&EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/540/R-96/535
September 1998
COGNISTERRAMET®
Lead Extraction Process
Innovative Technology
Evaluation Report
SUPERFUND INNOVATIVE
TECHNOLOGY EVALUATION
-------
-------
EPA/540/R-96/535
September 1998
COGNIS TERRAMET®
Lead Extraction Process
Innovative Technology Evaluation Report
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Printed on Recycled Paper
-------
Notice
The information in this document has been funded by the U. S. Environmental Protection Agency (EPA) under Contract No.
68-C5-0037 to Tetra Tech EM Inc. (formerly PRC Environmental Management, Inc.). It has been subjected to the Agency's
peer and administrative reviews and has been approved for publication as an EPA document Mention of trade names or
commercial products does not constitute an endorsement or recommendation for use.
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by the U.S. Congress with protecting the nation's land, air, and
water resources. Under a mandate of national environmental laws, EPA strives to formulate and implement actions leading
to a compatible balance between human activities and the ability of natural systems to support and nurture life. To meet this
mandate, EPA's research program is providing data and technical support for solving environmental problems today and
building a science knowledge base necessary to manage our ecological resources wisely, understand how pollutants affect our
health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is EPA's center for investigation of technological and
management approaches for reducing risks from threats to human health and the environment. The focus of NRMRL's
research program is on methods for the prevention and control of pollution to air, land, water, and subsurface resources;
protection of water quality in public water systems; remediation of contaminated sites and groundwater; and prevention and
control of indoor air pollution. The goal of this research effort is to catalyze development and implementation of innovative,
cost-effective environmental technologies; develop scientific and engineering information needed by EPA to support
regulatory and policy decisions; and provide technical support and information transfer to ensure effective implementation of
environmental regulations and strategies.
This Innovative Technology Evaluation Report has been produced as part of NRMRL's strategic long-term research plan. It
is published and made available by EPA's Office of Research and Development to assist the user community and to link
researchers with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
-------
Abstract
This Innovative Technology Evaluation Report documents an evaluation of the removal of lead from sands and fines fractions
of contaminated soils by the COGNIS TERRAMET® lead extraction process (COGNIS process). The COGNIS process was
developed by COGNIS, Inc., of Santa Rosa, California. The evaluation was performed under the U.S. Environmental
Prolection Agency's (EPA) Superfund Innovative Technology Evaluation (SITE) Program. The performance information
upon which this report is based is derived primarily from process samples that were collected during the 4-day SITE
demonstration period (August 2 to 5,1994) during operation of the COGNIS process on sands and fines separated from about
400 metric tons (tons) of contaminated soils. The contaminated soils were extracted from a small arms ammunition burning/
disposal area (Site F) at the Twin Cities Army Ammunition Plant (TCAAP) in New Brighton, Minnesota. Before, during, and
after the 4-day SITE demonstration sampling activity, the COGNIS system was operated in fulfillment of a full-scale
remediation contract for TCAAP Site F. About 20,000 tons of soil were processed under contract during 11 months of
operation. The COGNIS process was not specifically configured or operated during the 4-day SITE demonstration to
maximize lead removal, but rather to economically meet cleanup goals (for example, 300 milligrams per kilogram [mg/kg] for
lead). Also, the SITE demonstration sampling points, analytical procedures, and data reduction methods were established
independently, and differed substantially in some areas, from those employed for process control by COGNIS, Inc., and for
oversight of the full-scale TCAAP Site F soil remediation effort by the U.S. Army and the State of Minnesota.
The COGNIS process, as configured for TCAAP Site F, uses two parallel leaching systems — one for sand-sized particles,
and one for fines. The COGNIS process is expected to typically require pretreatment to classify the soil into sands and fines
fractions for the respective leaching systems, and to remove lead particulates amenable to size and density separation. During
TCAAP Site F remediation, Brice Environmental Services Corporation (BESCORP) performed the required separations. The
sands and fines streams produced by the BESCORP system were fed immediately and directly to the COGNIS process.
Some key results from the 4-day SITE demonstration include the following:
• The particle size distribution of the untreated soil consisted of 17 percent oversized material, 54 percent sands, and 28
percent fines.
• The 824 mg/kg mean lead concentration in untreated soil was reduced by 63 to 84 percent after treatment by both the
BESCORP and COGNIS processes.
• The mean daily lead concentration in the untreated sands stream (ranging from 292 to 434 mg/kg after BESCORP
process) was decreased by 21 to 45 percent after treatment by the COGNIS process.
• The mean daily lead concentration in the untreated fines stream (ranging from 342 to 548 mg/kg after BESCORP
process) was decreased by 72 to 83 percent after treatment by the COGNIS process.
• Of 35 untreated sands samples, 20 samples exceeded lead concentration of 300 mg/kg; of the 35 treated sands samples,
27 samples were less than or equal to a lead level of 300 mg/kg.
• Of the 35 untreated fines samples, 27 samples exceeded 300 mg/kg; of the 35 treated fines samples, all 35 samples were
less than 300 mg/kg.
• The mean daily lead concentrations of the treated soil (by combining 2 parts sands to 1 part fines, based on the sands to
fines ratio noted in the untreated soil) were less then 300 mg/kg on all 4 days during the SITE demonstration.
iv
-------
Contents
Acronyms, Abbreviations, and Symbols x
Conversion Factors xjj
Acknowledgments xjy
Executive Summary I
1 Introduction 4
1.1 Brief Description of SITE Program and Reports 4
1.2 Purpose of the ITER 5
1.3 Background 5
1.4 Technology Description 7
1.5 Key Contacts 8
2 Environmental Requirements Analysis 1 ]
2.1 ARARs Defined l j
2.2 Other Non-ARAR Requirements 12
2.3 Major ARARs and TBCs for the COGNIS Process 12
2.3.1 Chemical-Specific ARARs and TBCs for the COGNIS Process 12
2.3.2 Location-Specific ARARs and TBCs for the COGNIS Process 16
2.3.3 Action-Specific ARARs and TBCs for the COGNIS Process 16
2.3.4 Other Requirements of the COGNIS Process 17
2.4 Implementability of the Technology 18
2.5 Applicable Wastes 18
2.6 Key Features of the COGNIS Process 18
2.7 Availability and Transportability of Equipment 18
2.8 Site Support Requirements 20
2.9 Limitations of the Technology 20
3 Economic Analysis 22
3.1 Issues and Assumptions 22
3.1.1 Equipment and Operating Parameters 22
3.1.2 Site-Specific Factors 23
3.1.3 Base-Case Scenarios 24
3.2 Cost Categories 26
-------
Contents (continued)
3.2.1 Site Preparation Costs
3.2.2 Permitting and Regulatory Costs
3.2.3 Mobilization and Startup Costs ;
3.2.4 Capital Equipment Costs
3.2.5 Labor Costs
3.2.6 Supply Costs
3.2.7 Utility Costs
3.2.8 Effluent Treatment and Disposal Costs
3.2.9 Residual Waste Shipping and Handling Costs
3.2.10 Analytical Services Costs
3.2.1 1 Equipment Maintenance Costs
3.2.12 Site Demobilization Costs
3.3 Conclusions of Economic Analysis
4 Treament Effectiveness
4.1 Methodology
4.2 Physical and Chemical Analyses
4.3 Chemical Analyses
4.3.1 Feed Soil
4.3.2 Oversized Material
4.3.3 Untreated Sands
4.3.4 Treated Sands
4.3.5 Untreated Fines
4.3.6 Treated Fines
4.3.7 Lead Concentrates
4.3.8 Overall Treatment of Soils
4.3.9 Liquid Samples
4.3.10 Results of Toxicity Studies
4.4 Conclusions
5 Technology Status
6 References
26
26
27
27
28
28
29
30
30
31
31
31
33
35
35
39
41
41
45
45
45
48
51
51
51
52
54
55
59
73
VI
-------
Figures
1-1 TERRAMET Lead Extraction Process Sample Locations 9
4-1 Lead Concentration in Treated and Untreated Sands 47
4-2 Daily Average lead Concentration in Treated and Untreated Sands 47
4-3 Lead Removal Efficiency for the Sands Fraction 49
4-4 Lead Concentration in Treated and Untreated Fines 49
4-5 Daily Average Lead Concentration in Treated and Untreated Fines 50
4-6 Lead Removal Efficiency for the Fines Fraction 50
4-7 Lead Removal Efficiency of the COGNIS Process for Treating Sands and Fines 53
4-8 Lead Removal Efficiency of the Combined BESCORP and COGNIS Process 53
4-9 Blood-Lead Probability Density in Children (Age 0 to 6 Years) Exposed to
194Mg/kg Soil-Lead Concentration 55
4-10 Blood-Lead Probability Density in Children (Age 0 to 6 Years) Exposed to
371 Mg/kg Soil-Lead Concentration 56
4-11 Blood-Lead Probability Density in Children (Age 0 to 6 Years) Exposed to
824 Mg/kg Soil-Lead Concentration 57
vn
-------
Tables
ES-l Criteria Evaluation for the TERRAMET Lead Extraction Technology 3
1-1 Cleanup Goals for TCAAP Site F 6
2-1 ARARs and TBCs for the COGNIS Process 13
2-2 Other Requirements for the COGNIS Process 19
3-1 Costs Associated with the COGNIS Process 25
4-1 Outline of Sample Collection and Field Measurement Program 36
4-2 Particle Size Distribution Data 40
4-3 Lead Concentrations in the Feed Soil to the BESCORP Soil Washing Unit 42
4-4 Non-Lead Metal Concentrations in Soil Samples 43
4-5 Toxicity Characteristic Leaching Procedure Test Results 44
4-6 Lead Content in Treated and untreated Sands and Fines 46
4-7 Daily Average Lead Concentrations with Standard Deviations 48
4-8 Combined Average Lead Concentrations for COGNIS Process Soil 52
4-9 Average Lead Concentrations in the Liquid Streams 54
VIII
-------
Acronyms, Abbreviations, and Symbols
ARAR
ASTM
ATTIC
BESCORP
CAA
CAMU
CBR
CEC
CERCLA
CERI
CFR
CSCT
CWA
dL
DOT
ft2
ITER
kg
kW
kWh
L
LDR
MCL
MCLG
mg
mm
MPCA
MS
MSD
NA
NCP
Applicable or Relevant and Appropriate Requirements
American Society for Testing and Materials
Alternative Treatment Technology Information Center
Brice Environmental Service Corporation
Clean Air Amendments
Corrective Action Management Unit
California Bearing Ratio
Cation Exchange Capacity
Comprehensive Environmental Response, Compensation, and Liability Act
Center for Environmental Research Information
Code of Federal Regulations
Consortium for Site Characterization Technology
Clean Water Act
Deciliter
U.S. Department of Transportation
Square Foot
Innovative Technology Evaluation Report
Kilogram
Kilowatt
Kilowatt Hour
Liter
Land Disposal Restriction
Maximum Contaminant Level
Maximum Contaminant Level Goal
Milligram
Millimeter
Minnesota Pollution Control Agency
Matrix Spike
Matrix Spike Duplicate
Not Applicable or Not Available
National Contingency Plan
ix
-------
Acronyms, Abbreviations, and Symbols (continued)
ND Not Detected
PPE Personal Protective Equipment
ppm Part Per Million
PSD Particle Size Distribution
QAPP Quality Assurance Project Plan
RCRA Resource Conservation and Recovery Act
ROC RCRA Off-Site Coordinator
RPD Relative Percent Difference
SARA Superfund Amendments and Reauthorization Act
SD Standard Deviation
SDWA Safe Drinking Water Act
SG/RE Seed Germination/Root Elongation
SITE Superfund Innovative Technology Evaluation
SWDA Solid Waste Disposal Act
TBC To Be Considered
TCAAP Twin Cities Army Ammunitions Plant
TCLP Toxicity Characteristics Leaching procedure
TU Temporary Unit
VISnT Vendor Information System for Innovative Treatment Technologies
ug Microgram
urn Micron
-------
Conversion Factors
To Convert From
To
Multiply By
Length
Area:
Volume:
inch
foot
mile
square foot
acre
gallon
cubic foot
centimeter
meter
kilometer
square meter
square meter
liter
cubic meter
2.54
0.305
1.61
0.0929
4,047
3.78
0.0283
Mass:
pound
kilogram
0.454
Energy:
kilowatt-hour megajoule
3.60
Power:
kilowatt
horsepower
1.34
Temperature: (""Fahrenheit - 32) °Celsius
0.556
XI
-------
Acknowledgments
This report was prepared under the direction of Mr. Michael Royer, the EPA SITE project manager at the National Risk
Management Research Laboratory (NRMRL) in Edison, New Jersey. This report was prepared by Dr. Pinaki Banerjee, Ms.
Thcrese Gioia, and Mr. Jeff Swano of PRC Environmental Management, Inc. (PRC). The report was typed by Ms. Cheryl
Vaccarello; edited by Ms. Cheryl Finnegan of PRC; and reviewed by Ms. Ann Kern, Mr. Samuel Hayes, Mr. Michael Borst,
Mr. Richard Griffiths, and Mr. Ronald Turner of EPA, Mr. Dan Card of the Minnesota Pollution Control Agency (MPCA),
Mr. Pete Rissell of the U.S. Army Environmental Center, and Dr. Kenneth Partymiller and Mr. Peter Zelinskas of PRC.
Significant contributions were also made by the U.S. Army, COGNIS, BESCORP, Wenck Associates, Inc., Federal Cartridge
Corporation, and Mr. Tom Barounis from EPA Region 5. Bioassay testing was planned and executed by Ms. Lina W. Chang,
Mr. John Meier, and Dr. M. Kate Smith of Ecological Monitoring Research Branch, National Exposure Research Laboratory,
EPA.
xii
-------
Executive Summary
The COGNIS TERRAMET® lead extraction process
(COGNIS process) was evaluated at a Super-fund
Innovative Technology Evaluation (SITE) demonstration
at the Twin Cities Army Ammunition Plant (TCAAP) site
in New Brighton, Minnesota. The SITE demonstration
was conducted during the course of a much larger full-
scale remediation under the Resource Conservation and
Recovery Act (RCRA) hazardous waste management unit
closure process. The results of the demonstration and an
evaluation of the technology are provided below.
The U.S. Environmental Protection Agency (EPA)
developed the SITE Program in response to the Superfund
Amendments and Reauthorization Act of 1986. The
program's primary purpose is to maximize the use of
alternative treatment technologies. To this end, reliable
performance and cost data on innovative technologies are
developed during demonstrations where the technology is
used to treat a specific waste.
After completing each demonstration, EPA publishes an
Innovative Technology Evaluation Report (ITER)
designed to aid decision-makers in evaluating the
technology for consideration as an applicable cleanup
option. This report includes a description of the
technology and environmental requirements the technology
may need to meet, an economic analysis of treatment costs
using the technology, and the results of the demonstration.
The COGNIS process extracts and recovers lead and other
metals from contaminated soil, dust, sludge, or sediment.
In the full-scale remediation and SITE demonstration at
TCAAP, the COGNIS process was used in conjunction
with the Brice Environmental Services Corporation
(BESCORP) soil washing and size separation process.
The BESCORP process removes and washes oversized
material and then separates the remaining soil into sands
and fines fractions. The sands fractions are further treated
by density separation to isolate particulate metals as a
concentrate. Evaluation of the BESCORP process was not
an objective of this SITE demonstration. However,
measurements of feed soil to the BESCORP process and
product streams from it were necessary to characterize the
contaminated soil, the feed streams to the COGNIS
process, and the overall lead reduction produced by the
combined processes. The BESCORP process was
previously evaluated by the SITE demonstration program
(EPA/540/AR-93/503) during operation at an Alaskan
lead battery breaking site.
In the COGNIS portion of the process, the sands and fines
fractions are then separately leached by a proprietary
leachant. Leached metals are recovered through
electrochemical cells, producing a metal concentrate,
while regenerating the leachant for recycling. The
recovered lead can be recycled at a smelter. The washed
sands and fines are neutralized and blended with treated
oversized material before final disposition. The
wastewater can be disposed of in a publicly owned
treatment works (POTW) at the conclusion of treatment or
at the season'send.
The SITE demonstration showed that implementing the
remediation processes, which include site preparation,
mobilization, operation of the BESCORP and COGNIS
process equipment, waste disposal, and demobilization,
proceeded mostly without any major problems. The feed
soil was composed of 17 percent oversized material, 54
percent sand, and 28 percent fines. The BESCORP soil
washing process alone achieved lead removal efficiencies
ranging from 38 to 62 percent. The combined removal
efficiency of the COGNIS and BESCORP processes
ranged from 63 to 84 percent. The COGNIS sands
leaching process reduced lead concentrations by an overall
average of 28 percent and the COGNIS fines leaching
process reduced lead concentrations by an overall average
of 78 percent. Treated soil lead concentrations were all
less than 300 milligrams per kilogram (mg/kg). Three of
-------
the eight oversized material samples, separated and
washed by BESCORP, did not meet the maximum
allowable lead concentrations in the toxicity characteristics
leaching procedures extracts, and therefore require further
treatment before disposal. Recovered lead was recycled at
a lead smelting facility and the wastewater was discharged
to a POTW. The cost of soil treatment at TCAAP was
found to be $182 per ton for treating 10,000 metric tons
(tons) of soil; this unit cost includes the costs of soil
washing, soil leaching by the COGNIS process, site
preparation, and residuals shipping and handling.The
COGNIS process was also evaluated based on the nine
criteria used for decision-making in the Superfund
feasibility study process. This evaluation against the nine
criteria considers both the information generated from and
observations made during the SITE demonstration and the
overall TCAAP Site F remediation. Table ES-1 presents
the evaluation.
-------
Table ES-L Criteria Evaluation for the Terramet® Lead Extraction Technology
Criteria
Evaluation
Overall
Protection of
Human Health
and the
Environment
Compliance with
Federal ARARs
Long-Term
Effectiveness
and Permanence
Reduction of
Toxicity,
Mobility, or
Volume Through
Treatment
Short-Term
Effectiveness
Implementability
Cost
State Acceptance
Community
Acceptance
Protects human health and the
environment by eliminating
exposure to contaminants in
soil
Requires compliance with
RCRA treatment, storage, and
disposal regulations of a
hazardous waste
Effectively removes metals
contaminants from soil; treated
waste could be handled as fill
material; site may be suitable
for reuse after remediation
Significantly reduces the
volume of metals contaminants
in soil through treatment
Presents potential short-term
risks to workers and
community including exposure
to noise and contaminants
released to the air during
excavation and handling
Typically requires pretreatment
of contaminated soil to
produce separate sands and
fines fractions for treatment in
separate leaching circuits
$182 per ton of soil based on
10,000 tons of soil treated; this
includes the cost for soil
washing, removing ordnance,
preparing the site, and soil
leaching
Provides a permanent solution
• to contamination that is
preferable to other soil
remediation technologies
Minimal and manageable
short-term risks to the public
may increase community
acceptance
Prevents further contamination of
groundwater and off-site migration by
removing contaminants from soil
Excavation, as well as construction and
operation of on-site treatment unit, may
require compliance with local ARARs
Provides reliable, irreversible treatment of
contaminated soil
Reduces the toxicity of the soil by
removing lead to acceptable levels;
residual salt toxicity in treated soil may
require dilution in the form of rinsing or
rainfall to enable successful revegetation
Short-term risks are readily manageable
through common site health and safety
practices
High clay content in soil may cause clay
ball formation, and require increased
residence time for leaching and soil/water
separation; multiple metals present
treatment and recycling difficulties;
process is limited by cold or freezing
weather
Labor cost is major part of the total cost;
other significant cost items include
equipment, proprietary chemicals, and
residuals management
State regulatory authorities may require
that certain permits be obtained before
implementing the system, if conducted as
part of a RCRA corrective action; such as
a permit to operate a treatment system,
and a permit or approval to potentially
store contaminated soil for more than 90
days
Permanence and long-term effectiveness
of technology may increase community
acceptance
Requires measures to protect
workers and community during
excavation, handling, and treatment
of soil
Wastewater generated at the
conclusion of treatment requires
compliance with Clean Water Act
Involves some residuals treatment
or disposal (precipitated metals,
wastewater); process generates high
toxic concentrations of salts in
treated soil that can be effectively
reduced with rinsing
Lead in soil is reclaimed, resulting
in resource recovery
Achieves cleanup objectives in
fairly short amount of time
Availability of material and services
limited because of limited vendors;
equipment is transportable;
mobilization and demobilization
each take approximately 2 weeks
Disposal costs are reduced by lead
recovery smelters, depending on the
percent of lead in the material
Minnesota Pollution Control
Agency approved implementation
of the COGNIS process for soil
treatment at TC AAP as part of
RCRA corrective action process
Resource recovery (reclaiming lead)
may increase community
acceptance
-------
Section 1
Introduction
This section provides background information about the
Superfund Innovative Technology Evaluation (SITE)
program, discusses the purpose of this Innovative
Technology Evaluation Report (ITER) and describes the
COGNIS TERRAMET® lead extraction process (COGNIS
process). Additional information about the SITE
Program, this technology, and the demonstration site can
be obtained by contacting individuals listed in Section 1.5
of this report.
