|
Cost and Performance
Report:
|
Preparation of this report has been funded wholly or in part by the U.S. Environmental Protection Agency under Contract Number 68-W3-0001. It has been subject to administrative review by EPA headquarters and Regional staff and by the technology vendor. Mention of trade names for commercial products does not constitute endorsement or recommendation for use.
This report summarizes cost and performance data for a soil washing treatment application at the King of Prussia (KOP) Technical Corporation Superfund site. This site, located in Winslow Township, New Jersey, is a former waste processing facility that operated from January 1971 to April 1974. On September 28, 1990, a Record of Decision (ROD) was signed to conduct a remedial action for contaminated soil and sludge at KOP. A full-scale soil washing unit, owned and operated by Alternative Remedial Technologies, Inc. (ART) of Tampa, Florida, was used from June 28, 1993 to October 10, 1993 to treat 19,200 tons of soil and sludge at the site The soil and sludge were contaminated primarily with chromium, copper, and nickel. Maximum concentrations of these metals measured in the soil were chromium at 8,010 mg/kg; copper at 9,070 mg/kg; and nickel at 387 mg/kg. Average treatment unit feed concentrations were 660 mg/kg, 860 mg/kg, and 330 mg/kg, respectively. ART performed the soil washing operation under direct contract to the Potentially Responsible Party (PRP) committee who had received a Unilateral Administrative Order from the U.S. EPA in April 1991.
A treatability test of soil washing using soil from the KOP site was conducted in January 1992; the results from the treatability test indicated that the soil at KOP had an acceptable level of sand content and could be effectively treated by soil washing. A demonstration run was conducted in July 1992 when 164 tons of contaminated soil and sludge from the KOP site were processed through a full-scale unit in the Netherlands. The results from the demonstration run conducted in July 1992 further supported the feasibility of soil washing for treating soil from the KOP site to the ROD-specified cleanup levels.
For the full-scale remediation, ART operated the soil washing unit on a production basis with the goal of maintaining a 25 ton/hour throughput. The soil washing unit consisted of a series of hydrocyclones, conditioners, and froth flotation cells. The cleaned sand (product) and process oversize from the soil washing unit were redeposited on site while the sludge cake was disposed off site as a nonhazardous waste. Performance data showed that the cleaned sand and process oversize met the cleanup levels for 11 metals in this application.
This application was the first full-scale application of soil washing to remediate a Superfund site in the United States. In addition, a selective excavation technique was used to collect and identify contaminated soil and sludge for treatment in the soil washing unit, and the associated use of advanced on-site monitoring techniques. Selective excavation was performed through visual determination of contaminated material and confirmation of clean materials on site with an X-ray fluorescence instrument in an on-site laboratory. This excavation technique resulted in the processing of fewer tons of soil requiring soil washing than would have occurred with a less discriminating excavation technique.
Actual costs for the soil washing treatment application at the King of Prussia site, including off-site disposal costs, were approximately $7,700,000.
Table of Contents | Forward to Site Information
King of Prussia Technical Corporation Operable Unit 1
Winslow Township, New Jersey
CERCLIS #: NJD980505341
ROD Date: 28 September 1990
Type of Action: Remedial
Treatability Study Associated with Application? Yes (Refer to Appendix A
for additional information on treatability study and Appendix B for information
on demonstration run.)
EPA SITE Program Test Associated with Application? No
Operating Period: 6/28/93 to 10/10/93
Quantity of Soil Treated During Application: 19,200 tons
Historical Activity That Generated Contamination at the Site: Waste
processing facility
Corresponding SIC Code: 4953: Sanitary Services Refuse Systems
Waste Management Practices That Contributed to Contamination: Surface
impoundment/lagoon; and dumping unauthorized
Site History: The King of Prussia (KOP) Technical Corporation site is located in Winslow Township, Camden County, New Jersey, as shown in Figure 1. The site, a rectangular shaped, 10-acre parcel, as shown in Figure 2, is bordered to the northeast, northwest, and southwest by a dense pine forest of the state-owned 6,000-acre Winslow Wildlife Management Area. The southeast border is Piney Hollow Road. The Great Egg Harbor River, used for recreational purposes, is located approximately 1,000 feet southwest of the site. A drainage swale in the site is dammed by two fire roads; site runoff flows toward the river. The swale has been designated as a wetlands. The site is generally barren and sandy with sparse patches of tall seed grass. [1 and 9]
Figure 1. Site Location
Figure 2. Site Map [9]
The KOP Corporation began operating a waste recycling facility at this site in January 1971. The facility included six lagoons used to process liquid industrial waste. Industrial wastes were converted to materials that were intended to be marketed and sold as construction material and for other uses. Excess materials were transferred to other disposal locations. During its operation, it is estimated that at least 15 million gallons of acids and alkaline aqueous wastes were processed at this site. Site operations are believed to have ceased and site abandonment to have occurred in late 1973 to early 1974. In addition, between 1976 and 1988, illegal dumping of trash and hazardous materials was suspected to have occurred at the site. [1 and 9]
Soil and sediment at the site were determined to be contaminated with heavy metals. Prior to issuance of a ROD, cleanup activities at the site included excavation and removal for off-site disposal of buried plastic containers (carboys) and visibly-contaminated, surrounding soils located west of the lagoons. [1]
Regulatory Context: A ROD was issued for this site in September 1990 and defined five components of remedial activities pertaining to contaminated media, including the area relevant to this report (i.e., Component 1). These components included [1, 12]:
Component 1—The metals-contaminated soils adjacent to the lagoons, the sludge in the lagoons, and the sediment in the swale. (Operable Unit One)
Component 2—The buried drums and soils contaminated with volatile organic compounds located in the northwest section of the site. (Operable Unit Two)
Component 3—Two tankers and their contents located near the southeast sections of the site.
