrestricted disposal, pending further analyses.
A few important points about mineral processing technologies should be noted. Mineral processing makes particle-particle separations. No chemical bonds are broken, and no contaminants are destroyed. This is in contrast to many other remediation technologies, where a process such as incineration actually destroys the contaminants. In addition, mineral processing separations are based on differences in the physical properties of particles, so that no separation can be achieved if all particles are physically similar. Finally, the capacity and efficiency of most mineral processing operations decreases with particle size. Each individual mineral processing operation has a range of particle sizes for which the technology is effective. Further information on mineral processing methods is available from several sources, including Collins and Read (1979) and Somasundran (1979).
Selection and feasibility testing of mineral processing methods are described in an extensive handbook published by the Society of Mining Engineers (Weiss 1986). Bench-scale testing to verify mineral processing performance is inexpensive, and scale-up reliability is well documented. Most plants with capacities up to 2,700 tonnes/day are designed from laboratory studies without pilot-scale plant testing.
Debris Removal Technologies
Large debris is most likely encountered during mechanical dredging, especially in urban areas with unrestricted public access to the waterfront. Debris may be separated by the dredge operator as it is removed and placed into a barge, or it may be separated at the first transfer point where the sediments are placed into a disposal facility or loaded for transport. The advantages of removing debris at the first transfer point include: 1) mechanical equipment (i.e., cranes and backhoes) used for rehandling are typically smaller than the dredge, 2) more space is available to store debris, 3) it is easier to contain drippage, and 4) a properly designed site can also be used for decontamination.
Screens and Classifiers
Grizzlies and trommels are frequently used to remove small debris and are useful in sediment processing to capture driftwood, junk, or large rocks that would foul or damage other processing equipment. Vibrating or other moving screens are often chosen for separations of particles larger than about 100 um in diameter (Colman 1980; Reithmann and Burnell 1980).
Grizzlies are the simplest and coarsest devices for removing small debris. Their most likely application in sediment remediation would be to remove rocks and debris 5 cm or larger in diameter to prevent damage to subsequent equipment. A grizzly should always be used if there is a possibility of equipment damage from large rocks or foreign objects.
Trommels are used to remove gravel, rocks, or trash 1-10 cm in diameter from sediment prior to further processing. Difficulty has been reported with the formation of clay balls on trommel screens, effectively trapping fine particles that should pass through the device. If a significant clay fraction is present in the sediment, a water spray may be helpful to prevent the formation of clay balls. A log washer or similar disaggregating device might be used in conjunction with a trommel.
Vibrating screens are used to make particle size separations in sediments with particle diameters from 4,000 to 100 um. Hydrocyclones could also be used for separations in this range, usually with a lower unit cost. Selection of a vibrating screen over a hydrocyclone might be justified if variations in feed rate are anticipated, lower volumetric capacity is required, there is a wide variation in particle densities, or the feed solids content exceeds 25-30 percent.
Mechanical classifiers such as spiral or rake classifiers can also be used for separations in the same size range as hydrocyclones. A spiral or rake classifier might be selected for a sand-silt separation when a high solids content is required in the sand product (e.g., when sand is to be transported by belt conveyor). Mechanical classifiers are very sensitive to variations in the solids content of the feed material, and require a constant volumetric feed rate for reliable performance.
The selection of hydrocycloning pretreatment to reduce the volume of contaminated material to be treated is dependent on three factors. First, the contamination must be strongly distributed toward either the coarse- or fine-grained particles (usually the fines), so that the remaining fraction of the sediment is clean enough to be suitable for disposal without treatment or for unrestricted disposal (van Veen and Annokkée 1991). Second, the mass of the sediment must be sufficiently distributed toward the cleaner fraction so that an appreciable amount of clean material is recovered. As a general guideline, this would require that the contaminated material make up no more than about 40 percent of the total sediment weight. Third, the subsequent treatment to be used on the contaminated material must be as efficient and economical with a smaller volume of more heavily contaminated material as it would with the unseparated bulk sediment.