1.1 Brief Description of Program and
Reports
The SITE Program is a formal program established by the
U.S. Environmental Protection Agency (EPA) Office of
Solid Waste and Emergency Response (OSWER) and
Office of Research and Development (ORD) in response
to the Superfund Amendments and Reauthorization Act of
1986 (SARA). The SITE Program promotes the
development, demonstration, and implementation of new
or innovative technologies at Superfund sites across the
country.
The primary purpose of the SITE Program is to maximize
the use of alternatives in cleaning hazardous waste sites by
encouraging the development and demonstration of new,
innovative treatment and monitoring technologies. The
SITE Program consists of the following programs: the
Consortium for Site Characterization Technology (CSCT),
the Demonstration Program, and the Technology Transfer
Program.
Existing technologies that improve field monitoring and
site characterizations are identified in the CSCT program,
formerly the Monitoring and Measurement Technologies
Program. This program supports new technologies that
provide faster, more cost-effective contamination and site
assessment data. The CSCT Program also formulates the
protocols and standard operating procedures for
demonstrating methods and equipment.
The objective of the SITE Demonstration Program is to
develop reliable performance and cost data on innovative
technologies so that potential users may assess the
technology's site-specific applicability. Technologies
evaluated are either currently available or close to being
available for remediation of Superfund sites. SITE
demonstrations are conducted on hazardous waste sites
under conditions that closely simulate full-scale
remediation conditions, thus ensuring the usefulness and
reliability of the information collected. Technologies
chosen must be at the pilot- or full-scale stage, must be
innovative, and must have some advantage over existing
technologies. Data collected are used to assess the
performance of the technology, the potential need for pre-
and posttreatment processing of wastes, potential
operating problems, and the approximate costs. The
demonstrations also allow for evaluation of long-term
risks and operating and maintenance costs.
Cooperative arrangements between EPA, the site, and the
developer establish responsibilities for conducting the
demonstrations and evaluating the technology. The
developer is responsible for demonstrating the technology
at the selected site and is expected to pay any costs for
transport, operations, and removal of the equipment. EPA
is responsible for project planning, sampling and analysis,
quality assurance and quality control, preparing reports,
disseminating information, and the site is responsible for
transporting and disposing of treated waste materials and
site logistics.
The Technology Transfer Program disseminates technical
information on innovative technologies through various
activities in the Demonstration, and CSCT Programs.
These activities increase the awareness and promote the
-------
use of innovative technologies for assessment and
remediation at Superfund sites. The goal of technology
transfer activities is to develop communication among
individuals requiring up-to-date technical information.
The results of the COGNIS process demonstration are
published in two documents: the SITE Technology
Capsule and the ITER. The SITE Technology Capsule
provides relevant information on the technology,
emphasizing key features of the results of the SITE field
demonstration. Both the SITE Technology Capsule and
the ITER are intended to be used by remedial managers
making a detailed evaluation of the technology for a
specific site and waste.
1.2 Purpose of the ITER
The ITER provides information on the COGNIS process
and includes a comprehensive description of the
demonstration and its results. The ITER is intended for
use by EPA remedial project managers, contractors, and
other decision-makers for implementing specific remedial
actions. The ITER is designed to aid decision-makers in
evaluating specific technologies for further consideration
as an applicable option in a particular cleanup operation.
This report serves as a critical step in the development and
commercialization of a treatment technology.
To encourage the general use of demonstrated
technologies, EPA provides information regarding the
applicability of each technology to specific sites and
wastes. The ITER includes information on cost and site-
specific characteristics.
Each SITE demonstration evaluates the performance of a
technology in treating a specific waste. The waste
characteristics at other sites may differ from the
characteristics of the treated waste. Therefore, successful
field demonstration of a technology at a particular site
does not necessarily ensure that it will be applicable at
other sites. Data from the field demonstration may require
extrapolation to estimate the operating ranges in which the
technology will perform satisfactorily. Only limited
conclusions can be drawn from a single field
demonstration.
1.3 Background
In August, 1994, a demonstration of the COGNIS process
was conducted at the Twin Cities Army Ammunition Plant
(TCAAP) site in New Brighton, Minnesota. Since 1941,
operations at the TCAAP facility included the
manufacturing, testing, destroying, storing, and disposing
of small arms ammunition and related materials. Until the
1980s, Site F, which is located within the facility, was used
to destroy various explosives and tracer compounds.
Ordnance and scrap materials were also buried at Site F.
As a result, soils at the 10-acre site are contaminated with
lead and other metals. Site F is being addressed through a
Resource Conservation and Recovery Act (RCRA)
corrective action with the Minnesota Pollution Control
Agency (MPCA) serving as the lead agency for oversight.
TCAAP solicited proposals from technology vendors and
selected COGNIS to treat contaminated soils at Site F.
COGNIS began treating soils beginning in the fall of 1993
and treated about 20,000 tons of soil over an 11-month
period ending in July, 1995. TCAAP's contractor, Wenck
Associates, Inc. (Wenck), prepared a report documenting
the overall operation (Wenck 1995a).
The COGNIS process treats metal-contaminated soil
using two separate leaching systems. The first system
treats soil of larger particle size, defined as sands. The
second system treats soil of smaller particle size, defined
as fines. For the COGNIS process, sands are defined as
particles that range in size from less than 6.35 millimeters
(mm) in diameter to 106 microns (um) in diameter, and
fines are defined as particles of less than 106 um in
diameter. Contaminated soil may need to be pretreated
before treatment by the COGNIS process to achieve
acceptable particle size separation. The sands and fines
leaching systems use proprietary aqueous leaching
solutions to remove metals from soil particles. Metallic
ions are then recovered from the aqueous leachant The
recovered metallic ions are recycled at a smelter or
disposed of, and the aqueous leachant is recycled through
the treatment process.
During the demonstration, which was conducted as part of
an ongoing, full-scale cleanup, soils were excavated and
staged at Site F. The treatment equipment was placed at
Site D, 500 yards northwest of Site F. Before placing
equipment at Site D, the area was graded, paved, and
bermed. TCAAP also built containment areas where
treated soils were placed. Site D preparation occurred as
part of a previous incineration project at Site D. Soil
treatment was interrupted during the winter and began
again in early summer of 1994. The SITE demonstration
was conducted over a 4-day period, from August 2 through
5, 1994.
-------
As part of the remedial investigation at Site F, soil samples
were collected from various depths and analyzed for
several metals. The average concentration of lead detected
in the soil was approximately 700 milligrams per kilogram
(mg/kg). Other metals detected at levels above the cleanup
goal that are considered to be contaminants of concern at
the site include the following: antimony, cadmium,
chromium, copper, mercury, nickel, and silver. The
cleanup standards for this site are listed in Table 1-1.
These remediation goals are the enforceable standards
established by MPCA. The MPCA lead remediation goal
was established in accordance with a state standard that
existed at the time. This lead level in the soil does not
address leaching to groundwater at Site F, because lead
leaching to groundwater is not a concern based on depth to
groundwater and soil and groundwater sampling data.
Since that time the state standard changed from 300 mg/kg
to 100 mg/kg and EPA has established soil lead screening
levels of 400 mg/kg (residential use) and 1,200 mg/kg
(industrial use). The other metals remediation goals were
established at two standard deviations above calculated
background levels. Soils that meet the remediation goals
and are below the toxicity characteristics leaching
procedure (TCLP) standards are not considered RCRA
hazardous waste.
Table 1-1. Cleanup Goals for TCAAP Site F (mg/kg)
Metal Remediation Goals"
Notes:
a
Antimony
Cadmium
Chromium
Copper
Lead
Mercury6
Nickel
Silver*
Remediation
4.0
4.0
100.0
80.0
300.0
0.3
45.0
5.0
goals are the enforcea
standards
b Mercury and silver were not evaluated during this
project
The objectives of the SITE Program evaluation of the
COGNIS technology were:
To determine the effectiveness of the COGNIS
process in removing total lead from sands and fines
particles and leachable lead from oversized mate-
rials
•• To determine the potential need for pre- and post-
treatment of soils that are treated by the COGNIS
process
To evaluate implementability of the treatment sys-
tem
• To determine capital, operating, and maintenance
costs
COGNIS claims to have treatment systems capable of
removing most metals from contaminated soils. At
TCAAP Site F, the treatment system was configured to
treat the contaminants specific to Site F soil. Site F soil
was contaminated primarily with lead, but contained other
metals. The COGNIS process was proposed, selected, and
evaluated under the SITE Program as a lead extraction
process. Therefore, because lead was the primary
contaminant in Site F soils and lead leaching was the focus
of the SITE demonstration, lead is the critical parameter
for this project. The effectiveness of the COGNIS process
was evaluated by determining the degree to which total
lead in the feed soil, sands, and fines was reduced.
Reduction of TCLP leachable lead in oversized material
and treated soil was also evaluated. The effectiveness of
the process in reducing toxicity was assessed based on (1)
reduction in lead concentration in soil via the Uptake
Biokinetic Model, and (2) bioassay test results of various
types, such as seed germination, root elongation, and
earthworm mortality. Because the COGNIS process
removes metals other than lead, the demonstration also
monitored concentrations of other metals in various media.
The COGNIS process requires feedstock that is separated
into sands (<6.35 mm and >106 urn in diameter) and fines
(<106 urn in diameter). Site F soil required pretreatment
for particle size separation. BESCORP conducted
pretreatment processing of soils that consisted of soil
washing and size separation. BESCORP's pretreatment
system was monitored to provide background information
and key data for calculating COGNIS performance
parameters such as feed rate of soil and particle size
-------
distribution. During the SITE demonstration, clay balls
were observed in the oversize material from the
BESCORP system. These clay balls were not treated by
the COGNIS process. The need for posttreatment of
treated soil was determined by comparing concentrations
of lead and other metals with remediation goals. The need
for posttreatment of residuals other than the treated soil
was evaluated by reviewing the types and quantities of
residuals and whether or not they required additional
treatment or disposal. The types of residuals generated
during the SITE demonstration included live ordnance,
inactive ordnance components, oversized material from
the BESCORP process, lead concentrates from the
BESCORP and COGNIS processes, leachant, process
water, and rain water. Residuals handling is discussed in
Section 2.9 of this report.
The technology consists of a two-stage process.
BESCORP conducts the first stage, and COGNIS
conducts the second stage of the process. The first stage is
the physical separation stage during which oversized
material is removed by screening the untreated soil to
separate sands and fines. In addition to size separation, the
BESCORP process plays a role in removing particulate
lead, thereby reducing initial lead concentrations before
introduction to the COGNIS process. This preliminary
reduction saves COGNIS reagents and probably increases
the chances of meeting remedial goals at a given
processing rate. The second stage is the leaching and lead
recovery stage that results in the dissolution of metal
attached to the fines and sands and the recovery of the
metallic ions from the aqueous leachant. End products
include treated soils, rocks, debris, and metallic fragments
from soil washing; precipitated metals from the acid
leachant; and the leaching solution, which is recycled
through the process.
1.4 Technology Description
According to COGNIS, the system can treat most types of
metal contaminations including soluble ions or insoluble
metal oxides and salts bound by fine soil constituents such
as clays. These constituents can be treated by matching the
leachant with a specific substrate and type of metallic
contaminant;
A schematic of the COGNIS and BESCORP processes is
shown in Figure 1-1. The major components of the
combined system include the following:
Trommel. The trommel separates untreated soil that is
less than 6.35 mm (0.25 inch) in diameter from the
remaining soil. Soil less than 6.35 mm in diameter passes
through the screen and is processed further while the larger
oversized material is manually screened for ordnance and
debris and is then transferred to the treated soil pile.
Wet classifier. Fines are separated from sands within the
wet classifier. The fines are advanced to a clarifier and the
sands are transferred to the density separator.
Density separator. Each sands fraction is treated in a
density separator to remove lead and other dense particles.
The underflow is recovered as lead concentrate while the
overflow solids are advanced to the sands leaching unit.
Sands leaching unit. Sands are mixed with an acidified
leaching solution in the sands leaching unit. After
flocculation and clarification, the leachant containing lead
is advanced to the lead recovery cells. The leached sand is
neutralized and added to the treated soil pile.
Fines leaching unit. Fines are mixed with the leachant in
the fines leaching unit. The leachant containing lead is
flocculated and clarified and is advanced to the lead
recovery cells. The leached fines are neutralized and sent
to the fines dewatering unit.
Fines dewatering unit. The leached fines are dewatered
in the fines dewatering unit. The dewatered solids are
delivered to the treated soil pile while the leachant is
regenerated and reused as leachant.
Lead recovery cells. The lead-containing leachant passes
through lead recovery cells on which elemental lead is
deposited. The recovered lead is periodically washed from
the cells as lead concentrate. The regenerated leachant is
recycled and reused in the process.
The physical separation process begins when excavated
soil is hand-sorted for ordnance. Ordnance at Site F is
comprised of the metallic components associated with
military rifle ammunition. The soil is then moved from the
excavation site to a storage pile near the treatment
equipment. The soil is then loaded into a front-end loader
that is used to place contaminated soil into a feed hopper.
Soil continuously feeds from the feed hopper onto a
conveyor belt, where it is weighed and elevated into a large
rotating drum called a trommel. The trommel contains a
-------
screen with holes measuring 6.35 mm in diameter. The
soil within the trommel is sprayed with water and particles
measuring less than 6.35 mm in diameter pass through the
screen. The oversized material empties onto a conveyor
belt.
The smaller particles that pass through the holes within the
trommel are transferred into a classifier and are suspended
in an upward flow of water. The fines that are smaller than
106 fira in diameter exit at the top of the classifier, while
two sand streams consisting of particles larger than 106
Jim in diameter exit from the middle and bottom of the
classifier. The fines are advanced to a clarifier where a
flocculent is added and the mixture is allowed to settle.
The overflow from the clarifier is returned to the water
surge tank while the fines that settle to the bottom of the
clarifier are pumped to the fines leaching unit. The sand
streams then flow to a density separator (see Figure 1-1).
Each sand fraction is treated in a density separator to
remove lead and other dense particles. The underflow
from the density separation process is recovered as lead
concentrate. Overflow solids from the density separators
are combined and advanced by a sand screw to the sands
leaching unit.
Sands within the sands leaching unit are mixed with an
acidified leachant. The leachant, which contains lead, is
flocculated, clarified, and advanced to two electrochemical
cells that comprise the lead recovery system in the sands
leaching unit. Lead is deposited in the lead recovery cells
thereby preparing the leachant for reuse following
refortification with additional leaching agent. The leached
sand is (1) removed by a sand screw, (2) neutralized by
adding lime, and (3) added to the treated oversized
material on a conveyor belt. The resulting material is
transported to the treated soil pile on the conveyor belt.
The fines within the fines leaching unit are mixed with
acidified leachant in four mixer-clarifiers. Two streams of
leachant, both of which contain lead, are advanced to the
lead recovery system. The underflow solids from the last
mixer-clarifier are neutralized with lime and then
dewatered in a continuous centrifuge. The solids that
result from the centrifuge are transferred to the treated soil
pile, while the aqueous portion from the centrifuge is
returned to the leachant surge tank. The lead-containing
leachant within the lead recovery system passes through
six lead recovery cells where elemental lead is deposited
and lead-depleted leachant is produced. The lead-depleted
leachant is refortified with leaching agent and returned to
the leaching unit. The lead that collects in the cell cassettes
is periodically dewatered and drummed as lead
concentrate.
As part of the RCRA corrective action at Site F, upgrades
were made to the combined BESCORP/ COGNIS system.
The following upgrades were made in 1994:
• Spiral classifier
• Lifter bars to trommel
• Vegetative organic separation screens
• Acid storage tank
• Second centrifuge
The following upgrades were made in 1995:
• Log washer to deagglomerate clay balls, when re-
quired
• Second clarifier and pre-acidification vessel
• New flocculent
Fifth clarifier was added
1.5 Key Contacts
Additional information on the COGNIS process and the
SITE Program can be obtained from the following sources:
The COGNIS Process
Phillip Mattison
Henkel Corporation
300 Brookside Ave.
Ambler, PA 19002
215-628-1000
Fax: 215-628-1200
William E. Fristad
Parker Amchem.
32100 Stephenson Hwy.
Madison heights, MI 48071
248-588-4719
Fax: 248-583-2976
Lou Magdits
TERRAMET Mgr.
Raw Materials Manager
Buick Resource Recycling Facility
Hwy.KK
HC IBox 1395
-------
3
HN
o
u
t/2
89
I
I
P-I
i
a
IS
d
Q
W
ORDNANCE
REMOVED
FEE
UNTREATED OVERSIZE
UNTREATED CLAY BALL
TROMMEL
Key: Q Critical SITE samples
O Other samples (Non-critical
SITE and COGNIS)
"~" Primary Flows
WET
CLASSIFIERS
^
RECIRCULATED
^ WATER ^
^ w
.^
SANDS
|~~ LEACHING
(TROUGH)
" 1
/^v
(CLARIFIERj/l
.d
DENSI1
SEPARA1
(2 MINERAL.
|
FINES ^
• • ICIAF
1 SANDS V
'Y i^"^
"'OR I
ncs) ^
1 FI
JL I TA
FRESH (j£)
LEACHANT ^^ 1
^
"v/7 ^ irn>
1 A " LEAC
FROM (4CLAB
LEACHANT SURGE
NEUTRAL- r-N
IZATION "—i/
I] FLOCCULENT @
FINES ^ IWATI
nF.WATF.pmr: "^P^ r-
FLOCCULENT
FRESH
TANK /NH LEACHANT
O
FLOCCULENT
Q
1
(CENTRIFUGE
AND WATER TANK)
NEUTRAL-
V-1 IZATION
(Rl) (R2)
^ I T
LEAD
RECOVERY
(6 CELLS)
TO
LEACHING
UNITS
TREATED WHOLE SOIL
LEAD
CONCENTRATE
Figure 1 -1. Schematic of the COGNIS process.
-------
Boss, MO 65440
573-626-3476
Fax: 573-626-3405
1 magdits@misn.com
The SITE Program
Michael D. Royer
U- S. Environmental Protection Agency
Building #10 (MS-104)
2890 Woodbridge Ave.
Edison, NJ 08837-3769
732-321-6633
Fax: 732-321-6640
Annette Gatchett
LRPCD.NRMRL
U.S. Environmental Protection Agency
26 W. Martin Luther King Blvd.
Cincinnati, OH 45268
513-569-7697
Fax:513-569-7620
Information on the SITE Program is available through the
following on-line information clearinghouses:
The Alternative Treatment Technology Information
Center (ATTIC) System is a comprehensive, automated
information retrieval system that integrates data on
hazardous waste treatment technologies into a centralized,
searchable source. This database provides information
summaries on innovative treatment technologies. The on-
line access number is 513-569-7610. Technical support
number is 513-569-7272.
The Vendor Information System for Innovative Treatment
Technologies (VISITT) (Hotline: 800-245-4505) database
contains information on technologies offered by
developers.
The OSWER CLU-IN electronic bulletin board contains
information on the status of SITE demonstrations.
Homepage: http://www.clu-in.com.
Technical reports may be obtained by contacting the
Office of Research and Development Publications at 26
W. Martin Luther King Drive, Cincinnati, Ohio 45268, or
by calling 800-490-9198.