Component 4—The groundwater at the site contaminated with organics and metals. (Operable Unit Three)
Component 5—The surface waters, sediments, and biota of the Great Egg Harbor River.
EPA issued a Unilateral Administrative Order to the PRPs in April 1991 requiring the PRPs to implement the requirements of the ROD. The remedial activities for Component 1 were led by the PRPs with EPA oversight. [9]
Remedy Selection: The following six remedial alternatives were considered for remediation of Component 1 of the KOP site:
Soil washing was selected as the remedial alternative for Component 1. Soil was determined to provide a permanent solution by removing the contaminants from the site and thus protecting human health and the environment. In addition, the treated material could be redeposited to its original location to restore site topography. [1]
Site Management: PRP Lead
Oversight: EPA
Remedial Project Manager:
Gary Adamkiewicz (through May
1994)
John Gorin (June 1994 to Present)
U.S. EPA Region 2
26 Federal Plaza, Rm. 720
New York, NY 10278
(212) 264-7592
Treatment System Vendor:
Jill Besch/Mike Mann
Alternative Remedial Technologies, Inc.
14497 Dale Mabry Highway
Tampa, FL 33618
(813) 264-3506
Back to Executive Summary |
Table of Contents | Forward
to Matrix Description
Type of Matrix Processed Through the Treatment System: Soil (ex situ)/Sediment (ex situ)/Sludge (ex situ)
Primary Contaminant Group: Heavy metals
Investigations at the site were conducted by the New Jersey Department of Environmental Protection and by the PRPs. Samples of surface soil (<2 feet deep), subsurface soil (2 to 10 feet), and sediment were collected during the investigations to characterize the soil next to the lagoons, the sediments in the swale, and the sludges in the lagoons and adjacent areas. The results from this sampling indicated that beryllium, chromium, copper, nickel, and zinc are the primary contaminants in these areas. The highest concentration of surface contamination was located in the sediments at the bottom of the swale, with maximum concentrations of chromium at 8,010 mg/kg, copper at 9,070 mg/kg, and mercury at 100 mg/kg. The highest concentrations of subsurface contamination were located in a zone of sludge-like material at a depth of 3 to 4 feet northwest of and adjacent to the lagoons. The highest concentrations of contaminants in the sludge material were chromium at 11,300 mg/kg, copper at 16,300 mg/kg, lead at 389 mg/kg, and nickel at 11,100 mg/kg. Sampling results also indicated that the soils have infrequent and low concentrations of volatile and semivolatile organic compounds. Average soil concentrations were measured as 660 mg/kg for chromium, 860 mg/kg for copper, and 330 mg/kg for nickel. [1, 9, 12]
Listed below in Tables 1 and 2 are selected matrix characteristics which are considered to be the major matrix characteristics affecting cost or performance, and the values measured for each.
Table 1. Matrix Characteristics Affecting Treatment Cost or Performance [5, 10]
| Parameter | Value | Measurement Procedure |
|---|---|---|
| Clay Content and/or Particle Size Distribution | See Table 2 | Not available |
| Fines Content | 0.1 | Wet screening |
| Total Organic Carbon | Not measured | -- |
| Cation Exchange Capacity | Not measured | -- |
Table 2. Particle Size Distribution of Background Soil [5]
| Particle Size (microns) | Distribution (%) |
|---|---|
| >4,000 | 0 |
| 2,000 to 4,000 | 12.6 |
| 1,000 to 2,000 | 12.6 |
| 500 to 1,000 | 22.1 |
| 250 to 500 | 28.8 |
| 125 to 250 | 12.5 |
| 63 to 125 | 3.9 |
| 38 to 63 | 0.9 |
| <38 | 6.6 |
Back to Site Information | Table of Contents | Forward to Treatment System Description
Soil Washing
Screening
Technology Description
Excavation Description [7, 10]
Materials Handling: Selective excavation of metals-contaminated soils was completed using visual inspection and confirmed using an X-ray fluorescence (XRF) instrument in an on-site laboratory. Although 40,000 tons of material were excavated, only 20,000 tons exceeded the cleanup levels and required treatment through the soil washing unit. Selective excavation was identified as an appropriate technique for this site based on the findings of previous site investigation and excavation activities which indicated that the contaminants are associated within bands of sludge material and soils adjacent to the lagoons. Selective excavation of the soil and sludge in and adjacent to the lagoons and the swale area involved the following steps:
Excavation and blending of soils and sludges to maintain a constant ratio of soil to sludge involved the following three phases:
X-Ray Fluorescence: An X-ray fluorescence (XRF) instrument was used on-site during the excavation activities and during the soil washing operation for the analysis of chromium, copper, and nickel. An XRF instrument was also utilized during pre-remedial activities, including additional site characterization, the treatability study, and the demonstration run. For the treatability study and demonstration run, the XRF was calibrated with both synthetic and commercial standard reference materials. Confirmational analysis performed by an outside Contract Laboratory Program (CLP) laboratory indicated that the field results for chromium and copper were biased high by a factor of 1.3 to 2. It was determined that both synthetic and commercial calibration standards were not suitable for the concentrations and matrices encountered at the KOP site. Therefore, the XRF results relevant to the treatability study and demonstration run for this application were considered to be biased high by a factor of 1.3 to 2.
Based on a review of the confirmational analyses and calibration procedures used for the XRF instrument during the runs described above, the vendor modified the calibration standards. Calibration standards were developed for the full-scale application using samples of contaminated soil from the KOP site. Initial efforts to develop suitable calibration standards involved collecting contaminated soil from the site, manual homogenization, grinding, splitting and off-site laboratory analysis. Continuing studies for developing suitable standards resulted in refining the soil sample preparation method by replacing the manual homogenization, grinding, and splitting processes with mechanical processes for each item.