In the usual hydrocyclone application, it is the fine particles that carry the most contamination. Therefore, it is important in making a separation that the coarse product or underflow be as free of misplaced fine particles as possible. Some fine particles are always carried along with the water that exits the cyclone with the underflow, so the amount of this water should be kept to a minimum. Proper selection of the size and design of the apex nozzle will accomplish this. Another way of ensuring a clean underflow product is double-desliming, where the underflow product is subjected to a second hydrocyclone treatment, resulting in fewer misplaced fine particles. A final option recommended by at least one hydrocyclone manufacturer is to add clear water to the hydrocyclone just above the apex nozzle. The additional water forces some of the water containing misplaced fine particles back to the overflow, resulting in a cleaner underflow product.
The traditional methods for evaluating the feasibility of gravity separation in the laboratory are "sink-float" tests using a variety of dense liquids, such as bromochloromethane and tetrabromoethane (Mills 1985). A sediment sample can be separated into fractions of differing specific gravity using these liquids and specially constructed separatory funnels. These heavy liquids are suitable for density separations of sediment for metal contaminants. Density separations of organic contaminants can be predicted using water elutriation, in which closely sized material is allowed to settle against a rising current of water.
A density-based separation may be successful if contamination is found to reside disproportionately in a phase of different specific gravity than the bulk of the sediment matrix. For example, organic contaminants are frequently found attached to detrital material such as wood and leaf fragments. This material is much less dense than mineral matter and can be easily separated in a gravity separator. Most metallic phases are considerably denser than most sediment matrices, and can also be recovered. A specific gravity difference (between the phases to be separated) of about 0.4 is usually enough to effect a separation with most equipment.
The applicability of gravity separation to a contaminated sediment is dependent on the size of the sediment, sediment density, and the concentration criterion (C), defined as:
The feasibility of gravity separation for sediments of varying particle sizes is related to the concentration criteria in Table 6-8 (from Aplan 1980).
The use of froth flotation is warranted when most of the contamination is found in a phase (or phases) distinct from the bulk of the sediments. The most promising application would be with sediments containing an oily phase, where surfactants could be used to aid in detaching the organic-phase contaminants from sediment particles, followed by collection of the contaminants in an organic-laden froth. Another possible application might be in connection with a minerals industry-related site, where metal contamination is associated with a specific mineral phase. In this case, a flotation system could be designed to recover that phase.
Determining the feasibility of froth flotation for a given assemblage of particles involves two components. First, the phases present must be identified. In minerals processing, phases are usually identified using a combination of microscopic analysis and x-ray diffraction. Infrared spectroscopy might be used to identify principal organic phases. Second, bench-scale testing is used to identify surfactants and operating conditions for an effective separation. This is an expensive and time-consuming process relative to the characterization required for a particle size separation, for example. Accurate and complete knowledge of the identity of phases in the system will hasten and economize this process.
Only the low-intensity, rotating, drum-type separators and the WHIMS system appear to have significant applicability to sediment remediation, because they operate on wet material. The choice between these two devices is based on the particle size and magnetic susceptibility of the phase(s) to be recovered. Fine or paramagnetic material requires the WHIMS system. The low-intensity systems are generally applicable only when the material to be recovered is ferromagnetic.
The most practical method of evaluating the feasibility of magnetic separation is to conduct separability tests using laboratory-scale equipment.
There is considerable cost estimating guidance available on applications of mechanical and evaporative dewatering technologies to municipal and industrial sludges, and considerable cost data exist on applications of physical separation technologies in the mining and mineral processing industries. Most of these applications involve permanent installations that process large quantities of materials at controlled rates under near-ideal conditions. Sediment remediation will typically have none of these features. Cost information from wastewater and mineral processing operations will be provided in this document because it is the best or only information available, but applications to sediment remediation should be expected to be significantly more expensive.
Passive Dewatering Technologies
The capital costs for construction of CDFs are discussed in Chapter 8, Disposal Technologies. Capital costs for temporary diked facilities for dewatering can be estimated in a manner similar to that for CDFs. Although the design requirements may be less stringent for temporary facilities, one additional cost that would be incurred after the remediation is completed is the removal of the facility and decontamination of the site. Costs for sand drying beds may be adapted from guidance published for municipal sludge (USEPA 1985a). No cost data are available on the installation of wick drains at CDFs.