10
-------
Section 2
Environmental Requirements Analysis
The Comprehensive Environmental Response,
Compensation, and Liability Act of 1980 (CERCLA), as
amended by SARA, directs that for wastes left on site,
remedial actions must comply with federal laws and
regulations and more stringent state requirements that
apply or are relevant and appropriate under circumstances
of the release or potential release. Potentially applicable
or relevant and appropriate requirements (ARAR) are
identified throughout the remedial investigation and
feasibility study process and are selected in the record of
decision. Several technical factors must be considered in
determining whether the COGNIS process can be applied
to any particular site problem. Sections 2.1 through 2.3
provide analyses of the potential ARARs used in
evaluating the performance of the COGNIS process.
Sections 2.4 through 2.9 discuss technical factors that
should be considered in determining the applicability of
the COGNIS process.
2.1 ARARs Defined
ARARs are the federal and state environmental
requirements with which a remedial action at a CERCLA
site must comply. The more stringent of the state or
federal requirements will prevail, but only those state
requirements that are legally enforceable and consistently
enforced statewide may be established as ARARs (EPA
1988). Other nonpromulgated federal or state criteria,
advisories, and guidance, are "to be considered" (TBC).
TBCs do not have the same legal impact as ARARs, but in
some instances, TBCs may be useful in developing
CERCLA remedies if no ARARs exist for a particular
hazardous substance or for a particular situation.
An ARAR may be either applicable or relevant and
appropriate, but not both. According to the National Oil
and Hazardous Substances Pollution Contingency Plan
(NCP) (40 Code of Federal Regulations [CFR] Part 300),
the terms applicable and relevant and appropriate
requirements are defined as follows:
Applicable requirements are those cleanup standards,
standards of control, or other substantive environmental
protection requirements, criteria, or limitations promulgated
under federal or state environmental or facility siting laws
that specifically address a hazardous substance, pollutant,
contaminant, remedial action, location, or other
circumstance found at a CERCLA site. Only those state
standards that are identified by a state in a timely manner
and that are more stringent than federal requirements may
be applicable.
Relevant and appropriate requirements are those
cleanup standards, standards of control, and other
substantive environmental protection requirements, criteria,
or limitations promulgated under federal or state
environmental or facility siting laws that specifically
address a hazardous substance, pollutant, contaminant,
remedial action, location, or other circumstance at a
CERCLA site, address problems or situations sufficiently
similar to those encountered at the CERCLA site that their
use is well suited to the particular site. Only those state
standards that are identified in a timely manner and that are
more stringent than federal requirements may be relevant
and appropriate.
The NCP identifies the following three categories of
ARARs in 40 CFR §300.400(g):
Chemical-specific ARARs are usually health- or risk-
based numerical values or methodologies used to
determine acceptable concentrations of chemicals that
may be found in or discharged to the environment (for
example, maximum contaminant levels [MCL] that
establish safe levels in drinking water). The remediation
goals and targets listed in Table 1-1 in Section 1.3 are
chemical-specific ARARs.
11
-------
Location-specific ARARs restrict actions or contaminant
concentrations in certain environmentally sensitive areas.
Areas regulated under various federal laws include flood
plains, wetlands, and locations where endangered species
or historically significant cultural resources are present.
Action-specific ARARs are usually technology- or
activity-based requirements, limitations on actions, or
conditions involving special circumstances.
The major federal environmental statutes that typically
provide ARARs are RCRA, the Clean Air Act (CAA), the
Safe Drinking Water Act (SDWA), and the Clean Water
Act (CWA). Most states have passed laws similar to
federal environmental statutes and may have authorization
to implement federal laws. In addition, both federal and
state agencies have TBCs.
2.2 Other Non-ARAR Requirements
A number of requirements are not considered to be
ARARs because both the administrative and substantive
requirements are applicable to the remediation, and
therefore may not be waived. These additional
requirements include the Occupational Safety and Health
Administration (OSHA) worker protection requirements;
U.S. Department of Transportation (DOT) requirements
for transportation of hazardous materials; RCRA
requirements for off-site transportation of hazardous
waste (including compliance with the manifest
requirements); and the CERCLA Off-Site Rule.
CERCLA remedial actions and RCRA corrective actions
and closures must be performed in accordance with OSHA
requirements detailed in29 CFRParts 1900 through 1926.
Part 1910.120 is of particular importance because it
provides for the health and safety of workers at hazardous
waste sites. On-site construction activities at Superfund or
RCRA corrective action sites must be performed in
accordance with Part 1926 of OSHA, which provides
safety and health regulations for construction sites. State
OSHA requirements, which may be significantly stricter
than federal standards, must also be met.
The DOT has promulgated packaging and labeling
requirements for shipping hazardous materials. All off-
site shipments of hazardous materials must comply with
DOT requirements. In a related matter, all RCRA
manifesting requirements must be followed for off-site
shipments of hazardous waste.
The CERCLA Off-Site Rule regulates off-site disposal of
CERCLA wastes. CERCLA wastes (hazardous and
nonhazardous) can only be treated and disposed of at a
facility that is in compliance with the Off-Site Rule. Each
EPA regional office maintains a list of acceptable
facilities.
2.3 Major ARARs and TBCs for the
COGNIS Project
This section discusses the specific environmental
regulations pertinent to the operation of the COGNIS
process including the transport, treatment, storage, and
disposal of wastes and treatment residuals and analyzes
these regulations in view of the demonstration results. The
SITE demonstration of the COGNIS process was
conducted as part of RCRA closure of Site F. RCRA
closures must comply with applicable RCRA requirements.
This section also discusses potential ARARs when the
COGNIS process is used at a CERCLA site. Remedial
managers must also address state and local regulatory
requirements, which may be more stringent. The ARARs
and TBCs were identified using the operable unit
feasibility study that was conducted for TCAAP and are
based on current regulatory and policy developments.
This section presents ARARs and TBCs for soil and
surface water. ARARs and TBCs are presented for these
media because (1) contaminated soil is the primary
medium treated by the COGNIS process and (2) the wash
water used in the treatment system must be handled to
ensure that surface water is protected. Section 2.3.1
presents the major chemical-specific ARARs and TBCs,
Section 2.3.2 discusses the major location-specific
ARARs and TBCs, Section 2.3.3 presents the major
action-specific ARARs, and Section 2.3.4 discusses other
requirements that may apply to the COGNIS process.
Table 2-1 lists federal and state ARARs and TBCs and
discusses whether they are applicable to the COGNIS
process.
2.3.1 Chemical-Specific ARARs and
TBCs for the COGNIS Process
This section describes the chemical-specific ARARs and
TBCs for the COGNIS process. MPCA has a lead
abatement standard for residential property of 100 parts
per million (ppm) established in Part 4761.0300,
Subparagraph 4 of the Minnesota Rules. "Bare soil" in the
12
-------
Table 2-1. ARARs and TBCs for the COGNIS Process
Medium or
Action
Citation
Description
ARAR/TBC Basis
Soil
Water
Waste
characterization
Soil excavation
Minnesota R.
4761.0300, sub p. 4
SDWA40CFRPart
141 or state
equivalent
(Minnesota R.
4717.7150to
4717.7800)
RCRA40CFRPart
261 or state
equivalent
(Minnesota R. 7045)
CAA40CFRPart
50.6, and 40 CFR
Part52SubpartKor
state equivalent
RCRA40CFRPart
262 or state
equivalent
(Minnesota R. 7045)
Current Minnesota levels are 100
ppm for lead under its lead
abatement program
For lead, the MCL value of 15
micrograms per liter (^g/L) is the
maximum permissible level of a
contaminant in water that is
delivered to a free-flowing outlet to
the ultimate user of a public water
system. Minnesota has
promulgated Health Risk Limits
(HRL) for groundwater cleanup.
Identifying and characterizing
waste prior to and after treatment
Management of toxic pollutants
and paniculate matter in the air
Standards that apply to generators
of hazardous waste
State levels may be
relevant and appropriate
These requirements are
not ARARs or TBCs
because the COGNIS
process treats soil, not
groundwater.
Soil and waste to be
treated may be RCRA
hazardous wastes.
Fugitive air emissions may
occur during excavation
and material handling and
transport.
Soil excavation may
trigger RCRA generator
requirements.
State soil lead level at the time the closure plan
for Site F was approved was 300 mg/kg
Soil cleanup levels should consider the
possibility of leaching contaminants to
groundwater. Soil cleanup levels should be
established at a level that ensures leaching of
contaminants to groundwater does not exceed
SDWAMCLs. At SiteF.it was determined
that groundwater would not be affected by
leaching of lead.
Chemical and physical analyses must be
performed.
If necessary, the waste material should be
watered down or covered to eliminate or
minimize dust generation.
If possible, soils should be fed directly into the
wash unit for treatment
-------
I
Table 2-1. ARARs and TBCs for the COGNIS Process (corttlnoed)
Medium or
Action
Soil Excavation
(cent.)
Storage prior to
processing
Citation
CWA40CFRPart
122.26 or state
equivalent
(Minnesota R. 7045
and 7050)
RCRA40CFRPart
264
Description
Storm water runoff management
from landfills, construction sites,
and industrial sites consisting of
greater than 5 acres.
Standards applicable to the storage
of hazardous waste
ARAR/TBC Basis
Soil excavation and
treatment may involve
more than 5 acres.
Excavation may generate a
hazardous waste that must
Response
Silt fence would be constructed to control runoff
from the excavation area.
If in a waste pile, the material should be placed
on and covered with plastic that is tied down to
or state equivalent
(Minnesota R. 7045)
Waste processing
Waste
characterization
(treated waste)
RCRA 40 CFR Parts
264 and 265 or state
equivalent
(Minnesota R. 7045)
equivalent
RCRA 40 CFR Part
261 or state
equivalent
(Minnesota R. 7045)
Standards applicable to the
treatment of hazardous waste at
permitted and interim status
facilities
Standards that apply to waste
characterization
be stored in a waste pile,
containment building, or
corrective action
management unit
(CAMU). Waste piles are
land disposal units which
require compliance with
land disposal restrictions
(LDR).
Treatment of hazardous
waste may occur.
minimize fugitive air emissions and
volatilization. The time between excavation and
treatment should be kept to a minimum.
It must be determined if
treated soil is a RCRA
hazardous waste.
Equipment must be operated and maintained
daily. Air emissions must be characterized by
continuous emissions monitoring.
Chemical and physical tests must be performed
on treated soils prior to disposal.
-------
Table 2-1. ARARs and TBCs for the COGNIS Process (continued)
Medium or Action
Citation
Description
ARAR/TBC Basis
Response
On-site/off-site
disposal
Wastewater discharge
RCRA40CFRPart260
CAMU and temporary
unit (TU) rule or state
equivalent (Minnesota R.
7045)
RCRA40CFRPart268
or state equivalent
(Minnesota R. 704S)
CWA Act 40 CFR Parts
301,304,306,307,308,
402, and 403 or state
equivalent (Minnesota R.
7045)
Standards that apply to management of
remediation wastes
Standards that restrict the placement of
certain wastes in or on the ground
Standards that apply to discharge of
wastewater to a POTW or to surface water
body.
Untreated and treated soil can
be considered to be remediation
wastes.
Treated soil may be subject to
theLDRs.
Point source discharges of
wastewater are regulated
through the National Pollutant
Discharge Elimination System
(NPDES) program and
discharges to POTWs are
regulated through the
pretreatment program.
The CAMU designation allows for on-site disposal
without triggering LDR or minimum technology
requirements. TUs can be used to avoid stringent RCRA
treatment and storage technical requirements.
Wastes subject to LDRs must be treated to a specified
level or with a specific technology before being disposed
of on land.
Obtain discharge limitations for point source discharge to
surface water body and pretreatment standards for
discharge to a POTW. Discharge to a POTW requires a
permit while discharge to surface water may or may not
require a permit depending upon the proximity of the
surface water body to the CERCLA site.
-------
Minnesota rules refers to soil within the top 2 centimeters
of the ground surface. When the Site F closure plan was
approved, the residential bare soil standard was 300 mg/kg
of lead. MPCA therefore set the lead remediation goal at
that time at 300 mg/kg. Regardless of the 100 mg/kg
residential bare soil standard that now exists, Site F was
backfilled with treated soils meeting the remediation goals
and vegetated with 6 inches of top soil. The remediation
goals for lead and other metals are listed in Table 1-1 in
Section 1.3.
During the 11 months of site remediation, Wenck
Associates, Inc. (Wenck), subcontractor for TCAAP,
collected one sample from each batch of soil. Each batch
typically consisted of 30 tons of soil. Beginning in 1994,
soils from all batches that failed remediation goals were
tested with TCLP. If the leachate from TCLP analyses did
not meet maximum allowable concentrations, then that
batch of soil was reprocessed. Overall, 134 out of the 406
batches that were tested with TCLP did not meet
maximum allowable concentrations for lead, and required
reprocessing (Wenck 1995a). Atthe end of the project, all
treated soil met remedial goals.
EPA recommends site-specific health risk models that use
lead levels in soil to determine lead levels in blood and has
established a soil-lead screening level of 400 ppm to
indicate if a potential problem exists. If lead is present in
site soils at levels of 400 ppm or higher, then EPA
recommends collecting additional information in order to
run the Uptake Biokinetic Model to establish soil cleanup
levels. For current and probable future industrial land use
conditions, a remediation goal of 1,200 ppm for soil-lead
in an industrial setting would be considered to be
appropriate in accordance with EPA's new soil-lead
guidance (EPA 1994c).
In addition, the cleanup level for lead in soil should be
established to ensure groundwater does not contain lead in
excess of federal or state standards hi drinking water
through the leaching of residual lead in the remediated
soils to the groundwater. The federal action level of 15 u.g/
L of lead is the relevant and appropriate requirement for
lead in drinking water. State groundwater standards, such
as Minnesota HRLs, may be relevant and appropriate for
establishing soil cleanup levels. The state of Minnesota, as
may other states, has groundwater levels for some
contaminants, including HRLs for nonlead metals at Site
F.
2.3.2 Location-Specific ARARs and
TBCs for the COGNIS Process
No location-specific ARARs were identified for the
COGNIS process demonstration. The technology was not
demonstrated in a flood plain or near a wetland and the
area contained no identified historical or cultural
resources. Disposal of treated material may be subject to
state waste disposal facility siting requirements if the
treated material is categorized as solid or hazardous waste.
At TCAAP Site F, RCRA closure regulations were
followed because Site F was a permitted facility under
RCRA.
2.3.3 Action-Specific ARARs and TBCs
for the COGNIS Process
The principal action-specific requirements for CERCLA
sites are based on the regulatory definitions and
classifications of the materials at the site. This section
describes the waste classifications and indicates the
action-specific requirements associated with each type of
material that may be involved in implementing the
COGNIS process at a CERCLA site. These ARARs are
triggered by the particular remedial activities selected to
accomplish a remedy.
The presence of RCRA-defined hazardous waste
determines whether RCRA regulations apply to the
COGNIS process. If soils are determined to be hazardous
according to RCRA, all RCRA requirements regarding the
management and disposal of hazardous wastes will need to
be addressed. However, RCRA itself provides waivers to
some requirements, such as LDR treatment levels. RCRA
regulations define hazardous wastes and regulate their
transport, treatment, storage, and disposal. Wastes
defined as hazardous under RCRA include characteristic
and listed wastes. Criteria for identifying characteristic
hazardous wastes are included in 40 CFR Part 261 Subpart
C. Listed wastes from nonspecific and specific industrial
sources, off-specification products, spill cleanups, and
other industrial sources are itemized in 40 CFR Part 261
Subpart D.
After contaminated soils are treated by the COGNIS
process, the treated soils may still contain hazardous
constituents at levels above required cleanup action levels
or may still be considered a RCRA hazardous waste. Such
soils need to be properly managed. Proper management of
16
-------
soils that do not consist of hazardous waste, but whose
constituent levels exceed cleanup levels, may require
disposal and long-term management, including property
deed restrictions or notices. The type of disposal should be
tailored to the site-specific situation by considering a
variety of information including the types and levels of
constituents, future land use, and state requirements.
Treated soils still considered to be a hazardous waste are
subject to LDRs under RCRA unless a CAMU is
designated, or the soils may require further treatment,
recycling, or disposal. Applicable RCRA requirements
could include the following: a uniform hazardous waste
manifest if the treated soils are transported; restrictions on
placing the treated soils in land disposal units (if a CAMU
is not designated); time limits on accumulating treated
soils; permits for storing treated soils; and storage
requirements, such as storm water runoff prevention.
At most sites, ARARs relevant to soil excavation are
applicable because soils may need to be excavated and
stockpiled in the soil staging area. In addition, plastic
liners may be required under the stockpiled soil. At
TCAAP, runon and runoff from the soil staging area was
controlled by a bermed concrete pad. ARARs relevant to
underground injection wells did not apply at TCAAP
because all wastewater was discharged to a POTW. After
treatment, the soil was again stockpiled in the treated soil
holding area, which also consisted of a bermed concrete
pad. Some soils stockpiled in the holding area required
reprocessing due to exceedence of lead concentrations in
the TCLP leachate. However, following reprocessing,
treated soil did not meet either the RCRA or state
definitions of hazardous waste. Therefore, ARARs
applicable to the disposal of hazardous wastes are not
applicable to this SITE demonstration.
RCRA requirements may also apply to any concentrated
waste streams that result from the COGNIS process. In the
case of the COGNIS process demonstration at TCAAP,
the concentrated waste stream was sent under a hazardous
waste manifest to a lead smelting facility for recovery.
The COGNIS process uses water in applying its soil
washing technology. During the treatment process, water
is recycled through the system. When the treatment is
complete, some contaminated water remains and is
discharged to a POTW. In order to be discharged to a
POTW, the contaminated water must meet the
pretreatment standards established by the POTW. These
pretreatment standards are specific to the particular
POTW. Rainwater falling in the uncovered treatment and
storage area is pumped to a holding tank. Water from this
holding tank is occasionally used as make-up water.
During the demonstration, about 15,000 liters of rainwater
was used as make-up water. Wastewater was recycled
through the COGNIS and BESCORP processes. At the
season's end, wastewater including rainwater generally
did not meet the pretreatment standards and required
treatment before disposal. However, wastewater
generated during the season when the demonstration was
conducted met the pretreatment standards for discharge to
a POTW.
The COGNIS process is an ex situ technology that requires
excavation of contaminated soil. When contaminated soil
is excavated, the potential exists for the soil to contribute
to the transport of contaminants through surface water
runon and runoff during rainstorms. Storm water runon
and runoff must therefore be controlled at the site
excavation.
The CAA provides requirements for the control of fugitive
emissions. In accordance with the CAA, fugitive dust
emissions from the excavation and storage of contaminated
soil to be used in the COGNIS process must be controlled.
During the closure at Site F, dust emissions were
controlled by wetting down the soils with water. Sprinkler
systems were installed to suppress dust. Soil piles were
also covered with tarps while awaiting transport and
during transport between Sites F and D. Safety zones were
established and enforced and air monitoring was
performed.
2.3.4 Other Requirements of the
COGNIS Process
All technicians operating the COGNIS process are
required to have completed an OSHA training course and
must be familiar with all OSHA requirements relevant to
hazardous waste sites. For most sites, the minimum
personal protective equipment (PPE) for technicians will
include gloves, hardhats, steel-toed boots, and coveralls.
Depending on the types and concentrations of the
contaminants, additional PPE may be required. Noise
levels should be monitored to ensure that workers are not
exposed to noise levels above a time-weighted average of
85 decibels over an 8-hour day. If operation of the
COGNIS process causes noise levels to increase above
this limit, then workers will jbe required to wear hearing
protection.
17
-------
Shipping material off site for refining (concentrated waste
stream) or disposal (treatment residues) must be
conducted in accordance with applicable DOT regulations.
DOT regulations for packaging and labeling of hazardous
materials may apply.
Off-site disposal of wastes generated from the treatment
process must comply with the CERCLA Off-Site Rule.
The RCRA Off-Site Coordinator (ROC) in EPA regions
where wastes may be disposed of must be consulted to
ensure that any facility designated to receive waste from
the treatment process is acceptable under the CERCLA
Off-Site Rule. The CERCLA ROC was not consulted in
off-site disposal actions related to TCAAP Site F because
the action was a RCRA closure, not a CERCLA response.
Table 2-2 summarizes other non-ARAR requirements that
apply to the COGNIS process.