For the full-scale activities, three calibration standards, corresponding to concentrations less than, approximately equal to, and greater than the ROD-specified cleanup levels, were prepared for chromium, copper, and nickel using the refined technique and were used to calibrate the XRF instrument. The results obtained with the XRF using the mechanically prepared calibration standards showed no bias in the correlation with off-site confirmatory analysis.
Soil Washing System Description
The soil washing unit used to remediate the contaminated soil and sludge at the KOP site was constructed by a Swedish-based firm under contract to Alternative Remedial Technologies, Inc. The unit, shown in Figure 3, consists of four components: screening, separation, froth flotation, and sludge management (described below), and has a rated system throughput of 25 tons/hour.
Figure 3. Soil Washing Unit Used at KOP [6]
The soil washing unit was built off site as a modular system, and constructed at the site, as shown in Figure 4. Construction activities began on March 30, 1993, and were completed on June 1, 1993. Following completion, a slurry run, comprised of clean site soils and water, was conducted to monitor operation of the unit. To verify that the newly erected unit was capable of treating the contaminated soil to the ROD cleanup levels, a pilot run was performed form June 3 through June 9, 1993. The pilot run consisted of processing 991 tons of contaminated soil from Lagoons 1 and 6 and the sludge band area.
Figure 4. Remediation System Layout [12]
System operation included the following processes:
Screening: This stage consists of screening out the gross oversize fraction from the pile of material to be treated by means of a hopper and a vibrating grizzly (not shown on Figure 3). The gross oversize (greater than 8-inch material), which typically consists of concrete, tree stumps, and branches, is periodically removed from the hopper and staged. The material that passes through the grizzly is then directed to another mechanical screening unit, which consists of a double-decked, coarse vibrating screen with stacking conveyors, to remove process oversize (greater than 2-inch material) from the fall-through. The fall-through (<2 inch) is then subjected to wet screening with high pressure water nozzles. The wet screening breaks up clods, drops out pea-size gravel and forms a slurry. Gravel and other material is combined with the process oversize, while the slurry is further separated.
Separation: This stage consists of separating the screened soil/water slurry into coarse- and fine-grained material through the use of multi-stage hydrocyclones. The use of multiple cyclones achieves a separation efficiency of >99% of the sands and fines. The hydrocyclones have field-adjustable cone and barrel components to set and modify as necessary the "cut-point" between coarse- and fine-grained material. For this application, the hydrocyclone cut point was set at 40 microns (the distribution among size fractions showed a diminishing removal efficiency above 40 microns), determined using the results of the treatability study. The hydrocyclones were configured to minimize the volume of sludge cake requiring off-site disposal and to minimize the amount of fines in the clean product. The underflow containing coarse-grained material from the hydrocycloning steps was conditioned and directed to the froth flotation stage while the fine-grained material was processed into a sludge cake.
Froth Flotation: This stage consists of removing the contaminants from the coarse-grained material. The removal was done by means of air flotation treatment units. For this application, an air-flotation tank equipped with mechanical aerators was used. The coarse-grained material was pumped into the tank where a surfactant was added. The surfactant, selected based on the results of the treatability test, reduced the surface tension between the contaminant and sand. The contaminants "float" into a froth and were removed from the surface of the air flotation tank and were directed to the sludge management process. Surfactant dosing, slurry flow rate, and the height of the overflow weir were continuously monitored and adjusted as appropriate. The "cleaned" underflow sands were directed to a cyclone and sand dewatering screens, where dewatering occurs. Approximately 85% of the processed material (clean sand product) from the KOP site was used as backfill, while the water was recycled back to the wet screening section.
Sludge Management: This stage of the process consists of treating the overflow from the hydrocyclones. The overflow, consisting of fine-grained material and water, was pumped to banked Lamella clarifiers. A polymer, selected based on the results of the treatability test, was added prior to introduction to the Lamella. The clarified solids were directed to a sludge thickener and ultimately to a pressurized filter press, where the 15-20% solids influent was converted into a 50-60% dry solids filter cake. The filter cake was disposed off site as a nonhazardous waste. The water from the sludge management stage was returned to the wet screening area for reuse.
The major operating parameters affecting cost or performance for this technology and the values measured for each during this treatment application are listed in Table 3.
Table 3. Operating Parameters Affecting Treatment Cost or Performance [3, 10]
| Parameter | Value* |
|---|---|
| Moisture Content (of untreated soil) | ~15% |
| pH (of untreated soil) | ~6.5 |
| System Throughput | 25 tons/hr |
| Washing/Flushing Solvent Components/Additives | Polymer and Surfactant |
*Vendor provided approximate values for moisture content and pH but did not identify the specific polymer and surfactant used in this treatment application.
ART operated the soil washing unit at KOP on a production basis, with a goal of processing 25 tons/hour of contaminated materials, and monitored and adjusted 15 operational parameters. These parameters included the pH of the conditioners and make-up streams, metering of process streams (frother, conditioners, and polymers), cyclone feed rates, operational heights of process vessels (sumps and conditioner tanks), and operating pressures of pumps and cyclones. [6, 10]
A timeline for this application is shown in Table 4.