Activities associated with operating a CDF for dewatering may include water-level management, operation and maintenance of pumps and overflow weirs, and progressive trenching. At Corps CDFs around the Great Lakes, water level management is typically conducted by the dredging contractor (or subcontractor) and represents a relatively small effort. The cost of progressive trenching is highly site-specific. Haliburton (1978) estimates that the cost of implementing three trenching cycles over 2 years at a 100-acre (41-hectare) CDF would be approximately $128,000 (updated to January 1993 dollars). This cost assumes 70-percent operational efficiency, with administrative costs assumed to be 20 percent.
Mechanical Dewatering Technologies
Mechanical dewatering equipment may be purchased outright or leased. In addition, dewatering services are available on a contractual basis. If sediment dewatering is to be performed intermittently, or just once, contracted services may prove to be more cost effective. Contractors generally offer belt filter presses and recessed plate filters, although centrifuges are also sometimes available. Several vendors contacted during preparation of this document indicated "typical" pricing in the range of $3-$10 per hundred gallons ($0.79-$2.64 per hundred liters) of feed material. This can be expressed on a dry-ton basis if the feed solids concentration is known, as shown in Table 6-9:
Contractual costs are controlled by the quantity of the material to be processed, the dewaterability of the material, and the required cake solids concentration. The volume of slurry generated during a sediment remediation project might be considered moderately "large" when considering mobile dewatering. For example, 10,000 yd (7,600 m) of in situ sediments in a 10-percent slurry would result in a total volume of approximately 10 million gal (38 million L). Contaminant concentrations may influence cost as well.
Capital costs for construction of mechanical dewatering systems, based on municipal wastewater applications, are presented in Table 6-10. These costs include equipment purchase, installation, and housing costs. All equipment (except gravity thickener) is assumed to be housed in a building.
Operation and maintenance costs for mechanical dewatering include the following components:
- Maintenance of equipment and facilities
- Power requirements
- Chemical costs
The operating costs for specific mechanical dewatering systems are discussed in the following paragraphs. The costs of treating and disposing of wastewater streams resulting from dewatering are discussed in Chapter 9, Residue Management.
Belt Filter Press--Belt filter presses are probably the most energy conservative and, therefore, the most economical mechanical dewatering units to operate. The average power requirements range from 0.8 kW (1 hp) to 5.7 kW (8 hp) per meter of belt width. Replacement of the filter belts is one of the most common maintenance items. The main reasons for failure of the belts are tearing at the clipper seam, inferior quality belt material, ineffective tracking systems, and poor operation and maintenance. Average belt life is about 2,700 running hours with a range of 400-12,000 running hours (USEPA 1987b).
Process control is extremely important to ensure optimum performance of the dewatering system. By keeping accurate records (i.e., a log) the operator can determine how well the press is performing. In addition, preventive maintenance and waste minimization can be integrated to deter unnecessary shutdown and reduce chemical costs, respectively (USEPA 1987b).
Solid Bowl Centrifuge--Operating costs for centrifuge technologies depend on the solids capacity of the centrifuge and polymer dosage. Additional factors such as bowl speed and temperature can affect the final sludge cake. Particular attention should be focused on polymer dosage. Continual laboratory testing will minimize polymer dosage and maximize the dryness of the cake solids, thus minimizing costs (USEPA 1987b). In addition, replacement costs for centrifuge scrolls and bearings can be significant. Examples of operation and maintenance costs for centrifuges from two wastewater treatment works operated by the Metropolitan Water Reclamation District of Greater Chicago are shown in Table 6-11.
An evaluation of the costs of dewatering dredged material using mechanical dewatering methods was conducted by the USACE Buffalo District (1969) for various dredging volumes. The system consisted of slurried dredged material fed into solid bowl centrifuges by pipeline. The centrifuges were sized at 12,500 pounds (27,500 kg) per unit per hour, producing a cake of approximately 50-percent solids. A summary of the system costs is provided in Table 6-12. Total costs are based on a term of 10 years with a 4.625 percent annual interest rate. Operating costs are based on labor, utility, and maintenance.