2.4 ImplementabiIJty of the
Technology
Implementing remedial activities at hazardous waste sites
requires numerous activities to be performed prior to,
during, and after waste treatment. These activities include
preparing the site, arranging for utilities, mobilizing,
operating treatment equipment, disposing of residues, and
demobilizing. Information regarding all these activities
was collected to determine implementability.
Wenck prepared the site, which included grading the site,
constructing a treatment pad, and installing cells for
holding treated soil. TCAAP's contractors also arranged
for utilities to be connected and residual waste disposal.
COGNIS and BESCORP were responsible for mobilization,
demobilization, and operation of their respective
equipment. According to Wenck, mobilization and
demobilization activities were carried out on schedule
(Wenck 1995b). During the SITE demonstration, all
equipment operated with only minor mechanical
problems.
2.5 Applicable Wastes
COGNIS claims that its process can treat soil, sediment,
and sludge contaminated by lead and other heavy metals or
metal mixtures by using a physical separation process and
a chemical leaching and metal recovery process.
Appropriate sites include contaminated ammunition
testing areas, firing ranges, battery recycling centers, scrap
yards, metal plating shops, and chemical manufacturers.
The system can treat metallic lead as well as lead salts and
oxides. Certain lead compounds, such as lead sulfide, are
not amenable to treatment because of their exceedingly
low solubilities. The process can be modified to leach and
recover other metals, such as cadmium, zinc, copper, and
mercury, from soils. However, treatment through the
1995 season of the Site F RCRA closure had difficulty
meeting remediation goals (two standard deviations above
calculated background levels) for copper, mercury, or
antimony. Three hundred ninety of the 790 batches, about
49 percent, failed remediation goals set for one of the .
metals. The average concentrations for these metals in
treated soils were copper (116 mg/kg), mercury (0.29 mg/
kg), and antimony (1.72 mg/kg) (Wenck 1995a).
2.6 Key Features of the COGNIS
Process
The COGNIS process has several unique features, which
include the following:
The leachant is recycled and reused within the treatment
system. Because no process wastewater is generated
during processing, water is disposed of only upon
completion of treatment or at the season's end. The
process is a net consumer of water, but at TCAAP the
water exited the system as increased soil moisture, not as
process water requiring treatment.
The lead-containing leachant passes through lead recovery
cells on which elemental lead is deposited and later is
recovered as lead concentrate.
2.7 Availability and Transportability of
Equipment
The COGNIS process equipment is mounted on several
flat-bed trailers and transported to the site. Once on site, it
requires 8 days for a four-person operating staff to unload
and assemble the system and to hook up water and
electricity. Demobilization activities include
decontaminating on-site equipment, disconnecting utilities,
disassembling equipment, and transporting equipment off
site. Currently, COGNIS has one commercial unit. The
proprietary leaching solution used in the process is
available through COGNIS.
18
-------
Table 2-2. Other Requirements for the COGNIS Process
Title
Requirement
Rationale for
Implementation
OSHA Worker Protection
Requirements
29 CFR Parts 1904 and
1910
DOT requirements for
packaging, labeling, and
transporting hazardous
materials
CERCLA Off-Site Rule, 40
CFR Part 300
These regulations establish
requirements to protect
workers who could be exposed
to noise, hazardous wastes, or
other contaminants or hazards
at the remediation site.
These regulations establish
requirements for the
packaging, labeling, and
transporting of hazardous
materials off site.
Requires consultation with
region-specific ROC to
determine if proposed off-site
facilities are in compliance with
the CERCLA Off-Site Rule.
Compliance with 29 CFR
Part 1910.120 is required
for all sites undergoing
remediation by 40 CFR
Part 300.150.
Compliance with DOT
regulations apply to off-site
shipments of untreated
waste, treatment residuals,
and concentrated waste
streams considered to be
hazardous materials and
shipped off site.
Any off-site shipments of
waste from a CERCLA site
must comply with the
CERCLA Off-Site Rule.
19
-------
2.8 Site Support Requirements
2.9 Limitations of the Technology
Technology support requirements for providing suitable
site conditions, utilities, facilities, and equipment for the
COGNIS and BESCORP processes are discussed below.
Surface requirements for the process units include a level
graded area capable of supporting the equipment. The
foundation must support the weight of the process unit.
The process unit requires a concrete pad area measuring
approximately 8,000 square feet. The process unit
requires stable access roads that can accommodate
oversized and heavy equipment.
At some sites, pretreatment of soils, in addition to soil
washing, is required. Sites contaminated primarily with
lead typically contain ordnance or battery casings. The
presence of ordnance or battery casings requires sorting of
these materials before soil is transferred to the feed hopper
for washing.
Both the COGNIS and BESCORP processes require water
and electricity. Water is also needed to decontaminate the
equipment. Water consumption during processing
depends, among other variables, on the incoming soil
moisture content because treated soil will exit the plant
saturated with water. During the demonstration,
approximately 190 liters of water was used per 1 ton of
treated soil. A 275-ampere, 440-volt, three-phase
electrical circuit is also required.
Support facilities include a staging area for contaminated
soil and material handling, a treated soil holding area, a
rainwater holding tank, a leachant storage tank, and a drum
storage area. Treated soil is stored in roll-off containers or
a soil pile. Fifty-five-gallon drums containing lead
concentrate are stored in the drum storage area. In
addition, a tank storage area to store process water may be
required at some sites. All support facilities must be
designed to control surface runon and runoff. Support
equipment includes excavation and transport equipment
such as backhoes, front-end loaders, dump-trucks,
forklifts, roll-off boxes, and storage tanks.
Decontamination, changing, and other support facilities
should be made available to personnel. An on-site office
for reviewing operations information is also useful.
The COGNIS process would not typically operate as a
standalone technology. The process will likely need
integration with a soil washing process that removes
debris and separates sands and fines. Soil washing will
remove some lead particulates from the contaminated soil,
thereby playing a role in contaminant reduction. During
this demonstration, ordnance and lead shards within the
soil posed major difficulties, from both safety and
analytical precision perspectives. The presence of
unexploded ordnance within the soil required that
ordnance experts sort and remove ordnance from the soil
before treatment. The presence of lead shards within the
soil samples collected for analysis can result in "hot
spots," affecting the precision of the analysis. It was
decided that crushing the coarse grained samples such as
the feed soil, treated and untreated sands, and density
separated lead concentrates prior to analysis would create
a more homogenous mixture. During Site F remediation,
oversized material was not crushed prior to TCLP
analysis.
The COGNIS process is only effective for treating metal-
contaminated media and is not amenable to treating
organics. The contaminated media must be excavated and
staged before treatment. Another limitation posed by the
process is that residuals generated by soil treatment,
including used leachant and reclaimed lead, require
additional treatment and disposal. The following residuals
may be generated and may require additional treatment:
Reclaimed lead, which must be recovered at a smelter or
stabilized and disposed of in accordance with RCRA
LDRs
Process water remaining after treatment and rain water
collected from the storage and treatment area may need
treatment to meet POTW pretreatment standards before
discharge to a POTW
Oversized material from soil feedstock that may contain
ordnance and other hazardous substances, depending on
the site conditions, which must be disposed of
The clay content of the feedstock soil may limit the
operation of the COGNIS process. During the SITE
demonstration, clay balls formed during the BESCORP
soil washing and were removed from the system as
oversized material. Deagglomeration of clay balls that
20
-------
may be formed is necessary to ensure contaminated fines
are leached. In addition, as the clay content in the
feedstock soil to the COGNIS process increases, the
treatment rate will tend to decline due to additional time or
equipment required for fines-water separation in the
leaching system.
The elevated salt content of the treated soil due to the
addition of leachant may limit the options for placement of
the soil. Leachant and salt concentrations can be reduced
through rinsing or natural dispersion. Soil fertility will
likely be adversely affected for several months due to
elevated salt content and degradation of organic matter,
especially if natural dispersion is used to address elevated
salt concentrations. However, this temporary loss of
fertility should be viewed in light of other lead treatment
technologies, such as solidification and capping, which
permanently reduce soil fertility.
21
-------
Section 3
Economic Analysis
This economic analysis presents cost estimates for
operating the COGNIS process. Costs have been placed
into 13 categories, 12 of which are applicable to operating
the COGNIS process at RCRA and Superfund sites (Evans
1990). A thirteenth category was added to present the cost
of soil pretreatment before delivery as partially cleaned
sands and fines for further treatment by the COGNIS
process. Costs are presented in January, 1995 dollars.
Cost data were compiled from several sources including
the SITE technology demonstration at the TCAAP site
where the soil was contaminated with lead from ordnance
testing, burning, and destruction activities. Information
collected during the SITE demonstration was recorded in
logbooks and used in this analysis when applicable.
Wenck, the TCAAP oversight contractor, provided cost
and operating data specific to COGNIS operations at
TCAAP (Wenck 1995c). Cost data were also obtained
from various independent supply vendors to present actual
costs in preference to cost estimates for such items as
fencing and mobile trailer rentals. Other cost estimates
presented in this analysis are drawn from the R.S. Means
Company, Inc. (Means); construction cost data sources;
and engineering estimates (Means 1994). Cost data
obtained from COGNIS are used in this analysis when
independent costs could not be acquired and when
COGNIS was the only reliable source for costs of a
technology-specific nature.
This economic analysis discusses the issues addressed and
assumptions used in defining abase case scenario, each of
the 13 cost categories, and conclusions of the economic
analysis.
3.1 Issues and Assumptions
This section summarizes the major issues addressed and
assumptions used in this economic analysis of the
COGNIS process. Operating issues and assumptions
about the COGNIS process are based on information that
COGNIS provided. Issues are presented in text and
primary assumptions for each are presented as bullets at
the end of each subsection. Certain assumptions were
made to account for variable site and soil contamination
parameters, others were applied to simplify cost
estimating where complex engineering or financial
functions would be required. A hypothetical base-case
scenario, provided at the end of this section, was
developed from the issues and assumptions outlined
below.
3.1.1 Equipment and Operating
Parameters
This analysis provides cost estimates for treating TCAAP
Site F soils, which were primarily contaminated with lead.
As a result, the COGNIS system was operated to optimize
treatment of TCAAP Site F soil by cost effectively
achieving treatment goals established for Site F.
COGNIS will provide and operate the system for its clients
as part of an overall on-site soil treatment service. The cost
of using the equipment is not based on time but rather is a
lump-sum price of providing a complete soil treatment
service. That price is determined by the volume of the soil
remediation job. COGNIS will calculate the price on a
site-by-site basis.
A soil washing subcontractor will be needed to pretreat the
soil before the soil enters the system. COGNIS procures
the soil washing subcontractor and includes this cost in the
total price of providing a complete soil treatment service.
For this analysis, costs for a soil washing subcontractor are
presented as a separate cost category, because this analysis
focuses on the costs specific to using the COGNIS process.
COGNIS customers will incur soil washing costs as a
22
-------
result of using the treatment system, but the cost is not
directly associated with the operation of the COGNIS
process. While numerous soil washing subcontractors
exist, BESCORP was retained for the TCAAP SITE
demonstration. For this analysis, subcontractor costs are
based on using the BESCORP soil washing system.
A depreciation value is applied to the costs presented in
this analysis because COGNIS owns the equipment. In
accordance with EPA guidance, insurance and other
company-specific overhead costs, except for labor wage
rates, are not included in this estimate (Evans 1990). No
other equipment configuration alternatives are presented
in this section because the COGNIS system at TCAAP was
the first available commercial unit. This analysis presents
fixed and variable costs for operating the full-scale
commercial COGNIS unit.
Based on observations made at TCAAP, the combined
BESCORP and COGNIS system can continuously treat
contaminated soil at a rate of 12 tons per hour. This
analysis assumes the system operates at an average rate of
10 hours per day, 6 days per week at 90 percent efficiency.
This average includes unscheduled and weather-related
downtime. This results in a real treatment rate of about 11
tons per hour or 108 tons of soil per day or 648 tons per
week. At this rate, treating 10,000 tons of contaminated
soil through the combined BESCORP and COGNIS
system would require over 15 weeks to complete. The
estimated 15 weeks does not include time required for
mobilization and demobilization. Further, this system can
only operate during nonfreezing weather conditions. This
will affect the treatment season depending on a site's
geographical location (see the following section) unless
the entire system is operated in an enclosed, heated
building.
Equipment and operating parameter assumptions and rates
include the following:
• The COGNIS personnel and soil washing subcon-
tractor personnel are licensed to operate all heavy
equipment and are trained in proper health and
safety procedures
• The treatment system operates 10 hours per day, 6
days per week
The operating efficiency is 90 percent, including
unscheduled and weather-related downtime
• The real rate of treatment through the combined
BESCORP and COGNIS system is 648 tons per
week
• The lead removal efficiency is 75 percent, which
was observed at TCAAP for the combined
BESCORP and COGNIS system
• Treatment personnel work in Modified Level D
PPE
Rainwater is collected from the treatment pad,
stored in a tank, and used as process water
• Air emissions monitoring is not necessary
3.1.2 Site-Specific Factors
Site-specific factors affect the costs of using the COGNIS
system and can be divided into the following two
categories: contaminated soil factors and site factors.
Contaminated soil factors affecting costs include soil
mass, contaminant types and concentrations, treatment
goals, and regulatory requirements. Soil mass affects
variable costs by increasing the amount of supplies,
treatment, and labor (through an increase in the duration of
a treatment project) needed to complete a remediation
project. However, projects with larger soil masses tend to
achieve economies of scale, on a cost-per-ton basis,
because the relatively constant fixed costs are distributed
over the larger soil quantity. The mass of soil, number of
contaminants, concentration in the soil, and frequency of
sampling will affect time-dependant costs and sample
analysis costs. Regulatory requirements will determine
cleanup target levels. Treatment costs nearly double if
treated soil fails to meet treatment goals and requires a
second round of processing.
Site factors affecting costs include soil particle size
distribution, site area, accessibility, availability of utilities,
and geographical location. Soil clay content affects the
effectiveness of the soil washing system because high-clay
soils tend to form dense clumps that are not easily treated.
Soils with a large amount of readily cleanable, oversized
material will increase total throughput for the treatment
system because the soil washing subcontractor will be
treating this material. Conversely, the COGNIS process
will treat more material as the mass of finer soil particles
increases. Site preparation costs are affected by the mass
of contaminated soil, the size of the site, and the
23
-------
availability of utilities and access roads. Mobilization
costs are affected by the distance that treatment and rented
equipment must be transported from its point of origin to
the site. Cold weather climates will restrict the number of
months in a year during which treatment can be performed,
because freezing water prevents the soil washing and
treatment systems from operating.
Based on results from the SITE demonstration, the treated
soil achieved the 300 mg/kg standard set by MPCA on all
4 days during the SITE demonstration. Therefore, this
analysis assumes that treated soil meets MPCA
enforceable remediation goals and will not require
posttreatment.
Site-specific assumptions include the following:
Contaminated Soil Factors
• Treated soil is backfilled at the site
• Contaminant levels in the feed soil are 825 mg/
kg; with a removal efficiency of 75 percent for the
combined BESCORP and COGNIS treatment sys-
tem, 618 mg/kg is recovered
• Cleanup goal is 300 mg/kg
Site Factors
• Appropriate access roads exist
• Utility lines, such as electricity and telephone lines,
exist on site
• The site is located in the Midwest region of the
United States, which results in a treatment season
consisting of the 8 months from April 1 to No-
vember 30, inclusive
• Foreign objects are hand sorted from the soil by
the soil washing subcontractor prior to treatment
in the COGNIS system
• Oversized material is washed and separated by the
soil washing subcontractor and this material is
backfilled at the site
• Feed soil type is the same as TCAAP Site F soil
3.7.3 Base-Case Scenario
A hypothetical base-case scenario has been developed
using the issues and assumptions described in this section
for the purposes of formulating this economic analysis.
The costs presented in text are for a 10,000-ton treatment
job processed through the combined BESCORP and
COGNIS system. The table at the end of this section also
includes the costs for a 5,000-ton and a 50,000-ton
treatmentjob. Excluding mobilization and demobilization,
the 5,000-ton scenario will require nearly 8 weeks to
complete treatment. The 50,000-ton scenario will require
about 77 weeks to complete treatment, which will cover
nearly three 8-month treatment seasons and will require
several partial mobilization and demobilization efforts. As
a result, an explanation will be provided in the text when
certain costs associated with this scenario are higher than
those for the 10,000-ton scenario.
Additional assumptions used for this base case scenario
include the following:
• About 41.8 tons of lead concentrate will be recovered
from the entire 10,000 tons of soil
• Total costs for each category are rounded to the nearest
10 dollars
• The COGNIS system will be mobilized to the
remediation site from a location within 500 miles of the
site
• Front-end loaders and other soil-moving equipment will
be supplied and operated by the soil washing subcon-
tractor
• In addition to routine operations, total labor costs dur-
ing treatment operations include collecting samples and
providing maintenance
• Insurance and overhead costs are not included in this
cost estimate because they are used in a market envi-
ronment to competitively price the system. Cost evalu-
ation under the SITE Program typically presents the cost
of technologies, not the market dictated price, which
varies from job to job
24
-------
Table 3-1. Costs Associated with the COGNIS Process3
Cost Categories
Site preparation1"
Administrative
Treatment area preparation
Treatability study
Permitting and regulatory1"
Mobilization and startup1"
Transportation
Assembly
Shakedown
Equipment1
COGNIS treatment equipment
Rented treatment support equipment
Rented auxiliary equipment
Labor0
Supplies0
Proprietary chemicals
Disposable PPE
55-gallon drums
Diesel fuel
Utilities0
Electricity
Water
Effluent treatment and disposal0
Residual waste shipping and handling0
Lead concentrate
Debris
Disposable PPE
Analytical services6
Equipment maintenance0
Site demobilization0
Disassembly
Site restoration
Soil washing subcontractor0
Mobilization
Equipment
Labor
Supplies
Utilities
Maintenance
Site demobilization
Total costs
Total one-time costsw
Total variable costs0-1
Itemized
—
35,000
184,800
30,000
—
—
5,850
20,440
3,380
—
33,350
29,150
950
-
—
106,250
4,840
1,750
1,070
—
2,500
1,250
-
-
7,740
75,000
3,700
—
—
—
21,400
72,400
-
17,030
60,350
204,820
1,650
3,750
1,560
16,350
—
-
—
5.000
Total
$249,800
—
—
—
5,260
29,670
—
..
—
63,450
—
—
—
137,450
113,910
~
—
—
—
3,750
-
—
0
86,440
-
—
—
2,500
30,000
93,800
—
—
305,510
—
—
—
'
—
—
—
$1,121,540
$378,530
$437,500
10.000
Itemized
_
35,000
184,800
30,000
..
..
5,850
20,440
3,380
..
66,700
56,670
1,950
—
—
212,500
9,670
3,500
2,140
..
4,990
2,510
—
—
15,480
150,000
29,300
..
„
..
21,400
77,700
—
17,030
87,870
409,640
3,260
7,500
1,560
16,350
..
—
-
Total
$249,800
»
__
_,
5,260
29,670
—
—
._-
125,320
__
„
—
274,890
227,810
..
__
__
7,500
—
..
0
194,780
—
._
•
5,000
60,000
99,100
...
..
543,210
..
...
..
..
..
-.
..
$1,822,340
$383,830
$895,300
50,000
Itemized
35,000
184,800
30,000
„
12,050
20,440
10,140
333,500
281,810"
9,650
..
1,062,500
48,350
17,500
10,700
„
24,950
12,550
«
„
77,400
750,000
145,200
_„
„
-
40,600
120,100
..
31,310
313,010
2,050,860
15,280
37,500
1,560
16,350
'
-
Total
$249,800
15,780
42,630
624,960
„
„
1,376,240
1,139,000
—
—
__
37,500
—
0
972,600
._
-_
„
25,000
300,000
160,700
„
„
2,465,870
„
„
__
„
—
»
—
$7,410,080
$468,910
$4,775,300
Notes:
• Costs are in January 1995 dollars
b Fixed costs
c Variable costs
d Only a prorated cost is presented; the contractor may elect to buy the equipment for large jobs
° Fixed and variable costs combined
f Excluding soil washing subcontractor costs
25
-------
3.2 Cost Categories
Table 3-1 presents a breakdown of the costs considered
within each of the 13 cost categories. Costs for a soil
washing subcontractor are presented in Section 3.2.13, an
additional cost category for this technology. Table 3-1
presents a breakdown of costs for the base case scenario of
the 10,000-ton remedial action site; costs for 5,000- and
50,000-ton scenarios are provided for comparison
purposes.