Table 4. Timeline [1, 3, 7, 9, 11, and 12]
| Start Date | End Date | Activity |
|---|---|---|
| January 1971 | April 1974 | Operations at the KOP Technical Corporation conducted |
| September 1983 | -- | KOP added to National Priorities List |
| September 28, 1990 | -- | ROD signed |
| January 1992 | -- | Treatability test conducted |
| July 22, 1992 | -- | Demonstration run conducted |
| March 1, 1993 | November 4, 1993 | Site mobilization |
| March 30, 1993 | June 1, 1993 | Construction of soil washing unit |
| June 3, 1993 | June 9, 1993 | Pilot run conducted |
| June 28, 1993 | October 10, 1993 | Full-scale soil washing conducted |
| July 8, 1993 | October 13, 1993 | Off-site shipment of residual sludge |
| July 19, 1993 | October 10, 1993 | Backfilling of clean soils |
| October 11, 1993 | November 1, 1993 | Decontamination and disassembly of soil washing unit |
Back to Matrix Description | Table of Contents | Forward to Treatment System Performance
The 1990 ROD identified cleanup levels for 11 metals in the soils in the area adjacent to the lagoons, sediments in the swale, and sludges in the lagoons (Component 1 of the site remediation). These levels are presented in Table 5. [1]
Table 5. Soil Cleanup Levels [1]
| Constituent | Soil Cleanup Levels (mg/kg) |
|---|---|
| Arsenic | 190 |
| Beryllium | 485 |
| Cadmium | 107 |
| Chromium (total) | 483 |
| Copper | 3,571 |
| Lead | 500 |
| Mercury | 1 |
| Nickel | 1,935 |
| Selenium | 4 |
| Silver | 5 |
| Zinc | 3,800 |
The cleanup levels shown in Table 5 were developed based on risk to public health using carcinogenic and noncarcinogenic effects. The carcinogenic effects were assessed using the cancer potency factors developed by the U.S. EPA, and a cancer risk of less than 1 × 10-6. The noncarcinogenic effects were assessed using the hazard index approach, based on a comparison of expected contaminant intakes and Reference Doses. A hazard index of less than 1 was used to develop the cleanup levels from noncarcinogenic risks. The carcinogenic and noncarcinogenic risks were summed to indicate the potential risks associated with mixtures of potential carcinogens and noncarcinogens. [1]
Table 6 presents a summary of the treatment performance data for this application, corresponding to the four sampling points shown in Figure 3 and described below. Average concentrations and concentration ranges are provided for the untreated soil, process oversize, and clean sand, while only average concentrations are shown for the sludge cake.
Table 6. Treatment Performance Data [9,12]
Constituent |
Cleanup Level (mg/kg) |
Untreated (Feed) Soil Concentration (mg/kg) |
Process (Clean) Oversize Concentration (mg/kg) |
Clean Sand Product Concentration (mg/kg) |
Sludge Cake Average Concentration (mg/kg) |
|||
|---|---|---|---|---|---|---|---|---|
| Average | Range | Average | Range | Average | Range | |||
| Arsenic | 190 | 1 | N/A | 0.62 | 0.34 to 1.4 | ND (0.31) | ND (0.39) | N/A |
| Beryllium | 485 | 20 | N/A | 5.9 | 2.7 to 11 | 1.9 | 0.93 to 3.1 | N/A |
| Cadmium | 107 | 0.56 | N/A | ND (0.63) | ND (0.97) | 0.64 | ND (0.95) | N/A |
| Chromium | 483 | 660 | 500 to 5,000 | 172 | 81 to 310 | 73 | 37 to 94 | 4,700 |
| Copper | 3,571 | 860 | 800 to 8,000 | 350 | 170 to 580 | 110 | 52 to 158 | 5,900 |
| Lead | 500 | 22 | N/A | 6.5 | 3.1 to 14 | 3.9 | 2.6 to 6.1 | N/A |
| Mercury | 1 | 0.09 | N/A | ND (0.09) | ND (0.10) | ND (0.09) | ND (0.10) | N/A |
| Nickel | 1,935 | 330 | 300 to 3,500 | 98 | 58 to 150 | 25 | 18 to 38 | 2,300 |
| Selenium | 4 | 0.36 | N/A | ND (0.38) | ND (0.40) | ND (0.36) | ND (0.40) | N/A |
| Silver | 5 | 0.69 | N/A | ND (0.65) | ND (0.76) | ND (0.65) | ND (0.73) | N/A |
| Zinc | 3,800 | 150 | N/A | 48 | 27 to 76 | 16 | 9.4 to 22 | N/A |
N/A - Samples were not collected - see text.
ND - Not detected (detection limit shown in parentheses).
A review of the treatment performance data in Table 6 indicates that the process oversize and clean sand from the soil washing unit met the cleanup levels established for this application. As shown in Table 6, the average concentrations of beryllium, copper, lead, nickel, and zinc in the clean sand and process oversize were at least an order of magnitude lower than the cleanup levels. Cadmium, mercury, selenium, and silver were not detected in any process oversize samples; and arsenic, mercury, selenium, and silver were not detected in any clean sand samples.
The data in Table 6 show that chromium, copper, and nickel were concentrated in the sludge cake, with individual contaminants measured at levels greater than 2,000 mg/kg.
The available performance data characterize constituent concentrations in the untreated soil, process oversize, clean sand, and sludge cake residual. Data are not available for matching specific operating conditions with treatment performance.
The CLP SOW for Inorganic Analysis includes analysis of initial and continuing calibration checks, duplicates, matrix spike, and reagent blanks. No exceptions to the QA/QC protocol were noted by the vendor. [7]
Back to Treatment System Description | Table of Contents | Forward to Treatment System Cost
ART, Inc., was under contract to the PRPs to construct and operate the soil washing treatment at the site. ART used several subcontractors to assist in the application, including activities associated with excavation, construction, and materials handling. [7, 12]
Approximately $7.7 million were expended on the soil washing remediation at KOP, including all off-site disposal costs. [12]
No information is presented in the references available at this time to describe the items included in the $7.7 million value. Therefore, a cost breakdown using the interagency Work Breakdown Structure (WBS) is not provided in this report.
The cost data shown above were provided by the Project Coordinator for the PRPs, and are provided in the Remedial Action Report for this application. A detailed breakdown of the cost elements is not available at this time.
Back to Treatment System Performance | Table of Contents | Forward to Observations and Lessons Learned
- Increasing the bed length and redesigning the spray headers on the wet screen unit to prevent bypassing or short-circuiting of the feed soil;
- Using an alternate frother to reduce frothing;
- Load balancing to the hydrocyclones; and
- Selecting filtration-aided polymers to produce the densest sludge cake possible.