Filter Press--Proper sludge conditioning is a key component of an efficient and effective filter press operation. Routine evaluations and recordkeeping are recommended, because operating conditions may vary, leading to conditioner changes (USEPA 1987b). Operation and maintenance costs include the labor needed to operate the press, power to pressurize the feed material, and maintenance of the equipment. Most of the maintenance costs are for replacement of the filter cloths (USEPA 1985a). Requirements for power and materials costs, based on municipal wastewater experience, are shown in Table 6-13. Manpower and polymer requirements are a function of processing rate and dewatering characteristics, respectively.
No cost data are available on evaporation of sediments. In general, there is very limited information on evaporation of waste solids. Probably the best indication of evaporative costs are those for the Carver-Greenfield process discussed in Chapter 7, Treatment Technologies. Based on a hypothetical site with 21,000 tonnes of drilling mud wastes, with a solids content of 52 percent and an oil and grease content of 7-17 percent, processing costs have been estimated to range from $180-$200 per tonne of feed material (Schindler 1992).
Because physical separation technologies are economically applied on a large scale to ores of low value-to-mass ratio, they are among the least expensive processes in modern industry. For example, in processing copper, five or six separate mineral processing operations are performed, plus smelting and refining, at a rate of more than 91,000 tonnes/day, all on an ore that contains less that $10 worth of copper per tonne. It is important to note that large economies of scale are seen in mineral processing operating costs. The cost of treating a tonne of ore in a small operation may be 2-3 times the cost of treating the same amount in one of the larger facilities.
Mining industry costs for all major mineral-processing unit operations are well documented; however, considerable difficulty is encountered in applying these costs to an environmental remediation project. The U.S. Bureau of Mines has published and uses a cost estimating system to calculate capital and operating costs based on plant throughput by summing incremental costs of the unit operations and other contributions to cost. In sediment remediation, this system would appear to be most useful for larger projects, in excess of about 500 tonnes/day of sediment (U.S. Bureau of Mines 1987).
Debris Removal Technologies
Debris removal is an anticipated inconvenience during most maintenance dredging projects at Great Lakes harbors. Contractors are typically advised in dredging contracts to expect some debris and be prepared to remove it. Removal generally requires additional time by dredge operators to handle large debris and causes decreased production. The costs of debris removal are generally factored into the dredging cost estimates.
During sediment remediation, additional provisions may be necessary because of the highly contaminated nature of the sediments. Most of these costs can also be factored into the costs of other components. If the debris is removed by the dredge operator or during mechanical rehandling or transport, the costs will be reflected as decreased productivity. The costs of additional equipment and labor needed to store the debris and costs for decontamination are project specific.
Screens and Classifiers
Few data are available in the mining industry for these (coarse) size separations. Their cost is typically calculated as part of a larger grinding or mineral processing system. As an example, the operating cost for a washing and screening circuit consisting of a trommel, log washer, and vibrating screens, with ancillary equipment, is estimated to be $8.25/tonne. Such a circuit might be encountered in the gravel or crushed stone industries. With screens and classifiers, equipment costs are generally incidental to the costs of moving material to and through the system.
A typical hydrocyclone designed for soil or sediment remediation, which makes a separation at 75-150 um with a throughput of 18-55 dry tonnes per hour, would cost from $3,750-7,500 (1993 dollars), depending on the exact size and configuration (costs are adjusted from 1990 prices using ENR's CCI factor of 1.07). Because capacity is determined by hydrocyclone size, the cost increment for higher throughput would be linear (i.e., capacity would be increased by increasing the number of hydrocyclones). Pumping and support equipment must also be provided.
Operating costs for hydrocyclones are essentially the cost of pumping the slurry through the unit and costs for occasional replacement of the hydrocyclone liners. These costs are estimated at about $0.12-0.35 per dry tonne (1993 dollars; costs are adjusted from 1990 prices using ENR's CCI factor of 1.07). The highest costs associated with hydrocyclone applications are the manpower costs associated with operating the plant.
An evaluation of the costs of particle size separation of dredged material was conducted by the USACE Buffalo District (1969) for various dredging volumes. The system consisted of a dredged material slurry pumped from a wet well (equalization basin) into hydrocyclones. The underflow (fine fraction) was discharged to a CDF and the overflow (coarse fraction) passed through a spiral classifier before being disposed. A summary of the system costs is shown in Table 6-14. Total costs are based on a term of 10 years with a 4.625 percent annual interest rate. Operating costs are based on labor, utility, and maintenance.