Cost data associated with the COGNIS technology have
been assigned to the following 13 cost categories: (1) site
preparation; (2) permitting and regulatory; (3) mobilization
and startup; (4) equipment; (5) labor; (6) supplies; (7)
utilities; (8) effluent treatment and disposal; (9) residual
waste shipping and handling; (10) analytical services; (11)
equipment maintenance; (12) site demobilization; and
(13) soil washing subcontractor. Each of these cost
categories is discussed below.
3.2.1 Site Preparation Costs
Site preparation costs include administrative, treatment
area preparation, and treatability study costs. For this
analysis, administrative costs for site preparation, such as
legal searches, access rights, and other site planning
activities, are estimated to be $35,000 (Means 1994).
Site preparation costs may not be incurred at every site. At
TCAAP, an area was prepared at Site D to house an
incineration unit. This area was later made available for
treating soil from Site F. However, for this analysis, it was
assumed that a suitable treatment area needs to be
prepared. Treatment area preparation includes erecting a
security fence and constructing a bermed concrete pad on
which to site the COGNIS process, the soil washing
equipment, and soil staging and storage areas. The bermed
concrete pad consists of the following three separate areas:
an 8,000 square foot (ft2) area for treatment equipment; a
6,000 ft2 soil staging area for contaminated soil; and an
8,000 ft2 storage area for treated soil. Each area will be
connected to form one large area, sealed with an epoxy
coating, and equipped with berms and sumps to prevent
cross contamination and to define clean-area boundaries.
Each area will have specific design features applicable to
its function. Construction costs are estimated to be $8 per
ft2 Based on a quote from a fencing company, the fence
will cost about $8,800 including all materials and labor.
This analysis assumes that treatment goals will be met. As
a result, the cost of restricting access to treated soil is not
presented.
COGNIS conducts treatability studies in two phases. The
first phase tests the feasibility of the metallurgical aspects
of the process to determine if the contaminant can be
adequately leached and removed under practical
Conditions. COGNIS charges about $ 10,000 for this phase
of testing. The second phase tests the physical handling
properties, such as settling rates and dewatering
conditions, to be incurred under treatment conditions.
These tests cost about $20,000. Total treatability study
costs are estimated to be $30,000.
Total site preparation costs are estimated to be $249,800.
3.2.2 Permitting and Regulatory Costs
Remedial actions must be consistent with ARARs of
environmental laws, ordinances, regulations, and statutes,
including federal, state, and local standards and criteria. In
general, ARARs must be determined on a site-specific
basis.
Permitting and regulatory costs include fees for highway
permits for oversized vehicles traveling no more than 500
miles. These fees are assumed to total about $5,000.
Oversight costs incurred by regulatory agencies are not
included in this analysis because such costs depend on the
extent of oversight activities, which are negotiable
between the site owner and the regulatory agencies.
Furthermore, such oversights extend over several tasks
and may not be based on the volume of media requiring
remediation or the tune required for remediation.
Permits will be needed to discharge process water to a
POTW. According to Wenck, the cost for discharging the
total amount of water generated at TCAAP to the POTW
was about $8 per 4,000 liters of water (Wenck 1995b). The
combined BESCORP and COGNIS plants contain about
1.3 x 10s liters of water. The cost for discharging this water
to a POTW is $260.
Total permitting and regulatory costs are estimated to be
$5,260. Costs for the 5,000-ton scenario will be the same
as the 10,000-ton scenario. Costs for the 50,000-ton
scenario includes fees for highway permits to be paid three
times and the cost for discharging process water to the
POTW three times, for a total cost of $15,780.
26
-------
3.2.3 Mobilization and Startup Costs
Mobilization and startup costs include transporting the
COGNIS system and rental equipment to the site,
assembling the COGNIS system, and performing the
initial shakedown of the treatment system. COGNIS will
provide trained personnel to deliver, assemble, operate,
and maintain the COGNIS system and any necessary
support equipment. COGNIS personnel are assumed to be
trained in proper health and safety procedures for a
hazardous waste site, so training costs are not incurred as
a direct startup cost.
Transportation costs are site-specific and will vary
depending on the location of the site in relation to all
equipment vendors, including COGNIS. COGNIS will
mobilize its equipment to a customer's site. For this
analysis, the COGNIS equipment is assumed to be
transported 500 miles. COGNIS charges $5.50 per mile
for a total cost of $2,750. Permit costs for oversized
vehicles were presented in Section 3.2.2. Vendors of
rented treatment system support equipment (see Section
3.2.4) add delivery charges to the equipment. According
to independent vendors, delivery charges are about $ 1,000
for two mobile office trailers; $300 for an acid tank; and
$1,800 for heavy equipment that includes one excavator,
one dump truck, and two front-end loaders or backhoes.
The cost for the initial supply of proprietary acid is
covered in Section 3.2.6, Supply Costs. Total
transportation costs will be about $5,850.
Assembly costs include the costs of unloading equipment
from the trailers, assembling the COGNIS system, and
hooking up water and electricity. The four-person crew
that operates the system will work two 10-hour days
unloading and assembling the system. Total labor costs,
including overtime, will be about $11,440 (see Section
3.2.5 for labor rates). A crane must be rented for 4 days.
Vendors typically charge about $130 per hour or about
$1,000 per day. Therefore, the total cost of renting the
crane for 4 days is $4,000. After assembly, equipment
must be hooked up to water and electrical utilities, which
is estimated to cost about $5,000. Total assembly costs are
estimated to be $20,440. Some of the assembly costs such
as crane rental and connecting utilities are shared between
COGNIS and BESCORP. There are several other
activities where costs are to be shared by COGNIS and the
soil washing subcontractor. However, due to the low cost
of such activities, there is minimal impact on COGNIS' or
BESCORP's cost for each ton of soil treated. Therefore,
all such costs are assigned to the prime contractor,
COGNIS.
Initial shakedown costs include filling the system with
water to check for leaks and verifying that the electrical
and air-powered equipment is working. The four-person
crew that operates the system will work two 10-hour days
to provide initial startup activities. Total shakedown costs,
including overtime, will be about $3,380.
For the 50,000-ton scenario, rented auxiliary equipment-
will need to be delivered to the site at three different times.
In addition, equipment will need to go through initial
shakedown activities three separate times. As a result,
these partial mobilization and startup activities costs are
incurred three times. Depending on site location and soil
conditions, the contractor may elect to operate the
COGNIS system 24 hours per day on a regular operating
day, without seasonal downtime. This analysis uses a
prorated cost based on activities observed during site
remediation at TCAAP, where seasonal downtime was
encountered and the system was operated for 10 hours per
day, 6 days a week.
Total mobilization and startup costs for treating 10,000
tons of soil are estimated to be $33,670.
3.2.4 Equipment Costs
Equipment costs include the COGNIS treatment system,
rented support equipment, and rented auxiliary equipment.
COGNIS will provide the complete COGNIS treatment
system to its clients as part of the overall on-site soil
treatment service. The equipment is not leased to clients
on a time-based rate, but rather as a lump sum included in
the price of providing the treatment service. That price is
determined by the size of the soil remediation job and is
not based on time, and therefore will be calculated by
COGNIS on a site-by-site basis. COGNIS estimates the
capital equipment depreciation for this base case analysis
costs about $66,700.
Rented treatment system support equipment includes an
air compressor, centrifuge, acid tank, two 8-foot by 20-
foot mobile trailer offices, and heavy equipment. Rental
equipment vendors provided the following costs: $1,500
per month for an air compressor; $12,000 per month for a
centrifuge; and $745 per month for an acid tank. Mobile
27
-------
office trailers cost about $640 per month to rent. At a
minimum, heavy equipment will include one 1.5-cubic-
yard hydraulic backhoe excavator; one 12-cubic-yard
dump truck; and two 1-cubic-yard capacity, hydraulic
front-end loaders or backhoes to be used for soil
excavating, transporting, and regrading activities,
respectively. Based on Means estimates and equipment
vendor quotes, this heavy equipment can be rented for a
total cost of about $12,000 per month. The equipment is
necessary for the duration of the remediation and for an
additional 5 weeks to cover earth-moving activities during
mobilization and demobilization. To treat 10,000 tons of
soil, costs for rental support equipment are estimated to be
$113,350. Rented treatment system support equipment
costs are to be shared by COGNIS and the soil washing
subcontractor. Therefore, COGNIS' share of this cost is
$56,670. In the 50,000-ton case, where treatment covers
several seasons, the contractor may elect to operate the
COGNIS system in a different manner to reduce costs.
One option would be to buy the equipment. Only a
prorated cost was used in this analysis.
Rented auxiliary equipment includes a steam cleaner and a
portable toilet. The steam cleaner will be used to
decontaminate equipment and to keep the work areas
clean. A steam cleaner can be rented for $300 per month.
A portable toilet can be rented for $200 per month and will
also be needed during mobilization and demobilization
activities. Total rental auxiliary equipment costs are
estimated to be $1,950. This cost is also expected to be
shared between COGNIS and the soil washing
subcontractor. In the 50,000-ton case, where treatment
covers several seasons, purchasing the auxiliary
equipment would be more cost effective, but this analysis
does not consider this option.
Total equipment costs for treating 10,000 tons of soil are
estimated to be $125,320. Equipment costs for the 5,000-
ton and 50,000-ton scenarios are amortized costs based on
the amount of soil treated.
3.2.5 Labor Costs
Labor costs include the operating staff needed to operate
and maintain the COGNIS system, and site managerial
staff. Wage rates in this analysis include overhead and
fringe benefits. Operating staff includes one plant
manager (at $45 per hour), one foreman (at $35 per hour),
and two technicians (at $25 per hour). Managerial staff
includes one project manager (at $55 per hour), one health
and safety officer (at $45 per hour), and one administrator
(at $25 per hour). Labor costs required for site preparation,
mobilization, and demobilization are discussed under
Sections 3.2.1 and 3.2.3. Excavating and soil hauling is
provided by the soil washing subcontractor.
For this analysis, the total treatment time needed to treat
10,000 tons of soil will be over 15 weeks. To accomplish
this, the COGNIS system will be operated 10 hours per
day, 6 days per week. Wage rates increase to time-and-
one-half for every weekly hour over 40 hours. Labor costs
to treat 10,000 tons of soil are estimated to be $ 140,140 for
operating staff and $134,750 for managerial staff. No
lodging or per diem costs are included in this analysis
because it is negotiable and may not always be charged to
a customer.
Total labor costs for treating 10,000 tons of soil are
estimated to be $274,890.
3.2.6 Supply Costs
Supplies include proprietary leachants that are mixed at
the treatment site, disposable PPE, 55-gallon steel drums,
and diesel fuel. The quantities of all supplies will depend
on the amount of soil treated, the oversized material-to-
soil ratio, the lead concentration in feedstock soil, and the
duration of the treatment activities.
According to COGNIS, the types and quantities of
proprietary acid and leachant will vary in cost depending
on site-specific treatment goals. Based on the
contaminated soil and treatment goals for the TCAAP
project, COGNIS estimates the cost of these proprietary
agents will be about $21.25 per ton of soil treated. For this
analysis, total proprietary agent costs will be about
$212,500.
The four-person operating staff will operate the COGNIS
system wearing modified Level D disposable and
nondisposable PPE. Other personnel will wear PPE when
entering the treatment area. Disposable PPE includes
gloves, booties, and disposable coveralls. The operating
staff will change disposable PPE about four times per day
at a cost of about $6.50 per change or $104 per day. The
total time required to treat 10,000 tons of soil will be 93
days. Total disposable PPE costs will be $9,670.
Steel 200-liter (55-gallon) drums are needed to store
recovered lead prior to off-site smelting and to dispose of
28
-------
used PPE. The rate of lead concentrate production is
dependent on the rate of lead recovery, which can vary
within a site. Thus, the number of 55-gallon drums needed
will be highly site-specific arid will be difficult to
determine in advance. Lead concentrates consist of those
recovered through the density separation process and the
cell lead concentrates. A total of 2,695 kg of lead
concentrate was collected during the demonstration.
Based on observations made during the demonstration,
one 55-gallon drum can hold approximately 300 kg of lead
concentrate. Thus, about nine 55-gallon drums will be
filled with lead concentrate during every 6-day week. This
estimate compares well with the amount of lead
concentrate recovered during the entire remediation, when
about 15 drums of lead concentrates were recovered per 6-
day week.
About one 55-gallon steel drum per day will be required to
dispose of used PPE. The 55-gallon steel drums cost about
$15 each. To treat 10,000 tons of soil, a total of 93 55-
gallon drums will be needed to dispose of PPE and a total
of 14 drums will be needed for disposal of lead
concentrate. Total drum costs for this analysis are
estimated to be $3,500.
Diesel fuel is used to operate all earth moving equipment
and the air compressor. Fuel costs are site-specific and will
vary depending on the market price of diesel fuel and the
extent of equipment usage at the site. Diesel fuel is
estimated to be consumed at a rate of 460 liters per week.
Diesel fuel costs about $0.30 per liter for a total cost of
about $2,140. Costs for drums and diesel fuel are expected
to be shared between COGNIS and the soil washing
subcontractor.
Total supply costs are estimated to be $227,810.
3.2.7 Utility Costs
Utilities used by the COGNIS system include electricity
and water. This analysis assumes that electrical power
lines and water lines will be available at the site. Electrical
and water costs can vary extensively depending on the
geographical location and local utility rates. This analysis
assumes that the rate of electricity and water consumption
is constant and based on the amount of soil treated. The
costs presented in this section are for the COGNIS system
without soil washing. The soil washing utilities costs are
presented in Section 3.2.13.
Electricity is used to operate the COGNIS system, the soil
washing equipment, the mobile office trailers, and the
centrifuge; however, most of the electric power is
consumed by the treatment equipment. COGNIS
considers the exact electrical requirements of its
components to be confidential information. COGNIS
provided an estimate that at TCAAP, the combined power
requirement for the COGNIS system and soil washing
equipment is 121 kilowatts (kW) (i.e., 275 amps at 440
volts)
The actual electric energy usage by the COGNIS system
and the soil washing subcontractor at TCAAP between
August and September, 1994, averaged 21,280 kW-hr per
month (Federal Cartridge Company 1994). Assuming 240
hours of operation per month, the average electrical power
requirement is about 90 kW. This analysis takes the higher
of the two estimates (i.e., 121 kW from COGNIS instead of
90 kW based on the cited meter readings) and assumes the
COGNIS system uses about one-half of this electrical
power, which is 60 kW, which is drawn for 60 hours per
week for a total of 15.4 weeks. Based on an average cost
of power in the midwestern region of the United States,
Activity
Costs Per Ton Summary
5,000 Tons
10,000 Tons
50,000 Tons
Leaching activities only (includes equipment,
labor, supplies, and utilities)
Leaching and soil washing
All activities, including leaching, soil washing,
site preparation, and residual handlinga
$70
$157
$224
$70
$138
$182
$70
$124
$148
Note:
Profit and insurance are not addressed
29
-------
power 5s assumed to cost $0.09 per kWh including demand
and usage charges. Total electricity costs, excluding soil
washing activities, to treat 10,000 tons of soil are estimated
to be about $4,990.
Water is used for both the COGNIS system and for soil
washing activities. During the demonstration, the water
meter recorded a total of 53,000 liters of water during 3
days, in which about 324 tons of soil was treated. This
would result in a project water consumption rate of about
160 liters of water per ton of soil for the combined
BESCORP and COGNIS processes. This usage includes
plant process water as well as dust control at the excavation
site. Water usage can also be estimated from an increase in
moisture content in treated soil to that in the feed soil.
Average moisture content of the feed soil was 6 percent.
Average moisture content of the treated soil, combining
moisture contents of treated sands and fines in a 2 to 1 ratio,
was 29 percent. Therefore, a net increase of 23 percent in
moisture content in the treated soil was realized. The
treated soil with increased moisture content does not
include the oversized material. Therefore, assuming that
the feed soil is composed of 17 percent oversized material,
10,000 tons of soil results in 8,300 tons of treated soil with
about 1.9 million liters of water, and a water consumption
rate of 190 liters per ton. The water consumption rates
estimated by the two procedures are comparable. For this
analysis, the more conservative rate of 190 liters per ton of
soil was selected. It was also assumed that BESCORP and
COGNIS use equal amounts of water. Therefore, a water
consumption rate of 95 liters per ton of soil was assumed
for both COGNIS and BESCORP. Water costs are
assumed to be $0.01 per gallon. Total water costs,
excluding soil washing activities, to treat 10,000 tons of
soil are estimated to be $2,510.
Rainwater will be collected from the treatment area, stored
in tanks, and used for process water. At TCAAP, about
15,000 liters of rainwater was used for process water
during the demonstration. This constitutes about 40 liters
of water per ton of soil treated. This would indicate that the
combined BESCORP and COGNIS processes used 160
liters per ton (based on water meter readings) plus the 40
liters per ton of rainwater for a total of 200 liters per ton.
Because this value is comparable to the 190 liters per ton
value and the use of rainwater can vary widely, this
analysis considers 190 liters of water used per ton of soil.
Total utility costs are estimated to be $7,500.
3.2.8 Effluent Treatment and Disposal
Costs
The COGNIS system at TCAAP did not produce effluent
during the SITE demonstration requiring further treatment
or disposal. Rainwater from the treatment area was
collected in sumps and stored in a tank and used as process
water. Hence, it is not treated separately or discharged
frequently. It will be discharged only upon completion of
treatment or, in case of large soil treatment projects, during
seasonal shutdown. Discharge may also be necessary after
extraordinarily large precipitation events when the amount
of rain water collected is greater than storage capacity and
process water requirements. Process wastewater is
neutralized, treated, and discharged to a POTW at the
season's end or at the completion of treatment. At
TCAAP, this cost was included in the cost of obtaining the
discharge permit. Continuous monitoring of perimeter
dust at TCAAP did not identify potentially harmful air
emissions.
As a result, no cost will be incurred for effluent treatment
and disposal.
3.2.9 Residual Waste Shipping and
Handling Costs
The residuals produced by using the combined BESCORP
and COGNIS system to perform lead extraction include
metal concentrate, sorted debris, and drummed disposable
PPE. The total amount of residuals generated depends on
the amount of debris (such as ordnance, wood, twigs, and
metal fragments) in the soil and whether posttreatment of
soils will be needed. The actual amount of metals
concentrate generated will depend on the concentration of
the contaminant in the soil. The total costs for disposing of
metals concentrate will depend on whether a recycling
market exists, and if so, recycling may partially offset total
disposal costs.
This analysis assumes lead concentrate will be generated
from the combined soil washing and leaching system as the
only metals residual. Disposal of lead waste from TCAAP
incurred a processing fee and received a metals credit. The
value of the metals credit is based on the lead content of the
waste and the current market price for lead. At TCAAP, a
lead smelter processed the lead concentrate for metal
recovery. The smelter operates an EPA-certified
laboratory to test incoming materials for lead content.
30
-------
Disposal costs are $400 per ton for processing and a metals
credit of 80 percent of the current metals market price for
lead. Wastes delivered in drums incur an additional $15
fee per drum (Gopher Smelting 1995). Based on disposal
of 41.8 tons of lead concentrate in 140 55-gallon drums,
net lead concentrate disposal costs, inclusive of lead
content credits, are about $300 per ton plus a $15 drum
disposal fee. Total disposal costs for lead concentrate for
this analysis are estimated to be $15,480.
At many sites where lead contamination is present, debris
such as battery casings, ordnance, and other oversized
material will need to be sorted from the soil before soil
washing and lead extraction. The amount of debris
generated and disposal costs will be highly site-specific.
Based on observations made during the demonstration,
this analysis defines debris as the ordnance, oversized
wood, and metal fragments sorted from the feedstock soil
after excavation and assumes the soil washing
subcontractor will hand sort the debris prior to washing.
At TCAAP, about 50 kg of debris was generated for every
ton of soil treated. While oversized wood and metal
fragments could probably be classified as nonhazardous
materials and disposed of in a solid waste landfill,
ordnance will require special handling. For this analysis,
it was assumed that debris will require disposal in a RCRA
Subtitle D landfill. For this analysis, disposal costs will be
about $300 per ton, including transportation, which is the
average cost of disposing of RCRA hazardous waste in a
Subtitle D landfill. Therefore, about 500 tons of debris
will be disposed of for a total cost of $150,000.