Back to Treatment System Cost | Table of Contents | Forward to References
This case study was prepared for the U.S. Environmental Protection Agency's Office of Solid Waste and Emergency Response, Technology Innovation Office. Assistance was provided by Radian Corporation under EPA Contract No. 68-W3-0001.
Back to Observations and Lessons Learned | Table of Contents | Forward to Appendix A - Treatability Study Results
| Identifying Information | |
| King of Prussia Superfund
Site Winslow Township, New Jersey |
CERCLIS#: NJD980505341 ROD Date: 28 September 1990 |
| Historical Activity at Site - SIC Codes: | 4953 Sanitary Services-Refuse Systems |
| Historical Activity at Site - Management Practices: | Waste processing facility |
| Site Contaminants: | Metals, primarily chromium, copper, and nickel |
| Type of Action: | Remedial |
| Did the ROD/Action Memorandum include a contingency on treatability study results? | No |
| Treatability Study Information | |
| Type of Treatability Study: | Laboratory screening, bench-scale testing, and pilot scale testing |
| Duration of Treatability Study: | January 15, 1992 to March 27, 1992 |
| Media Treated: | Soil (ex situ) |
| Quantity Treated: | 188 kg |
| Treatment Technology | Soil washing |
| Target Contaminants of Concern: | Chromium, copper, and nickel |
| Conducted before the ROD was signed? | No |
| Additional treatability studies conducted: | None identified at this time |
| Technology selected for full-scale application: | Yes |
| Treatability Study Strategy | |
| Number of Runs: | A minimum of 1 test was conducted for each unit of the soil washing system, with additional tests performed where necessary. The entire system was run 3 times during the process simulation tests. |
| Key Operating Parameters Varied: | Hydroclone Test: surfactant concentration,
pH, retention time, pretreatment Fines/Sludge Handling Test: polymer |
| Treatability Study Results | |
| Range of Concentrations of Metals in Soils Treated During Pilot-Scale (Process Simulation) Runs: | Cu: 62 ppm to 1,500 ppm Ni: 18 ppm to 86 ppm Cr: 13 ppm to 130 ppm |
Treatability Study Objectives
The treatability study on the King of Prussia Technical Corporation Superfund site soil consisted of the following three steps:
The laboratory screening step was performed to characterize the soil and to collect enough information to make a soil washing feasibility determination. The bench-scale testing step was performed to select and optimize the appropriate treatment unit operations for the separation and removal of target metals from the coarse-grained and fine-grained source fractions. The pilot-scale testing step was performed to determine the system operating conditions, equipment lists, utility, chemical, and personnel requirements, and to refine the capital and operating cost estimates for the full-scale operation. [5]
Treatability Study Test Description [5]
Soil was collected from eleven locations at the KOP site in January 1992. One 5-gallon bucket of soil/sediment was collected, packed and shipped to the Heidemij Reststoffendiensten treatability lab located in the Moerdijk, Netherlands for treatability testing. [5]
Laboratory Screening: Soil characterization efforts included the chemical analyses of the initial (influent) soil samples for chromium, copper, nickel, mercury, and silver. These metals were analyzed using the Dutch equivalent to SW-846 7000 series methods. Each influent soil was physically screened/sieved to define the particle size distribution. Each fraction was analyzed for chromium, copper, and nickel to determine contaminant concentrations. Scanning electron microscopy was performed to determine the physical form of the contaminants.
Bench-Scale Testing: Tests were performed on hydrocycloning, flotation, gravity separation, and sludge management by coagulation, thickening, and dewatering unit operations using soil from lagoons 1 and 6.
The hydrocycloning operation test involved processing the soil through a 5" hydrocyclone test unit at different cut points and screening/sieving the underflow and overflow fractions.
The flotation tests involved selecting a suitable surfactant and concentration and retention time for this unit operation. One sample of the sludge band soil following wet screening was used for the flotation studies, which included varying surfactant concentrations, pH, retention time, and pretreatment (attritioning scrubbing).
The gravity separation operation test involved the use of a standard lab separator/shaking table to divide a wet-screened sample of the sludge band soil and lagoon composite soil to promote additional source separation.
The sludge operation test involved four organic polymers at four dosage concentrations on the overflow (fines and water) from the hydrocycloning test.
Pilot-Scale Testing: For this test, each of the optimum unit operations evaluated in the previous steps were combined into a batch feed process system. The system consisted of a vibrating screen, three hydrocyclones, a froth flotation cell, and a spiral concentrator. Three process simulation test runs were designed and conducted for the lagoon 1 soil, lagoon 6 soil, and the sludge band soil. The sand and sludge generated from the simulation runs were collected and analyzed. The sludge cake was further subjected to a TCLP analysis for chromium.