A typical gravity separation circuit, employing Humphreys spirals, in a mineral processing plant is estimated to have an operating cost of $6.05/tonne. The capital cost for a 91-tonne/day Humphreys spiral circuit is estimated to be $270,000.
Based on mineral processing industry experience, the capital cost of a froth flotation plant designed to process 91 tonnes/day is estimated to be $750,000 (Allen, in prep.). Operating costs for froth flotation are about twice those for gravity separation, because of the cost of reagents. Many of the surfactants proposed for sediment treatment are rather expensive and would drive the operating costs even higher.
Magnetic separation plants are used in the iron-ore industry and are quite large. No data are available for magnetic separation plants that operate at capacities lower than about 1,900 tonnes/day. Generally, magnetic separation plants will be more costly to build than gravity separation facilities, but will be about equal in cost to operate.
Estimating Contaminant Costs
While methods for predicting contaminant losses from passive dewatering technologies (primarily CDFs) are fairly well developed, a priori methods for predicting contaminant losses from mechanical dewatering and physical separation technologies do not exist. For these technologies, mechanisms for contaminant loss can be identified, and controls can be installed to minimize loss.
Passive Dewatering Technologies
Contaminant losses from passive dewatering systems are expected to be comparable to those experienced at CDFs. Chapter 8, Disposal Technologies, and Myers et al. (in prep.) provide further discussion of these losses.
Mechanical Dewatering Technologies
The mechanisms for contaminant loss from mechanical dewatering systems will include volatilization and leakage/spillage of solids or water. Systems that are housed can be equipped with controls to collect and route all leakage/spillage for treatment as necessary. Leakage/spillage would most likely be washed into a wet well and pumped to the water residue treatment system.
If the sediments have significant concentrations of volatile or semivolatile contaminants, controls can be implemented to capture and treat any contaminant losses. Contaminant losses will ultimately be limited to the quantity of emission permitted by the regulatory agencies and the residuals generated during the treatment of the off-gas (e.g., spent carbon). Volatilization losses from systems that cannot be housed (i.e., gravity thickeners) may be estimated using the same methods used for CDFs (Chapter 8, Disposal Technologies).
Active Evaporative Technologies
Contaminant loss mechanisms for active evaporative technologies would be similar to those for mechanical dewatering technologies. Because the sediments are heated, volatilization is more likely to be significant, and more elaborate controls would be required.
Debris Removal Technologies
The mechanisms for contaminant loss during debris removal include sediment drippage during handling, volatilization, and wash water. If debris is separated during dredging, there are few controls that can be implemented other than having an adequate storage container for debris. If debris is separated during rehandling (between components), drippage can be controlled using drip aprons or by constructing a low-permeability, drained rehandling area. Drippage from a rehandling area and wash water from debris decontamination should be collected and routed for treatment.
Screens and Classifiers
Contaminant losses from screens and classifiers are the result of volatilization, splashing, or spillage. Mechanical classifiers can readily be fitted with covers to recover volatile contaminants; because these devices require a quiescent flow regime, it is not expected that volatile losses would be much greater than those from sediment in place. Significant losses are not expected from grizzlies. The mixing in trommels and the high-frequency vibration of some moving screens may impart sufficient energy to effect contaminant volatilization; however, substitution of reciprocating or gyratory screens would reduce this possibility.
Contaminant losses from hydrocyclone treatment are expected to be minimal, because the hydrocyclone is an enclosed unit, and material is transferred to and from the hydrocyclone by pumping through rigid pipes. It is possible that some contaminants could be volatilized in the turbulence of the hydrocyclone, but provisions can be made for capture of the escaping gases.
Contaminant losses from gravity separation devices are expected to be relatively low. An exception to this may be volatile losses from shaking tables or other flowing-film concentrators. These losses could be controlled if the equipment was enclosed or housed in a building with air capture and treatment capability.
The most likely loss pathway for froth flotation is volatilization of organic contaminants, which results from forcing quantities of air through the sediment pulp. Ventilation hoods can be fitted on flotation cells to capture volatile emissions.
Contaminant losses from magnetic separations will be no greater than from any other simple materials-handling operation, because no heating or significant increase in air-slurry interface is involved.