Drummed disposable PPE is assumed to fill a total of 93
drums. These drums will require off-site disposal at a
licensed hazardous waste landfill. This analysis assumes
the drums will be transported off site in two separate loads.
Transportation costs are estimated to be $700 per trip, and
disposal costs will be about $300 per 55-gallon drum.
Total disposal costs for drummed PPE are estimated to be
$29,300.
Total residual waste shipping and handling costs are
estimated to be $ 194,780.
3.2.10
Analytical Services Costs
Samples are collected to determine if clean-up levels are
being achieved. If clean-up levels are not met, the results
are useful for optimizing the leaching system. Sampling
frequency and quantity are highly site specific and are
based on treatment goals, contaminant concentrations, and
regulatory agency requirements. Analyses will typically
be performed for the metals of concern at a site. This
analysis assumes that one sample will be collected for
every 50 tons of soil treated and will be analyzed for one
metal of concern, which is lead. Costs for analyzing lead
following SW 846 Methods 3020 and 6010 (EPA 199la)
are about $25 per metal per sample (General Testing
Corporation 1993). This analysis does not include costs
for TCLP analyses or an on-site laboratory because these
elements will vary from site to site.
Total analytical costs are estimated to be $5,000.
3.2.11 Equipment Maintenance
Costs
Maintenance labor has been included in labor costs (see
Section 3.2.5). COGNIS estimates that maintenance
material costs will be about $6 per ton of soil treated.
COGNIS incurs costs for design adjustments, facility
modifications, and equipment replacements, because
COGNIS is responsible for remediating faulty design
problems.
Total maintenance costs of equipment used to treat 10,000
tons of soil are estimated to be $60,000.
3.2.12
Site Demobilization Costs
Site demobilization includes treatment system disassembly
and site restoration activities. Disassembly includes
shutdown, decontamination, rental equipment return, and
utilities disconnection. According to COGNIS, the four-
person crew that operates the system will work ten 10-hour
days to decontaminate, disassemble and load the system.
Total labor costs, including overtime, will be about
$ 14,3 00 (see Section 3.2.5 for a breakdown of labor costs).
A crane must be rented for 4 days at about $ 1,000 per day
for a total cost of $4,000. All rented treatment system
support equipment will be returned for a total cost of
$3,100. Total disassembly costs are estimated to be
$21,400.
Site restoration activities include demolishing constructed
pads, placing topsoil over treated areas, and reseeding
backfilled and bare spots. Demolition of the concrete pad
will be about $3.05 per square foot for a total cost of
$67,100 (Means 1994). About 205 tons of soil will be
needed to cover an area measuring 50 feet by 50 feet with
31
-------
6 inches of soil. The cost of topsoil and grading is about
$35 per ton (Means 1994). Reseeding will cost about
§39.25 per 1,000 square feet including mulch and fertilizer
to cover an estimated 2-acre area (Means 1994). Total site
restoration costs are $77,700.
For the 50,000-ton scenario, rented support equipment
will need to be returned three separate times in order to
avoid rental costs for the nontreatment months. In
addition, multiple shutdowns will occur: two partial
shutdowns in addition to the one complete system
shutdown. Labor for the partial shutdowns is assumed to
require the four-person operating staff to work five 10-
hour days. As a result, these portions of site
demobilization costs will be higher in this case.
Total site demobilization costs are estimated to be
$99,100.
Soil Washing Subcontractor Costs
This section presents the costs of subcontracting a soil
washing service. This section is typically not presented in
the economic analysis section of an ITER because it is not
a specific cost category. It is also specific to the COGNIS
process in that a soil washing service is required prior to
treating soil in the COGNIS system. A soil washing
service is required before leaching to remove oversized
material (<6.35 mm) and to separate sands and fines from
each other. Soil washing costs are provided in this section
in order to focus the analysis of the preceding 12 sections
solely on the COGNIS system. As such, this estimate is
brief and its primary purpose is to illustrate the types of
costs that may be incurred.
The following subsections present only the cost categories
that are expressly incurred as a result of using a soil
washing service to augment the lead extraction service.
For example, no site preparation costs will be incurred
because it is assumed that the site will be prepared to
support lead extraction activities, which are accounted for
in Section 3.2.1. At some sites, the soil washing contractor
may be the primary contract and will be responsible for
activities that are common to both soil washing and
leaching. Incremental costs of adding an acid leaching
unit to soil washing are listed in Table 3-1.
Actual costs of procuring a soil washing service will vary
depending on several economic conditions, including
local competition levels, local wage rates, and
subcontractor comparative advantages. This soil washing
cost analysis uses all base-case scenario assumptions and
additional assumptions when necessary. BESCORP
served as the soil washing subcontractor for the TCAAP
project. Much of the cost data provided in this analysis is
based on using a BESCORP commercial-scale unit
because they are published cost figures (EPA 1994a).
Total costs are presented for treating 10,000 tons of soil.
This analysis assumes that soil washing will maintain the
same rate of soil throughput as the COGNIS system, which
is 108 tons of soil per day.
For this analysis, total subcontractor costs for soil washing
are estimated to be $486,540 which are detailed in the
sections below.
Soil Washing Mobilization and Startup Costs
Subcontractor mobilization and startup costs will include
transporting soil washing equipment to the site and the
labor for assembling the equipment and for conducting
initial shakedown activities.
The soil washing equipment is assumed to be transported
500 miles at $5.50 per mile for a total cost of $2,750.
Assembly and shakedown requires a 10-person crew that
operates the equipment, working seven 6-hour days for a
total cost of $14,280 (EPA 1994a). For the 50,000-ton
scenario, shakedown activities will occur two additional
times but are estimated to be completed in one-half the
amount of time previously required.
Total costs for mobilization of soil washing equipment are
estimated to be $17,030.
Soil Washing Equipment Costs
The soil washing treatment system must be modified for
site-specific conditions on a case-by-case basis. The soil
must first be characterized and other factors, including
contaminant type and level and cleanup criteria, must be
considered when designing the system. The costs of these
activities are presented in Section 3.2.1. The BESCORP
commercial-scale unit appropriate for the base-case
scenario is estimated to use a depreciation value of about
$31,200 (EPA 1994a). This equipment is assumed to
operate at a rate that meets the COGNIS system feed rate.
The soil washing subcontractor is assumed to operate
heavy equipment. The rental cost of this equipment is
$56,670 for a total equipment cost of $87,870.
32
-------
Soil Washing Labor Costs
Labor costs include the site managerial staff and the
operating staff needed to operate and maintain the soil
washing equipment and the heavy equipment. Wage rates
in this analysis include overhead and fringe benefits.
Operating staff includes one excavation site presorter with
ordnance experience (at $40 per hour), two post-washing
sorters (at $35 per hour), one post-washing sorter with
ordnance experience (at $40 per hour), three system
operators, including laboratory personnel (at $35 per
hour), and two equipment operators (at $35 per hour).
Managerial staff includes one project manager (at $55 per
hour). Manual sorting activities are assumed to be
incurred at lead-contaminated sites. The soil at lead-
contaminated sites is typically mixed with debris, such as
battery cases and bullets, that must be removed from the
soil before washing. Because the TCAAP project is
located on a military base and ordnance was being hand
sorted, two of the sorting personnel had ordnance
experience and were not BESCORP personnel. As a
result, labor costs presented in this analysis may be higher
than at other sites requiring fewer hand sorters. All staff
are assumed to work the same hours as COGNIS
personnel.
Total labor costs for soil washing, including overtime
hours, are estimated to be $409,640.
Soil Washing Supplies Costs
Supplies required by the soil washing subcontractor
include proprietary chemicals. Proprietary chemicals for
soil washing will vary among subcontractors. Proprietary
chemical usage observed during the SITE demonstration
constituted about one 55-gallon drum of chemicals during
every day. The cost of these chemicals will be about $35
per 55-gallon drum.
Total costs for soil washing supplies are estimated to be
$3,260.
Soil Washing Utilities Costs
Soil washing activities use electricity and water. This
analysis assumes the COGNIS system and the BESCORP
equipment used the same amount of electricity and water.
Based on observed electricity usage at the SITE
demonstration, the BESCORP system is assumed to draw
60 kW. Power is estimated to cost $0.09 per kWh
including demand and usage charges. Based on water
usage at the SITE demonstration, total water usage is
assumed to be 95 liters per ton of soil treated. Water is
estimated to cost $0.01 per gallon.
Total utility costs will be $4,990 for electricity and $2,510
for water.
Soil Washing Equipment Maintenance Costs
Equipment maintenance costs are estimated to be 5
percent of the capital equipment. Although this cost is
expected to vary over time, the variations are expected to
be insignificant.
For this analysis, total equipment maintenance costs are
estimated to be $ 1,560.
Soil Washing Site Demobilization Costs
Site demobilization includes equipment shutdown,
disassembly, decontamination, and return transportation
of 500 miles. This analysis assumes the 10-person
operating staff that assembled the system will perform site
demobilization, which is assumed to require five 8-hour
days to complete. All soil washing equipment is assumed
to remain on site during the seasonal downtime incurred in
the 50,000-ton scenario.
Total demobilization costs are estimated to be $16,350.
3.3 Conclusions of the Economic
Analysis
This analysis presents cost estimates for treating lead-
contaminated soil using the COGNIS lead extraction
treatment system. The base-case scenario used for this
analysis includes treating 10,000 tons of soil containing
825 mg/kg lead to a residual concentration of 206 mg/kg.
No pretreatment other than soil washing, hand sorting of
foreign objects, and removing oversized material is
required. Soil washing costs are presented individually in
order to focus on the direct costs associated with using the
COGNIS process.
For the base-case scenario, total costs including soil
washing are estimated to be $1.8 million. Excluding soil
washing activities, total costs are estimated to be $1.3
million. Total fixed costs for leaching are $383,830. Of
this amount, $249,800 or 65 percent is for site preparation
33
-------
costs. Estimated total variable costs for leaching are
$895,300. Of this amount, labor, supplies, and residual
waste disposal costs account for about 31, 25, and 22
percent, respectively.
Total costs per ton of soil treated, presented in the
following table have been divided into three categories:
(1) leaching activities only, (2) leaching and soil washing
activities only, and (3) all costs considered in this analysis.
The estimated cost per ton of soil treated is $70. Costs for
leaching and soil washing activities only is estimated to be
$138. The cost per ton of soil treated for all activities of
this analysis is estimated to be $182.
Wenck reported costs monitored during the 11 months of
remediation (Wenck 1995a). Wenck divided costs into
three primary categories: pretreatment, treatment, and
posttreatment. Pretreatment and posttreatment activities,
which include site preparation, soil excavation, oversight,
sampling and analyses, and waste handling and disposal,
were primarily conducted by Wenck. The price of soil
treated by BESCORP and COGNIS was reported at $179
per ton of soil treated (COGNIS 1996).
This analysis of a base-case scenario using the COGNIS
process shows that large jobs that require more than one
treatment season to complete will incur additional costs
due to the need to mobilize, start up, and demobilize
certain equipment more than one time. Table 3-1 shows
the estimated costs of treating 50,000 tons of soil. Based
on the assumptions used in the base-case scenario,
remediation of 50,000 tons of soil would require two entire
8-month treatment seasons plus 3 additional months.
Despite higher fixed costs, the following economies of
scale are realized: the estimated cost per ton of treated soil
is $148 per ton and $124 per ton, excluding site
preparation and residual handling.
34
-------
Section 4
Treatment Effectiveness
This section documents the methodology, discusses the
physical and chemical analytical results, and draws
conclusions for evaluating the effectiveness of the
COGNIS process in removing lead from soils. Treatment
effectiveness for the SITE demonstration did not consider
removal of other metals of concern. All sampling and
analyses were conducted in accordance with the EPA-
approved quality assurance project plan (QAPP)
developed for this project. All sample analyses met the
quality control acceptance criteria listed in the QAPP.
4.1 Methodology
Samples were collected from several locations in the
treatment process. These include the material in the feed
hopper, the washed oversized material exiting the
trommel, the washed (untreated) sands and fines entering
the COGNIS process, the treated sands and fines exiting
the COGNIS process, and the lead concentrate exiting the
density separator and the electrochemical cells. Throughout
this report, the term "untreated" sands and fines are used to
define sands and fines treated through BESCORP's
washing process and entering COGNIS's leaching
process. Feed soil samples were screened with a 6.35-
mm-diameter sieve; the portions remaining in the sieve
were designated as untreated oversized material. During
the demonstration, it was observed that the oversized
material is composed of rocks and clay balls. Some of
these clay balls did not break down in the trommel. Clay
balls were hand sorted from the treated and untreated
oversized material and samples were collected. Table 4-1
presents sampling locations and frequencies. Sample
locations are also shown in Figure 1-1.
During the demonstration, feed soil samples were
collected once during each hour from a point immediately
before soil enters the soil feed conveyor. This soil was
hand screened through a 6.35-mm-diameter screen. Once
during every 2 to 3 hours, screened samples were
composited so that three composite samples were
collected each day to be analyzed for lead. Hourly samples
were composited to create one sample per day that was
analyzed for other metals, TCLP metals, percent moisture,
particle size distribution (PSD), load bearing capacity,
cation exchange capacity (CEC), and pH. In addition, on
the third day, a sample was collected for bioassay tests to
evaluate the toxicity of treated and untreated soil. All
compositing was conducted. following the procedure
described in ASTM Method C 702-75 (ASTM 1994).
Samples of the treated and untreated sands and fines were
collected once during every hour to be analyzed for lead.
Hourly samples were composited into one daily sample to
be analyzed for other metals, TCLP metals, PSD, percent
moisture, CEC, pH, and alkalinity. One sample per day
was collected for determining load bearing capacity by
combining four parts treated sands with one part treated
fines. In addition, on the fourth day, a sample of the mixed
sands and fines was also collected for bioassays. It should
be noted that TCAAP and MPCA believed that treated soil
generated during the entire remediation effort would be
composed of four parts sand and one part fines. TCAAP
and MPCA determined if remediation goals were attained
by analyzing a mixture of four parts of treated sands to one
part of treated fines and comparing the contaminant levels
in the mixture with the remediation goals.
Oversized material samples were collected hourly and
were composited to collect one sample per day to be
analyzed for TCLP metals. Samples of clay balls were
also collected to be analyzed for TCLP metals and total
lead content. One sample per day was collected from the
coarse sands and the density separated lead concentrates
for lead analysis. Samples of lead concentrates were
collected once at the conclusion of the demonstration.
Liquid samples from the leaching circuits were collected
once per day to be analyzed for lead.
35
-------
Table 4-1. Outline of Sample Collection and Reid Measurement Program
ON
1.
2.
3.
4.
5.
6.
Sample and Location
Feed soil (A)
Untreated oversize soil
(Bl)
Treated oversize soil (B2)
Untreated clay ball (B1L)
Treated clay ball (B2L)
Treated sandsf(C)
Parameter
Lead6
Metals
TCLP"
PSD
% Moisture
Load bearing capacity
Alkalinity
CEC
PH
Bioassay"
Mass'
TCLF
TCLPC
Lead"
TCLP
Lead'
TCLF
Leadc
Metals
TCLP
PSD
% Moisture
CEC
pH
Alkalinity
Bioassayg
Load bearine capacity8
Sampling
Method
C
C
C
C
C
C
C
C
C
C
G
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
Sample
Matrix
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
No. of
Samples
12
4
4
4
12
3
4
4
4
1
NA
4
4
4
4
4
4
34
4
4
4
34
4
4
4
1
3
Blanks'
ld
jj
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4d
1<
NA
NA
NA
NA
NA
NA
NA
NA
Matrix
Spikes'
2
1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
NA
NA
NA
Matrix Spike
Duplicates'1
2
1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
NA
NA
NA
Total
Samples
17
7
4
4
12
3
4
4
4
1
NA
4
4
4
4
4
4
46
7
4
4
34
4
4
4
1
3
-------
Table 4-1. Outline of Sample Collection and Field Measurement Program (continued)
7.
8.
9.
10.
11.
Sample and Location
Treated fines'1 (D)
Density separated lead
concentrate1 (F1-F5)
Cell lead concentrate (GI-
GS)
Sands before leachingf
(KT)
Fines before leaching (P)
Parameter
Lead*
Metals
TCLF
PSD
% Moisture
CEC
PH
Alkalinity
Bioassay8
Load bearing capacity8
Leadc
Mass
Leadc
Mass
Lead0
Metals
PSD
% Moisture
CEC
PH
Alkalinity
Lead0
Metals
TCLF
PSD
% Moisture
CEC
PH
Alkalinity
Sampling
Method
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
Sample
Matrix
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
No. of
Samples
35
4
4
4
35
4
4
4
1
3
15
15
3
1
33
4
4
33
4
4
4
33
4
4
4
33
4
4
4
Blanks'
4"
ld
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
NA
Matrix
Spikes6
4
1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
NA
Matrix Spike
Duplicates6
4
1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
4
1
NA
NA
NA
NA
NA
NA
Total
Samples
47
7
4
4
35
4
4
4
1
3
15
15
3
1
45
7
4
33
4
4
4
45
7
4
4
33
4
4
4
-------
r
Table 4-1. Outline of Sample Collection and Field Measurement Program (continued)
Ul
00
12.
13.
14.
15.
16.
17.
18.
Notes:
NA
C
TI
Sample and Location
Pregnant sands leachant
(L)
Regenerated leachant from
dewatering (N1-N6)
Pregnant fines leachant
(R1.R2)
Leachant (X)
Water (Y)
Neutralizer (Z)
Hocculent (YY)
Parameter
Lead0
Lead'
Metals
PH
Lead1
Mass
Volume
Mass
Mass
Sampling
Method
G
G
G
G
G
TI
TI
Tl
TI
= Not applicable
= Composite sample
= Time integrated sample
Sample
Matrix
L
L
L
L
L
S
L
S
S
G
S
L
No. of
Samples Blanks'
16
46
46
46
17
4
1
4
4
=
2
4
2
NA
1
NA
NA
NA
NA
Grab sample
Solid sample
Liquid sample
Matrix
Spikes"
2
4
NA
NA
1
NA
NA
NA
NA
Matrix Spike
Duplicates*
2
4
NA
NA
1
NA
NA
NA
NA
Total
Samples
22
58
48
46
20
4
1
4
4
One solid stream equipment rinsate blank was generated from me sampling equipment after final decontamination (after each run).
Field duplicates, field blanks, and matrix spikes/matrix spike duplicates (MS/MSD) were analyzed for 5 percent of the total number of samples per
matrix type. The quality control analyses concentrated on the primary sampling points.
Critical parameters; TCLP tests were conducted for lead, chromium, and mercury.
Eight rinsate blanks for all parameters; the SITE team analyzed only those blanks necessary to give a blank to sample frequency of 5 percent.
COGNIS monitored the mass of the feed soil by using a scale on the conveyor belt
An additional four composite samples, one during each day, were collected from these media for grinding, screening, and lead analysis.
These tests were conducted by compositing treated sands and treated fines.
-------
The feed soil, untreated and treated sands, and density
separated lead at this site all contained lead shards.
Concerns arose regarding homogeneity of these samples.
Laboratory analysts typically collect a 1 - to 2-gram sample
from the sample container to be analyzed for metals. Lead
shards present in these containers can weigh more than 2
grams. Therefore, potential bias may be introduced when
collecting and analyzing samples. It was determined that
grinding all of the soils in the sample container from these
sample streams before analysis would be the most
appropriate method for obtaining representative samples.
A Retsch/Brinkmann centrifugal grinding mill, Model
ZM-1, with a carbide-coated rotor, cyclone collection
chamber, and carbide-coated ring sieve, was used.
Approximately 1 kg of dry weight sample was ground for
2 minutes. Between samples, all parts of the grinder that
had come in contact with the sample were washed with
soap and water, rinsed with hot water, and then rinsed with
methanol. About 500 grams of laboratory grade Ottawa
sand was ground and analyzed after every 10 samples.
An experiment was conducted to determine if a
representative sample could be collected from ground
material. Sand samples were spiked with lead balls and
ground to pass through a sieve with 425 urn mesh. Ten
replicate samples of the ground material were collected
and analyzed for lead. The results indicated that the
grinding procedure produces a well mixed sample. The
recovery ranged from 84 to 93 percent and met the MSD
analysis acceptance criterion of 25 percent relative percent
difference (RPD) established for this project in the QAPP.