Treatability Study Performance Data
Laboratory Screening Step: The particle size distribution curves in Figure A-1 developed during the laboratory screening show the relative amounts of coarse and fine-grained sized materials in the soil and sludge tested. The concentrations of metals in each size fraction of the lagoon composite sample is shown in Table A-1. These results indicate that lagoons 1 and 6 and the sludge band area contained native soil material that might be amenable to soil washing treatment; however, lagoon 4 consisted exclusively of non-soil material with a high fines content and would not likely be amenable to soil washing treatment. Only soil from lagoons 1 and 6 and the sludge band area were further subjected to bench-scale testing. [5]
Figure A-1. Particle Size Distribution Curves
Table A-1. Particle Size Distribution and Contaminant Concentrations
[5]
Lagoon Composite Sample
| Concentration (ppm) | ||||
|---|---|---|---|---|
| Size Fractions (microns) | Distribution (%) | Cu | Ni | Cr |
| >40,000 | 0.7 | |||
| 10,000 to 40,000 | 3.8 | 18,000 | 3,900 | 1,600 |
| 4,000 to 10,000 | 2.4 | 18,000 | 3,200 | 1,700 |
| 2,000 to 4,000 | 2.5 | 9,400 | 1,700 | 1,300 |
| 1,000 to 2,000 | 7.4 | 6,100 | 1,300 | 1,500 |
| 500 to 1,000 | 12.3 | 2,200 | 450 | 560 |
| 250 to 500 | 12.7 | 2,600 | 560 | 710 |
| 125 to 250 | 7.8 | 7,600 | 1,600 | 1,700 |
| 63 to 125 | 7.1 | 13,000 | 2,900 | 2,500 |
| 38 to 63 | 10.8 | 12,000 | 2,700 | 2,500 |
| 20 to 38 | 2.5 | 16,000 | 3,800 | 4,200 |
| <20 | 29.9 | 12,000 | 3,400 | 4,400 |
| TOTAL | 100 | 9,215* | 2,227* | 2,407* |
*Calculated
Bench-Scale Testing: The results from the bench-scale test indicated that, for the hydrocycloning operation, a cut point for the KOP soil washing unit would be set at 40 microns. Also, for the flotation studies, a surfactant concentration of 240 gr/ton and a naturally-occurring pH with pretreatment by attrition scrubbing would provide the best flotation results. For the gravity separation tests, the results indicated that gravity separation would not be effective for treatment of KOP soils, because poor separation occurred and no shifts in contaminant concentrations were observed. Also, for the sludge operation, Mogul FL-5009 would lead to the best pre-settling performance and Mogul XH-1990 would lead to the best dewatering performance. A filter cake with a dry solids concentration of 52% was produced with a plate and frame filter press during the bench-scale test. [5]
Pilot-Scale Testing: The mass balance/recovery results from the pilot-scale testing indicate that the process simulation equipment treated the KOP soils to meet the target cleanup goals. The sludge from each process simulation run did not exceed the chromium TCLP limit; therefore, the sludge would not be considered a RCRA hazardous waste. [5]
Treatability Study Observations and Lessons Learned
Back to References | Table of Contents | Forward to Appendix B - Demonstration Run Results
Demonstration Run Objectives
A demonstration run using soil from the King of Prussia (KOP) Technical Corporation Superfund site was performed to confirm the findings of the treatability study and to expand upon the operating parameters relating to full-scale operations. Also, a successful demonstration run would reinforce the selection and application of the ROD-specified remedy, and thereby potentially streamline the review by EPA and hasten actual construction of the full-scale unit. [6]
Demonstration Run Description
Soil was selectively excavated from the KOP site in May 1992, in accordance with an EPA-approved excavation plan. The goal of the selective excavation was to excavate soils for the demonstration run that were representative of site conditions and also be biased high, with respect to the level of contamination, to confirm the ability of the treatment system to achieve the treatment standards. Approximately 164 short tons of soil were excavated from areas in and around lagoons 1 and 6, the swale and sludge band. An on-site x-ray fluorescence (XRF) instrument was used to screen targeted soils for excavation and to quantitatively determine the concentrations of copper, chromium, and nickel in the excavated soil. [6]
The excavated soil was placed into 200 1-ton super sacks. A composite sample of soil from each sack was analyzed with the XRF to ensure that the soil contained at least one metal above the ROD cleanup requirements. The sacks were then properly labelled for shipment of hazardous waste and transported to the Port of Newark, New Jersey. The sacks were loaded onto a ship of the Mediterranean Lines, transported to the Port of Rotterdam, and ultimately trucked to the Heidemij Restoffendiensten soil washing facility in Moerdijk, Netherlands for the demonstration run. The soil was screened and blended at the facility on July 18, 1992 and processed through the unit on July 22, 1992. The duration of the demonstration run was seven hours. The process residuals were returned to the United States on October 20, 1994, again through the Port of Newark. The oversize and product were returned to the KOP site as clean material and staged for restoration of the site, while the sludge cake was disposed at the GSX Pinewood Treatment, Storage, and Disposal Facility.
Pre-Processing Activities: The contents of each of the 200 super sacks were screened at 4 cm using a Grizzly vibrating bar to remove the gross oversize, which was weighed, combined, staged, and bagged for transport back to the U.S. The screened material was carefully blended and mixed to create a single feed pile.
Feeding: The feed pile was loaded into an apron feeder using a front-end loader. The feed rate was controlled as the material was fed to the feeder conveyor and into the first process unit.
Screening: The feed soils were screened to 2 mm using a vibrating wet screen. Oversize material was removed via conveyor, staged, and rebagged for return to the site. The soil/slurry underflow from the wet screening was then pumped to separation unit.
Separation: The underflow was processed through a 10" Mozley hydrocyclone, with subsequent processing of the fines and water and the coarse-grained material through separate 5" Mozley hydrocyclones. All three hydrocyclones were adjusted at a cut point of 40 microns. The underflow (coarse-grained material) from the separation unit was further processed through a froth flotation device while the fines were managed through a sludge dewatering unit.
Froth Flotation: The sand treatment train consists of a contact scrubber, where the surfactant is added, a froth flotation cell where treatment occurs, and a sand dewatering screen. The froth was further directed to the Lamella clarifiers. The sand was dewatered on an oscillating sand dewatering screen. The dewatered sand was moved by conveyor belt to a staging area where it was weighed and bagged.
Sludge Dewatering: The fines and water from the separation unit are processed through a flocculation unit, where coagulant was added and thickened on the Lamella clarifiers. The solids were dropped into the bottom hopper and the sludge was pumped to a belt filter press. The sludge was dewatered and moved to a staging area where it was weighed and bagged. During this demonstration run, 14 feed pile samples, 6 process oversize samples, 1 pre-flotation product sample, 22 sand product samples, 6 sludge cake samples (for total metals) and 2 sludge cake samples (for TCLP metals) were collected. The samples and split samples were analyzed primarily for chromium, copper, and nickel using CLP protocols by D.C. Griffith laboratory located in the Netherlands, and by IEA laboratory in North Carolina.