4.2 Physical and Chemical Analyses
Physical and chemical parameters that determine
characteristics of the soil that may affect treatment were
monitored for this project. These parameters were also
monitored to determine if significant changes occur in soil
characteristics due to treatment. The physical and
chemical parameters that were monitored include PSD,
percent moisture, CEC, pH, alkalinity, load bearing
capacity, and mass.
PSD data for the feed soil, treated and untreated sands, and
treated and untreated fines are presented in Table 4-2.
According to COGNIS, treatability data shows that fines
are defined as particles smaller than 106 |im in diameter.
For this project, particles larger than 6.35 mm in diameter
were defined as oversized material. Sands are defined as
particles that are smaller than the oversized material yet
are larger than the fines. PSD data available from the
laboratory were used to determine percent sands, fines,
and oversized material.
The PSD data showed that, over the 4-day demonstration,
the average composition of the feed soil was 17 percent
oversized material, 54 percent sand, and 28 percent fines.
These data show that the sands to fines ratio for the soil
tested during the demonstration is about 2 to 1. Although
lead concentrates obtained through the density separation
process were removed from the treated sand stream, the
weight of this material is insignificant when compared to
the weight of the treated sand. Therefore, it is likely that
the treated soil is also composed of a sands to fines ratio of
almost 2 to 1.
The untreated and treated sands fractions were similar in
PSD composition with average PSD distributions of 90.5
and 93.5 percent sand content, respectively. The average
PSD distribution for the untreated fines was 80 percent
fines and 20 percent sands, which is different from that
found in the treated fines. The treated fines stream is
composed of 68 percent fines and 32 percent sands. The
increase in sand sized particles in the treated fines samples
may be due to the addition of flocculents in the fourth
mixer-clarifier found in the fines leaching process. The
flocculents result in the formation of agglomerates. Large
amounts of sands were not expected in the fines fractions.
The terms sands and fines for this project were defined
based on project definition of sands composed of particles
larger than 106 um in diameter.
Moisture content was monitored in the feed soil, treated
and untreated sands, and treated and untreated fines
samples. Moisture content in the feed soil ranged from 3.9
to 13.4 percent, with an average of 6 percent. Moisture
content in the untreated samples was generally about 50
percent, indicating that at this stage the soils were carrying
excess water from the washing process. Average moisture
content in the treated sands and fines was 23.2 and 41
percent, respectively. Combining two parts sands to one
part fines in treated soil yields a moisture content of 29
percent. This indicates an increase of 23 percent in the
moisture content of treated soil compared to that in the
feed soil.
The California Bearing Ratio (CBR) test was used to
determine the effect of soil washing and leaching on the
load bearing capacity of the soil. The CBR is primarily
used to determine the strength of the soil for use as a
39
-------
Table 4-2. Particule Size Distribution Data (percent)
Sample Stream
Feed soil
Untreated sands
Treated sands
Untreated fines
Treated fines
Day
1
2
3
4
1
2
3
4
1
2
3
4
I
2
3
4
1
2
3
4
Oversized
Material'
20
24
8
17(18)"
2
2
1
1
2
2
3
4
0
0
0
0
0
0
0
0
Sands*
50
48
60
58(59)
92
87
92
92
95
93
94
94
20
20
16
23
30
29
30
38
Fines"
30
28
32
25(23)
6
11
7
7
3
5
3
2
80
80
84
77
70
71
70
62
Notes:
Particles > 6.35 mm in diameter
6.35 mm < particles > 106 /an
Particles < 106 toa
Values in parentheses are for results of duplicate samples analyses
structural fill material. Feed soil and treated soil samples
were collected during days 2, 3, and 4 of the
demonstration. On day 2, a duplicate sample of the feed
soil was also tested. The CBR values for the feed soil
indicate that it is suitable for use as fill material. The CBR
increases by 2 to 4 times after soil washing and leaching.
The greatest increase in CBR occurs at the higher
composition rates (that is, up to 100 percent composition).
The increase in CBR indicates that due to removal of
oversized material, the treated soil is more compact
compared to the feed soil, and its load bearing capacity
increases significantly after being washed. This indicates
that the treated soil can be used as structural fill.
Alkalinity and pH were monitored to characterize soils.
Liquid stream samples showed pH levels of about 2 in
most samples. Only two samples collected from the sand
leaching circuit had a pH above 3. The COGNIS leaching
process is designed to be effective at a low pH. These pH
values above 3 show that COGNIS may have been unable
to lower the pH to optimum levels in one of the six
leaching circuits. In treated sands and fines, pH levels
were all near or above 7, indicating that the COGNIS
process was able to neutralize the treated soil. Alkalinity
in the feed soil ranged from 1,800 to 3,700 mg/kg, with an
average value of 2,700 mg/kg. The pH in the feed soil
ranged from 7.9 to 8.1. The average alkalinity of the
treated sands and fines was 4,250 and 5,325 mg/kg,
indicating that COGNIS was able to counter the effect of
acidic leachant on the soils.
Average CEC values in the sand stream were 2 and 4
40
-------
milliequivalent (meq) per 100 grams (g) compared to 20
meq/100 g in the fines. CEC values for sand streams were
much lower than those for the fines. CEC values for feed
soil samples were comparable to each other (ranging from
8.1 to 11.7 meq/100 g).
Mass of feed soil monitored by COGNIS indicated that
about 432 tons of soil was processed by the combined
BESCORP and COGNIS processes during the 4 days of
the SITE demonstration. This corresponds to a soil
treatment rate of about 11 tons per hour (12 tons per hour
with 90 percent efficiency), which corresponds well with
the 9.3 tons per hour rate observed during the entire
remediation (Wenck 1995a). Based on PSD data, 17
percent of this soil consists of oversized material, and was
not treated by COGNIS. Therefore, based on laboratory
data on oversized material, although BESCORP treated
432 tons of soil, COGNIS treated about 360 tons of soil
during the 4-day demonstration. However, the laboratory
data on oversized material is expected to only include
rocks and not the clay balls. The amount of oversized
material in the feed soil was monitored on four occasions
during the demonstration. The amount of oversized
material in the feed soil, composed of rocks and clay balls,
was measured at 15,27,36, and 42 percent, for an average
composition of 30 percent. Also, the percent of rocks and
clays monitored in the feed soil and the treated and
untreated material indicate that rocks comprised from 21
to 33 percent of the oversized material. The rest of the
oversized material, therefore, are composed of clay balls
which was not measured in the laboratory. It apppears that
the percentage of oversized material hi the feed soil was
higher than the laboratory reported value of 17 percent.
Therefore, the actual amount of soil treated by COGNIS
during the 4-day SITE demonstration was less than 360
tons.
4.3 Chemical Analyses
Analytical results for all media are presented in this
section.
4.3.1 Feed Soil
The analytical results for lead in the feed soil to the
BESCORP soil washing unit are presented in Table 4-3.
The daily average lead concentrations range from a low of
530 mg/kg to a high of 1,027 mg/kg. The individual
composite samples collected during the demonstration
ranged from 380 mg/kg to 1,800 mg/kg, indicating
considerable heterogeneity in the feed soil. Lead
concentrations in all composite feed soil samples were
above the target remediation levels. The average lead
concentration in all feed soil composite samples is 824 mg/
kg-
Table 4-4 presents the analytical results for other metals in
feed soil to the BESCORP soil washing unit; treated and
untreated sands; and treated and untreated fines from and
to the COGNIS process. Data presented in Table 4-4 show
that chromium and nickel concentrations in untreated and
treated sand samples are approximately 4 to 5 times higher
than in the feed soil. The ring sieves and the carbide-
coated rotor associated with the grinder are made of
stainless steel that contains chromium and nickel
(Brinkmann 1996). Furthermore, high chromium, copper,
and nickel concentrations were observed in sand blanks
used in evaluating the grinder. It appears that elevated
concentrations of chromium and nickel in sand samples
that were ground are due to contamination from grinding
equipment. Chromium, nickel, and copper data for feed
soils, which were also ground, are also of questionable
quality. Therefore, the impact of soil washing and
leaching on removal of chromium, nickel, and copper on
feed soil and sand-sized soils is not discussed further in
this report.
Concentrations of heavy metals in composite feed soil
samples collected daily show that antimony was not
detected above 9 mg/kg. Results presented in Table 4-4
indicate that cadmium was not detected in the feed soil or
any other medium sampled during the demonstration.
TCLP test results conducted with feed soil to the
BESCORP soil washing unit, oversized material from the
BESCORP unit and, treated and untreated fines, and
treated sands from the COGNIS process are presented in
Table 4-5. All four TCLP tests conducted with the feed
soil to the BESCORP soil washing unit showed lead
concentrations in excess of the 5 mg/L maximum
allowable concentration.
In all 29 TCLP extracts collected from material to and
from both the BESCORP soil washing unit and the
COGNIS process, cadmium and chromium concentrations
were below their maximum allowable concentration of 1
and 5 mg/L, respectively.
41
-------
Table 4-3. Lead Concentration in the Feed Soil to the BESCORP Soil Washing Unit (mg/kg)
Daily Average
Day
1
2
3
4
Overall Average
Composite
Number
1
2
3
1
2
3
1
2
3
1
2
3
Lead
Concentration
720
430
1,300
1,800
890
390
800
l,700(810)(950)b
740(890)°
380
750
460
Lead Concentration
±SDa
817 ±443
—
--
1,027 ±715
-
--
923 ±200
-
-
530 ± 195
~
-
824 ±424
Notes:
SD = Standard deviation
Values in the parentheses are for replicate soil samples; the daily average ± SD was
calculated following calculation of the average value for each composite sample
Values in the parentheses are for results of duplicate analyses
42
-------
Table 4-4. Non-Lead Metal3 Concentrations in Soil Samples (mg/kg)
Remediation Goals
4.0
4.0
100.0
80.0
45.0
Medium
Feed soil
Untreated sands
Treated sands
Untreated fines
Treated fines
Day
1
2
3
4
1
2
3
4d
1
2
3
4
1
2
3
4
1
2
3
4
Antimony
0.77
0.78
4.5
0.33
1.0(0.42)°
0.81
1.4
5.2
1.3
2.0
3.7
1.2
ND
0.53
ND
ND
ND
ND
ND
ND
Cadmium
NDb
ND
ND
ND
ND(ND)
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Chromium
37
37
48
43
160(140)
120
150
7.5
230
190
190
170
41
40
93
38
22
24
25
25
Copper
150
240
300
310
140(270)
460
340
79
280
370
160
91
360
480
360
340
54
93
78
81
Nickel
47
41
54
58
160(170)
140
170
9
200
160
160
140
40
44
38
36
23
22
27
28
Notes:
* Mercury was not analyzed because all samples exceed holding times
" ND = Not detected
c Values in the parentheses are for duplicate analytical results
d Sample may have been mislabeled and therefore is not used in the evaluation
Bolded values indicate samples may have been contaminated during grinding
43
-------
Table 4-5. Toxicity Characteristic Leaching Procedures Test Results (ug/L)
TCLP Maximum Allowable Concentration" (ue/L)
5.000
1,000
5,000
Medium
Feed soil
Untreated oversized material
Treated oversized material
Untreated clay balls
Treated clay balls
Untreated fines
Treated fines
Treated sands'
Day
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
Lead
6,200
17,000
6,300
12,000
NAC
3,400
1,800
140
11,000
5,300
410
NA
36,000(180)d
NA
1,200
170(8,600)e
NA
NA
230
ND(7,600)
780
2,700
1,600
1,600
100
150
140
240
3,400
7,100
4,600
5,200
Cadmium
ND"
ND
ND
ND
NA
ND
ND
ND
ND
ND
ND
NA
ND(ND)
NA
ND
ND(ND)
NA
NA
ND
ND
14
40
24
24
ND
ND
ND
16
ND
ND
ND
ND
Chromium
27
ND
20
ND
NA
11
ND
ND
10
ND
ND
NA
ND(ND)
NA
ND
ND(ND)
NA
NA
ND
ND
22
16
21
23
15
10
28
31
14
34
22
13
Notes:
Bolded value indicates sample exceeds maximum allowable TCLP concentration
ND = Not detected
NA = Not available
These two samples were different in appearance; one consisted of clay balls, the other was
made up of broken clumps.
Values in the parentheses are for duplicate soil samples
Untreated sand samples contained substantial amount of water, which could not be consistently
sampled. Therefore, TCLP analyses of untreated sand streams were not conducted.
44
-------
4.3.2 Oversized Material
Oversized material is defined as any material handled by
the soil washing subcontractor that exceeded 6.35 mm in
diameter, whether that was determined before (untreated)
or after (treated) the trommel portion of the BESCORP
soil washing process (see Figure 1-1). Daily composite
samples of untreated and treated oversized material were
collected for TCLP analysis. Oversized samples for TCLP
tests during the SITE demonstration and the overall site
remediation were not crushed. TCLP extracts of untreated
oversized material showed lead, cadmium, and chromium
at concentrations all below their respective maximum
allowable concentration. However, for treated oversized
material, lead concentrations in TCLP extracts were
greater than 5 mg/L in two of the three samples. It was
decided to collect clay ball samples on the third and the
fourth days and to analyze these samples for TCLP metals,
lead, and other metals. Lead concentrations in TCLP
extracts exceeded the maximum allowable concentration
in two of the six untreated samples and in one of the five
treated clay ball samples. The wide variation of
concentrations in the duplicate soil samples indicates the
heterogeneity of these clay balls. The lead content of the
untreated and treated clay balls ranged from 3 8 to 310 mg/
kg and 26 to 150 mg/kg, respectively. Cadmium and
chromium concentrations in TCLP extracts were either not
detected or were below the maximum allowable
concentrations. Metal concentrations in all treated
oversized clay ball samples were below remediation goals,
except for the treated sample from the third day. The
COGNIS process did not treat the oversized materials.
Exceeding the TCLP maximum concentrations in some of
the oversized material samples indicates that soil washing
alone was not always sufficient to clean oversized material
to concentrations considered nonhazardous under RCRA.
Some of this material may require further treatment before
final disposition.
4.3.3 Untreated Sands
Untreated sand is defined as the sand (particle size less
than 6.35 mm in diameter and greater than 106 p.m in
diameter) from the density separator portion of the
BESCORP soil washing unit. Samples were collected
from the untreated sand before treatment by COGNIS.
The untreated sands samples contained large amounts of
liquids that were decanted and filtered. The residues from
the liquid were added to the solid portion of the samples.
The liquid and the solid portions were analyzed separately.
The solid portions were weighed and the volume of the
liquids was recorded. Concentrations in the liquid
portions were much smaller than concentrations in the
solid portions. Only the results of analyses for solids are
listed. Table 4-6 presents the results of lead analyses when
hourly samples were collected from untreated sands. Lead
concentrations in all treated and untreated sand samples
are shown in Figure 4-1. The daily average lead
concentrations in treated and untreated sands are shown in
Figure 4-2. Lead concentrations in the untreated sand
samples ranged from 140 to 670 mg/kg. Lead
concentrations in untreated sand samples collected on day
2 were higher than lead concentrations observed on the
other days, and all concentration values for untreated sand
on that day were above the remediation goal of 300 mg/kg.
Thirteen of the 26 samples collected on days 1,3, and 4
were below the remediation goal. Considerable variations
occurred in lead concentrations in each day's samples.
The daily average lead concentrations for untreated and
treated sands and fines are presented in Table 4-7. The
average lead concentration in untreated sands was 339 mg/
kg.
As shown in Table 4-4, results of metal analyses of
untreated sands for the composite sample collected on day
4 are, for some metals, orders of magnitude different than
the analytical results for untreated sand samples collected
on other days. This sample may have been mislabeled, and
therefore was not used for this evaluation. Cadmium was
not detected in any of the untreated sand samples.
Antimony concentrations ranged from 0.81 to 1.4 mg/kg,
with none of the samples above the remediation goal of
4.0 mg/kg. The quality control samples indicated poor
recoveries for antimony; therefore, antimony values
reported in this document may be biased low.
4.3.4 Treated Sands
Lead concentrations in the treated sand samples ranged
from 71 to 640 mg/kg. Seven of the 35 treated sand
samples exceeded the remediation goal of 300 mg/kg. The
overall average lead concentration in treated sands
collected during the demonstration was 243 mg/kg.
A comparison of lead concentrations in untreated and
treated sands shows that during each of the 4 days of the
demonstration, lead removal efficiency was 44, 25, 27,
and 21 percent, respectively. Removal efficiency refers to
the estimate of the percentage of total lead that is removed
by the treatment process. The percentage removal is
45
-------
Table 4-6. Lead Content in Treated and Untreated Sands and Fines (mg/kg)
Day Hour
1 1
2
3
4
5
6
7
8
9
2 1
2
3
4
5
6
7
8
9
3 1
2
3
4
5
6
7
8
9
4 1
2
3
4
5
6
7
8
Untreated
Sands
180
220
140
180
320
330
630
360
270
670
500
380
360
380
460
400
430
330
380
NA
340
280
220
240
420
330
NA
220
270
180
290
340
490
410
240
Treated .
Sands
120
76
71
71
130
230
270
350
150
640
420
370
210
250
220
340
410
150
320
200
230
210
250
180
220
240
NA
180
200
300
290
260
290
240
160
Untreated
Fines
220
250
200
210
460
450
610
340
NA
770
720
780
510
430
430
490
340
410
440
550
490
570
450
390
510
520
NA
320
260
340
430
420
550
360
350
Treated
Fines
54
51
54
64
50
74
87
84
77
62
100
130
150
110
91
78
95
90
79
100
140
120
120
110
110
190
90
97
77
99
92
120
93
99
140
46
-------
800 —
en
300 ma/Kg
Day 1
Figure 4-1. Lead concentration in sand samples.
Day 2 I Day 3
TIME
Day 4
800—I
— +SD
Mean
SD
300 mg/kg
Day 1|Day 2|Day 3
TIME
Day 4
Figure 4-2. Daily average lead concentration in treated and untreated sands.
47
-------
Table 4-7. Daily Average Lead Concentrations with Standard Deviations (mg/kg)
Day
Untreated Sand Treated Sand Untreated Fines
Treated Fines
1
2
3
4
Overall Average ± SD°
292 ± 148
434 ±102
316 ±73
305 ± 104
339 ±122
163 ±99
334 ±149
231 ±42
241 ± 56
243 ± 113
342 ±150
548 ± 162
490 ±60
379 ± 88
443 ± 145
66 ±15
101 ±26
118 ±32
103 ±21
96 ±30
Note:
SD = Standard deviation
calculated using daily averages for lead in untreated and
treated material. The average lead levels were calculated
using nine samples collected over a 9-hour period in a day.
Removal efficiency was calculated using the following
calculation:
UMta(I
x 100 = Removal Efficiency
where:
UMkld = Average daily lead concentration in
untreated material
d = Average daily lead concentration in
treated material
The removal efficiencies are presented graphically in
Figure 4-3. It should be noted that the BESCORP and
COGNIS soil treatment process is conducted in a
continuous mode and is not conducted hi batches of soil.
Results of other metal analyses listed in Table 4-4 show
that antimony concentrations in treated sand ranged from
1.2 to 3.7 mg/kg; all treated sand samples met the
benchmark value of 4.0 mg/kg for antimony.
TCLP results show that cadmium and chromium
concentrations were either not detected or below the
maximum allowable concentrations. Leachable lead
concentrations were, however, detected in excess of 5 mg/
L in two of the four daily composite samples. Treated soils
composed of oversized material, sands, and fines were not
tested for TCLP leachable metals. TCLP data presented
for treated sands and treated fines do not allow any
conclusion regarding the leachability of treated soil.
4.3.5 Untreated Fines
Lead concentrations in all fines samples are presented in
Figure 4-4. The daily average lead concentrations in
treated and untreated fines are presented in Figure 4-5.
Lead concentrations in the untreated fines sample stream
ranged from 200 to 780 mg/kg. Results showed the three
highest lead concentrations in samples collected on day 2
when compared to samples collected during other days.
Only five of the 34 samples collected had concentrations
below the remediation goal. The daily average lead
concentrations ranged from 342 to 548 mg/kg, with an
overall average lead concentration of 443 mg/kg.