Demonstration Run Results
The results of the feed pile are presented in Table B-1; those of the clean sand product in Table B-2; and the sludge cake results are presented in Tables B-3 and B-4. These results indicate that the demonstration run was successful in meeting the stated objectives of treating the KOP soils to ROD-required levels with the soil washing unit configuration as recommended in the treatability study report.
Table B-1. Process Feed Material
[6]
King of Prussia Technical Site Demonstration Run
Moerdijk, The Netherlands
July 22, 1992
(all mg/kg)
| Sample | Cr | Cu | Ni | Dry Solids (%) | |||
|---|---|---|---|---|---|---|---|
| DCG | IEA | DCG | IEA | DCG | IEA | ||
| 1 | 790 | 872 | 1,600 | 1,470 | 433 | 409 | 83.5 |
| 2 | 745 | 1,600 | 415 | 83 | |||
| 3 | 705 | 759 | 1,300 | 1,080 | 408 | 357 | 85.5 |
| 4 | 705 | 1,400 | 420 | 85 | |||
| 5 | 910 | 982 | 1,850 | 2,170 | 660 | 639 | 82 |
| 6 | 815 | 1,900 | 473 | 85 | |||
| 7 | 855 | 1,080 | 1,500 | 1,310 | 460 | 368 | 83.5 |
| 8 | 710 | 1,250 | 393 | 86 | |||
| 9 | 735 | 675 | 1,250 | 1,110 | 435 | 378 | 86 |
| Average | 770 | 870 | 1,500 | 1,430 | 460 | 430 | 84.4 |
Per the agreed plan, all discrete process materials were mixed into a feed blend pile. Results of this activity were captured on video tape.
Efficiency of the blending operation and feed to the plant was measured via a series of nine (9) radial hollow stem auger borings, analyzed for contaminant metals chromium, copper, and nickel. In addition, five (5) samples were split for CLP analysis by IEA Laboratories in the United States.
Analysis of the nine samples by D.C. Griffith (DCG) showed good consistency with averages and ranges for each metal. CLP analysis by IEA on five split samples showed similar consistency and close agreement to the results generated by the Dutch laboratory. From these data, it was concluded that the feed pile was sufficiently blended to introduce a consistent feed to the process.
Table B2. Product Sand [6]
King of Prussia Technical Site Demonstration Run
Moerdijk, The Netherlands
July 22, 1992
(all mg/kg)
| Sample | Cr | Cu | Ni | Dry Solids (%) | |||
|---|---|---|---|---|---|---|---|
| DCG | IEA | DCG | IEA | DCG | IEA | ||
| 1 - 0900 | No sample taken, sand not discharging | ||||||
| 2 - 0930 | 98 | 195 | 41 | 90 | |||
| 3 - 1000 | 250 | 266 | 465 | 668 | 105 | 119 | 81 |
| 4 - 1030 | 185 | 370 | 73 | 83 | |||
| 5 - 1100 | 130 | 97 | 270 | 187 | 53 | 43 | 84 |
| 6 - 1130 | 115 | 240 | 46 | 84 | |||
| 7 - 1200 | 155 | 161 | 315 | 353 | 67 | 77 | 83 |
| 8 - 1230 | 76 | 145 | 33 | 84 | |||
| 9 - 1300 | 150 | 129 | 305 | 258 | 63 | 66 | 84 |
| 10 - 1330 | 140 | 280 | 54 | 84 | |||
| 11 - 1400 | 140 | 183 | 310 | 428 | 65 | 98 | 84 |
| 12 - 1430 | 235 | 520 | 120 | 81 | |||
| 13 - 1500 | 185 | 455 | 87 | 83 | |||
| 14 - 1530 | 205 | 465 | 97 | 86 | |||
| 15 - 1600 | 220 | 195 | 445 | 429 | 91 | 99 | 83 |
| 16 - 1630 | 205 | 430 | 89 | 83 | |||
| Average | 170 | 170 | 350 | 390 | 70 | 80 | 84 |
| Treatment Requirement | 483 | 3,571 | 1,935 | ||||
Table B-3. Sludge Cake Results
[6]
King of Prussia Technical Site Demonstration Run
Moerdijk, The Netherlands
July 22, 1992
(all mg/kg)
| Sample | Cr | Cu | Ni | Dry Solids(%) | |||
|---|---|---|---|---|---|---|---|
| DCG | IEA | DCG | IEA | DCG | IEA | ||
| 1 | 4,400 | 7,300 | 2,300 | 44 | |||
| 2 | 4,400 | 4,470 | 7,400 | 7,330 | 2,300 | 2,360 | 46 |
| 3 | 4,700 | 4,760 | 8,100 | 7,950 | 2,700 | 2,670 | 46 |
| 4 | 5,500 | 9,300 | 3,200 | 44 | 44 | ||
| Average | 4,750 | 4,615 | 8,030 | 7,640 | 2,630 | 2,515 | 45 |
This table tabulates the results of the produced sludge cake. The sludge cake contains the treated contaminants and will be disposed at an appropriate off-site facility.
Table B-4. Sludge Cake Results - TCLP
Metals [6]
King of Prussia Technical Site Demonstration Run
Moerdijk, The Netherlands
July 22, 1992
IEA Analyses Only
| TCLP Metal | Regulatory Standard (mg/L) | Results Sample Number (mg/L) |
|||
|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | ||
| Arsenic | 5 | <0.61 | <0.61 | <0.62 | <0.63 |
| Barium | 100 | <14 | <17 | <48 | <37 |
| Cadmium | 1 | <0.12 | <0.12 | <0.12 | <0.12 |
| Chromium | 5 | 2.1 | 1.8 | <0.65 | <0.67 |
| Mercury | 0.2 | <0.02 | <0.03 | <0.02 | <0.02 |
| Lead | 5 | <0.65 | <0.71 | <1.0 | <0.96 |
| Selenium | 1 | <0.11 | <0.11 | <0.11 | <0.11 |
| Silver | 5 | <0.60 | <0.60 | <0.60 | <0.63 |
The TCLP Metal Analyses confirm that the produced sludge cake does not exceed TCLP regulatory standards. The sludge cake is not the product of the treatment of any listed RCRA hazardous waste and does not demonstrate any hazardous characteristics.
Demonstration Run Observations and Lessons Learned
Back to Appendix A | Table of Contents | Forward to Appendix C - Full-Scale Analytical Results
Table C-1. KOP Production Composites
Process Oversize [12]
Constituent |
ROD Cleanup Level (mg/kg) | Date Sampled (week of) (mg/kg) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 7/2 | 7/8 | 7/16 | 7/23 | 7/30 | 8/6* | 8/13** | 8/27*** | 9/10 | 9/24 | 10/8 | 10/11 | ||
| Arsenic | 190 | 0.43 B | 0.34 U | 0.32 U | 0.36 U | 0.39 U | 0.45 B | 0.82 B | 0.50 B | 0.98 | 1.4 B | 0.76 B | 0.66 B |
| Beryllium | 485 | 5.3 | 3 | 3.1 | 2.7 | 2.7 | 6.8 | 7.4 | 7.2 | 9.6 | 11 | 7.3 | 4.5 |
| Cadmium | 107 | 0.36 U | 0.36 U | 0.57 U | 0.47 U | 0.45 U | 0.59 U | 0.57 B | 0.80 U | 0.80 U | 0.80 U | 0.80 U | 0.97 B |
| Chromium | 483 | 120 | 98 | 110 | 81 | 92 | 210 | 210 | 220 | 280 | 310 | 200 | 130 |
| Copper | 3,571 | 230 | 190 | 250 | 180 | 170 | 380 | 330 | 420 | 520 | 545 | 580 | 320 |
| Lead | 500 | 9.6 | 3.1 | 3.4 | 3.5 | 3.1 | 6.2 | 4.5 | 6.9 | 14 | 12 | 8.3 | 5.6 |
| Mercury | 1 | 0.09 U | 0.10 U | 0.09 U | 0.10 U | 0.10 U | 0.09 U | 0.10 U | 0.08 U | 0.10 U | 0.10 U | 0.10 U | 0.10 U |
| Nickel | 1,935 | 72 | 72 | 79 | 58 | 58 | 120 | 97 | 120 | 150 | 150 | 110 | 77 |
| Selenium | 4 | 0.36 U | 0.34 U | 0.32 U | 0.36 U | 0.39U | 0.39 U | 0.20 U | 0.20 U | 0.40 U | 0.40 U | 0.40 U | 0.40 U |
| Silver | 5 | 0.72 U | 0.72 U | 0.76 U | 0.63 U | 0.60U | 0.79 U | 0.60 U | 0.60 U | 0.60 U | 0.60 U | 0.80U | 0.60 U |
| Zinc | 3,800 | 29 | 28 | 34 | 26 | 27 | 69 | 50 | 71 | 76 | 68 | 59 | 39 |
*Last IEA Result
**First ITCorp Result
***Beginning of Two Week Composite
Table C-2. KOP Production Composites
Clean Sand [12]
Constituent |
ROD Cleanup Level (mg/kg) | Date Sampled (week of) (mg/kg) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 7/2 | 7/8 | 7/16 | 7/23 | 7/30 | 8/6* | 8/13** | 8/27*** | 9/10 | 9/24 | 10/8 | 10/11 | ||
| Arsenic | 190 | 0.36U | 0.37U | 0.34 U | 0.33U | 0.36 U | 0.36U | 0.39 B | 0.20U | 0.22 B | 0.36 B | 0.24 B | 0.20 B |
| Beryllium | 485 | 2.8 | 1.8 | 1.5 | 0.93 | 0.96 | 1.7 | 3.1 | 2.l | 2.6 | 2.3 | 1.9 | 1.8 |
| Cadmium | 107 | 0.36 U | 0.34 U | 0.49U | 0.53 U | 0.55U | 0.54 U | 0.76U | 0.80 U | 0.80 U | 0.95 B | 0.80 U | 0.80 U |
| Chromium | 483 | 73 | 58 | 63 | 38 | 37 | 62 | 94 | 61 | 70 | 63 | 57 | 44 |
| Copper | 3,571 | 150 | 100 | 100 | 61 | 52 | 85 | 140 | 110 | 158 | 150 | 150 | 100 |
| Lead | 500 | 6.1 | 3.9 | 3.3 | 3.3 | 2.6 | 2.6 | 3.4 | 3.5 | 4.3 | 3.4 | 3.4 | 3.6 |
| Mercury | 1 | 0.08 U | 0.09 U | 0.09 U | 0.08 U | 0.09U | 0.10 U | 0.10 U | 0.10 U | 0.10 U | 0.10 U | 0.10 U | 0.10 U |
| Nickel | 1,935 | 32 | 28 | 30 | 20 | 18 | 27 | 36 | 32 | 38 | 27 | 23 | 21 |
| Selenium | 4 | 0.36 U | 0.37 U | 0.34 U | 0.33 U | 0.36U | 0.36 U | 0.20 U | 0.20 U | 0.2 | 0.40 U | 0.40 U | 0.40 U |
| Silver | 5 | 0.73 U | 0.08 U | 0.65 U | 0.71 U | 0.73U | 0.71 U | 0.57 U | 0.60 U | 0.60 U | 0.59 U | 0.60U | 0.60 U |
| Zinc | 3,800 | 16 | 15 | 17 | 11 | 9.4 | 17 | 23 | 18 | 22 | 19 | 15 | 12 |
*Last IEA Result
**First IT Corp Result
***Beginning of Two Week Composite