Cadmium was not detected in any of the samples and
antimony was detected in only one sample at a
concentration of 0.5 mg/kg. Copper concentrations ranged
from 340 to 480 mg/kg, indicating that all results were
above the remediation goal. Chromium and nickel
concentrations ranged from 38 to 43 mg/kg and 36 to 44
mg/kg, respectively. This indicates that all results are
below remediation goals.
TCLP tests indicated lead, cadmium, and chromium
concentrations in all extract samples were below the
maximum allowable concentrations.
48
-------
100 —
80 —
60 —
LLI
O
or
111
20
Day 1
Day 2
Day 3
Day 4
Figure 4-3. Removal efficiencies.
O)
.X
O)
O
cc
LU
O
O
O
Q
<
ai
800—\
600 —
400 —
30O mg/kfl
200 —
Figure 4-4. Lead concentrations in fines samples.
. 49
-------
800—I
UNTREATED
h
lean
-SD
300 mg/kg
Day 1 I
Day 2 | Day 3
TIME
Day 4
Figure 4-5. Daily average lead concentration in treated and untreated fines.
100 —
SO-
li 60 —
UJ
o
cc
UJ
°" 40 —
20 —
n —
' ',"'
', /•-
t ;' ''
', ?Jl
» "/• v
.. , V
-, '''*',
*"'•• ,.
^.i. <.^ '•
-T:-.
.J.*1 * f.f
' ' ** .
^ *' J' '"
! ' -v s '^ '
r^!^["i
%' '' '••
'•^'^•vVfl
• / * s' f •
.f f f f
f
f j
f
f f
*
ff
•" '_
s's •"••
ff f f
f
f
-/'- -
/**•••* !
•V
^ 1 •• '
^ *'
' 'i
^ ^ ^
•• ^ 1
' -• '
..
^ i
^ -** "" ^>
.- ^ ^ /
*•
*!• ^ **
" ''•• -•
<• ^
^ .
•*• 1- v'
/
"' , ',:
'
'
_. ^
1
"* f f
'«'
^ '
••
•* t
1
'' ''*'' '
'^'
*/''
' "f
Figure 4-6. Removal efficiencies.
Day 1
Day 2
Day 3 Day 4
50
-------
4.3.6 Treated Fines
Data presented in Table 4-6 show that lead concentrations
during the 4 days of the demonstration ranged from 50 to
190 rag/kg, with only one sample being above the
remediation target and all samples being below the
remediation goal. The daily average concentration ranged
from 66 to 118 mg/kg, with an overall average of 96 mg/
kg. Comparison of daily average lead concentrations in
untreated and treated fines show that removal efficiencies
of the fines leaching process ranged from 72.8 percent on
day 4, to 81.6 percent on day 2. The removal efficiencies
are presented in Figure 4-6.
Data for non-lead metals analyses presented in Table 4-4
show that antimony and cadmium were not detected in the
daily composite samples of treated fines. Copper,
chromium, and nickel concentrations were all below their
respective remediation goals. Average removal
efficiencies in the fines leaching process are 40, 80, and
36.7 percent, respectively. Lead, cadmium, and
chromium concentrations in the TCLP extracts were all
below their respective maximum allowable concentrations
of 5, 1, and 5 mg/L, respectively.
4.3.7 Lead Concentrates
The lead concentrates were recovered through the
BESCORP density separation process and the COGNIS
electrochemical deposition process. All recovered lead
was transported to a lead smelting facility.
The density separated lead concentrates were composed of
five different size fractions (designated as Fl through F5).
The smallest (Fl) and the largest particle size fractions
(F5) were deposited in the same container and were
analyzed as a combined size fraction. Based on
observations made during the demonstration, the mass of
the F5 fraction was much smaller than the mass of the Fl
fraction. The total weight of Fl and F5 sample streams
(1,029 kg) was more than the F2, F3, and F4 streams
combined. Approximately 1,300 kg of density separated
lead concentrate was collected during the week of the
demonstration. During the demonstration, average lead
concentrations for fractions Fl through F4 werE 45,500
mg/kg; 49,500 mg/kg; 31,111 mg/kg; and 4,200 mg/kg,
respectively. During the demonstration, approximately 48
kg of lead was collected through the density separation
process.
The lead concentrate samples from the sands and fines
leaching circuit were sampled at the end of the
demonstration to determine the amount of lead recovered
by the COGNIS process. Approximately 694 kg of lead
concentrate, with an average concentration of 74,000 mg/
kg was recovered from the sands leaching circuit.
Approximately 714 kg of lead concentrate, with an
average concentration of 156,700 mg/kg, was recovered
from the fines leaching circuit. The COGNIS process
recovered a total of about 153 kg of lead. These lead
recovery values from the sands and fines leaching circuit
may not reflect actual recovery during the demonstration.
COGNIS periodically cleans the lead cassettes in the
leaching circuits. During the week of the demonstration,
COGNIS did not recover any leached lead. Lead was
recovered on the Monday following the week of the
demonstration. The weights of lead concentrates reported
in this paragraph were based on data collected on the
Monday following the demonstration. However, it is not
known when lead was recovered before the demonstration.
Therefore, this at least represents 6 days of activities at the
site. During the 4 days of the demonstration, at least 100
kg of lead was recovered.
4.3.8 Overall Treatment of Soils
Daily average values for treated soils were determined by
combining daily average lead concentrations in treated
sands and treated fines in a 2 to 1 sands to fines ratio. A 2
to 1 sands to fines ratio was used because data from the
demonstration indicate that the soil from the TCAAP Site
F soil is made up of 2 parts sand to 1 part fines. These data
are presented in Table 4-8.
The data collected during the SITE demonstration show
that the soil treated through the combined BESCORP and
COGNIS processes were less than 300 mg/kg during all 4
days. It should be noted that TCAAP and MPCA assumed
that the treated soil was composed of four parts sands and
one part fines. TCAAP and MPCA determined attainment
of remediation goals and targets by a sampling,
preparation, and analysis protocol that was independent
from the SITE demonstration. Under their protocol, four
parts treated sands to one part treated fines were mixed and
the contaminant concentrations were compared with the
remediation goals. When daily average lead concentrations
for the SITE demonstration were combined at a 4 to 1 ratio,
the values were comparable to those listed in Table 4-8
with all treated soil concentrations below the remediation
goals. Comparison of the TCAAP and SITE data over the
51
-------
demonstration week shows that TCAAP data (generated
by a different sampling and analysis protocol) was
substantially lower (76 mg/kg) than SITE data (194 mg/
kg). Evaluation of data in Table 4-8 shows that the daily
average removal efficiencies through the COGNIS
process range from 40.9 percent on day 4 to 57.6 percent
on day 1. Comparing the daily average lead
concentrations of untreated sands and fines with those for
the feed soil indicates that size separation and soil washing
through BESCORP's process achieved lead removal
efficiencies ranging from 37.7 percent on day 4 to 62.2
percent on day 1. Comparison of the treated soil data with
the daily average lead concentrations in feed soil shows
that the combined removal efficiencies of the BESCORP
and COGNIS processes range from 63.2 percent on day 4
to 84 percent on day 1. These average removal efficiencies
are presented in Figures 4-7 and 4-8. These lead removal
efficiency values compare well with those reported by
Wenck (1995a) of 55 to 90 percent observed during
remediation. Also during the demonstration, three of the
eight oversized material samples (36 percent) did not meet
the maximum allowable concentrations determined by
TCLP testing. Wenck (1995a) reported that 33 percent of
the samples tested for TCLP did not meet maximum
allowable concentrations after the first pass through the
treatment system and were successfully treated by a
second pass through the treatment system.
The feed soil and the treated material contained oversized
particles. However, the removal efficiencies presented in
this report are limited to the sands and fines, which were
treated by the COGNIS process. Observation of COGNIS'
operations made before the SITE demonstration showed
minimal contamination associated with the oversized
material. It was assumed that contamination associated
with the oversized material would be insignificant during
the demonstration. Therefore, attainment of cleanup
objectives for oversized material during the SITE
demonstration and the overall site remediation period was
designed to be determined by TCLP.
4.3.9 Liquid Samples
Liquid samples consisting of the pregnant sands leachant,
pregnant fines leachant, and the regenerated leachant,
were collected to provide information on the lead recovery
system. One pregnant sands leachant stream and two
pregnant fines leachant streams were introduced into six
electrochemical cells; two of the cells received pregnant
sands leachant and four received pregnant fines leachant.
Data for daily average lead concentrations are presented in
Table 4-9. Concentrations in the pregnant sands leachant
stream varied by about a factor of 5, ranging from 21 to 100
mg/L. The daily average lead concentrations ranged from
51.5 to 80.5 mg/L, with an overall average of 66.5 mg/L.
Average concentrations in the two fines streams were 40.4
Table 4-8. Combined Average Lead Concentrations for COGNIS Process Soil3* (mg/kg)
Day
Untreated Sands and Fines
Treated Sands and Fines
1
2
3
4
Overall Average
309
472
374
330
371
131
256
193
195
194
Notes:
Includes size separated soil from the BESCORP soil washing unit that was feedstock for
the COGNIS process
Combining sands and fines data at a ratio of two parts sand and one part fines
52
-------
100
80 —
60 —
Ul
O
cc
UJ
20 —
Day 1
Day 2
Day 3
a ^2,
Day 4
Figure 4-7. Average removal efficiencies of the COGNIS process.
100 —
80 —
t 60H
UJ
o
40
20 —
life
f''i#>~;
',/*•>*
V»*
f £Cf ^ f
lA&f-r'S ,'
f'P/??'^
yw * j^
>A jf f *
^
^r"
tf^**-
I* •• .* \
* •£ V"
(>»' .5
-%''/•-
»$^
t&fr»'
%(&"
•$'&<'!.
Day 1
Day 2
Day 3
Day 4
Figure 4-8. Average removal efficiencies of the COGNIS and BESCORP processes.
53
-------
Table 4-9. Average Lead Concentrations in the Liquid Streams (mg/L)
Day
1
2
3
4
Overall
Average
Pregnant
Sands
Leachant
51.5
71.5
80.5
62.5
66.5
Pregnant Fines
Leachant
No.l
29.5
51.5
46.0
34.5
40.4
Pregnant Fines
Leachant
No. 2
17.0
27.5
25.5
21.0
22.8
Regenerated
Leachant from
Sands
3.6
2.9"
14.3
7.5
7.0
Regenerated
Leachant
from Fines
2.9
3.9
7.0
4.7
4.6
Note:
Two of three data points were below the detection limit. The average was calculated by
assigning a value of 25 (tgfL, one-half the lead detection limit, to the nondetected samples.
and 22.8 mg/L. Overall patterns in all pregnant liquid
streams are similar; concentrations on days 1 and 4 are
comparable but lower than those on days 2 and 3. This
pattern corresponds well with the trend observed in the
feed soil and the untreated sands and fines soil streams
where lead concentrations were also higher on days 2 and
3 when compared to days 1 and 4.
Lead concentrations in the regenerated leachant streams
are also highly variable. Lead concentrations ranged from
below the detection limit to 26 mg/L hi the sands leaching
system and from 1.6 to 26 mg/L in the fines leaching
system. Most of the concentrations were below 10 mg/L;
concentrations above 10 mg/L were found only on days 3
and 4. Comparing lead concentrations in pregnant streams
with the regenerated streams shows a reduction of 89
percent in the sands leaching system and a reduction of 85
percent in the fines leaching system.
Lead removed from the leaching system was recovered as
lead concentrates. The wastewater generated at the
season's end met pretreatment standards for disposal in a
POTW.
4.3.10
Results of Toxicity Studies
In addition to meeting treatment goals and measuring
removal efficiencies, reduction of toxicity is another
indicator of determining effectiveness of a treatment
process.
Uptake Biokinetic Model Results
EPA recommends using the Uptake Biokinetic Model for
lead to determine the total lead uptake for children 0 to 6
years old, from such exposure pathways as inhalation, diet,
soil and dust ingestion, and maternal exposure. However,
the Uptake Biokinetic Model was not used at Site F
because RCRA closure regulations require a background-
based cleanup. A blood lead concentration, in micrograms
per deciliter (ng/dL), can then be estimated based on the
total lead uptake. EPA (1991b) recommends a maximum
blood-lead concentration of 10 ng/dL, which is at the low
end of the range of concern for adverse health effects in
children (10 to 15 ug/dL). The Uptake Biokinetic Model
(EPA 1994b) estimates the percentage of population that is
expected to develop blood-lead concentrations above and
below 10 ug/dL. The Uptake Biokinetic Model was run at
54
-------
the average feed soil concentration of 824 mg/kg, the
average concentration in soil treated through BESCORP's
process of 371 mg/kg, and the average concentration in
soil treated through the COGNIS process of 194 mg/kg.
The model was run with default values for lead
concentrations in air, drinking water, plant intake, and
maternal contribution. The results of the model runs are
presented in Figures 4-9 through 4-11. The figures show
that for exposure to soil lead concentrations of 824 mg/kg,
371 mg/kg, and 194 mg/kg, percentages of children
expected to have blood-lead concentrations of less than 10
u.g/dL are 68,94, and 99 percent, respectively. It is evident
that the percent of the population exposed to blood-lead
concentration in excess of 10 ng/dL decreases as the soil is
treated through the BESCORP process and further
decreases following treatment by COGNIS.
Bioassay Results
Bioassay tests were conducted on samples from the feed
soil and from mixed treated sands and fines in a ratio of
four parts sands to one part fines. In addition, tests were
also conducted after rinsing the treated soil with water to
simulate conditions encountered after about 3 months of
exposure to rains when treated soil is used as fill material.
The earthworm 14-day acute toxicity bioassay using
Eisenia fetida and Lumbricus terrestris and the seed
germination/root elongation (SG/RE) tests using lettuce,
oats, and two local grass seeds were conducted by
following EPA's National Risk Management Research
Laboratory (NRMRL) Standard Operating Procedures to
evaluate the toxicity of the three soil samples. The
earthworm toxicity test results show that the feed soil and
the treated soil after rinsing were not toxic to the
earthworm, while the treated soils without rinsing showed
significant toxicity. In the SG/RE tests, lettuce showed
significant inhibitions with all three soils, while the treated
soil without rinsing was the most toxic. For tests with oats
and grass seeds, the treated soil without rinsing was the
most toxic and the treated soil after rinsing was the least
toxic. Plant genotoxicity bioassays conducted with Allium
(common onion) showed that the feed soil and the treated
soil induced similar concentrations of genetic damage to
root tip cells. However, after rinsing, the treated soil was
found not to be genotoxic. These results show that in
general, toxicity increases after treatment with the
COGNIS process, and decreases after treatment and
rinsing. High concentrations of salt generated during the
remediation process seems to be responsible for the
increased toxicity of unrinsed soil in both plants and
earthworms.
4.4 Conclusions
Based on the SITE demonstrations, the following
conclusions may be drawn about the performance of the
COGNIS and BESCORP processes for soil contaminant
matrices found at TCAAP:
The combined removal efficiencies of the
BESCORP and COGNIS processes ranged from
63.2 to 84 percent, with an overall average removal
efficiency of 76.4 percent. Treated soil (combin-
ing treated sands and fines data) met the lead
remediation goal of 300 mg/kg on all 4 days of
the demonstration.
Daily average lead removal efficiencies of the
COGNIS process alone range from about 41 per-
cent to about 58 percent.
• The COGNIS sands leaching process achieved
lead removal efficiencies ranging from 21 to 45
percent. Lead concentrations in untreated sands
ranged from 140 mg/kg to 670 mg/kg. Lead con-
centration in treated sand ranged from 71 mg/kg
to 640 mg/kg. The overall average lead concen-
tration in untreated sand is 339 mg/kg. The over-
all average lead concentration in treated sand is
243 mg/kg. Based on an average concentration,
lead was reduced in the sand leaching process by
96 mg/kg. Of the 35 samples collected of untreated
sand, 20 samples exceeded 300 mg/kg. Of the 35
samples collected of treated sand, 7 samples ex-
ceeded 300 mg/kg. However, due to differences
in sampling, preparation, and analysis protocol,
SITE demonstration data differs from MPCA com-
pliance data. Nine percent of the compliance
samples did not meet the MPCA remediation goal
for lead.
• Lead removal efficiencies in the fines leaching
process ranged from 72.8 to 81.6 percent. Lead
concentrations for untreated fines ranged from 200
mg/kg to 780 mg/kg with an overall average of
443 mg/kg. Lead concentrations in treated fines
ranged from 50 mg/kg to 190 mg/kg. Based on
average concentrations, lead was reduced in the
fines leaching process by 347 mg/kg. Of the 35
samples collected of untreated fines, 27 samples
exceeded 300 mg/kg. Of the 35 samples collected
of treated fines, all samples were below 300 mg/
kg-
55
-------
IS
v 5
CutofF: 10.0pg/dL
% Above: 0.97
% Below: 99.03
G. Mean: 3.4
4 6 8 10
BLOOD LEAD CONCENTRATION G»E/dL)
0 TO 84 Months
Figure 4-9. Blood-lead probability density in children (age 0 to 6 years) exposed to 194 mg/kg soil-lead concentration.
• i • . • L
Cutoff: 10.0|ig/dL
% Above: 6.03
% Below: 93.97
G.Mean: 4.9
2 4 6 8 10 12 14 16 18 20 22 24 26
BLOOD LEAD CONCENTRATION Gig/dL)
0 TO 84 Months
Figure 4-10. Blood-lead probability density in children (age 0 to 6 years) exposed to 371 mg/kg soil-lead concentration.
56
-------
I
a J
2
PH
Cutoff: 10.0|ig/dL
% Above: 32.10
% Below: 67.90
G.Mean: 8.3
10 15 20 25 30 35 40 45
BLOOD LEAD CONCENTRATION Qig/dL)
0 TO 84 Months
Figure 4-11. Blood-lead probability density in children (age 0 to 6 years) exposed to 824 mg/kg soil-lead concentration.
Three of the eight TCLP extracts of the oversized
material exceeded the maximum allowable con-
centrations for lead. Therefore, some of this over-
sized material may need further treatment to be
removed from consideration as a hazardous waste
under RCRA.
The COGNIS process likely requires pretreatment
of soils because it separately treats sand and fines.
For TCAAP Site F remediation, pretreatment for
the COGNIS process included BESCORP's soil
washing and size separation process. In addition,
removal of ordnance and other debris from feed
soil is considered to be pretreatment.
BESCORP's physical separation and soil wash-
ing stage was effective in removing oversized
material, recovering lead concentrates, and sepa-
rating sands and fines. However, BESCORP was
not successful in addressing clay balls. Clay balls
were removed from the system as oversized mate-
rial. Clay balls may represent a significant source
of lead. Because clay balls by passed treatment,
the COGNIS process treated a reduced amount of
fines. According to COGNIS, clay ball formation
was not a typical problem during the long-term
treatment. The BESCORP process achieved lead
removal efficiencies ranging from 37.7 percent to
62.2 percent during the 4 days of operation.
TCLP results for the treated sands indicate
exceedance of the lead maximum allowable con-
centration of 5 mg/L in two of the four samples.
There were no exceedances of the TCLP maxi-
mum allowable concentrations for lead, cadmium,
or chromium in the four treated fines samples.
It is evident from the Uptake Biokinetic Model
runs that toxicity of soil was reduced due to treat-
ment. Bioassay results indicated that increased
soil toxicity for earthworms and plants was due to
the increased salt content of the treated soil com-
pared to the feed soil. After treatment and rinsing
that simulates rainfall, soil is less toxic.
57
-------
COGNIS and BESCORP processes generate sev-
eral treatment residues. The wastewater, gener-
ated at the conclusion of treatment or at the
season's end, requires treatment prior to disposal.
The lead concentrates can be recycled at lead
smelting facilities. Treated soils at TCAAP met
the MPCA enforceable remediation goals. Some
of the oversized material did not meet the RCRA
enforceable standards for TCLP extracts and may
require further treatment. Ordnance and other
debris removed from the soil may require disposal
at off-site facilities.
Activities such as site preparation, arrangements
for utilities, waste disposal, mobilization, and de-
mobilization were carried out on schedule and ac-
cording to plan. All BESCORP and COGNIS
equipment operated smoothly during the SITE
demonstration.
Based on treating 10,000 tons of soil, the cost of
treatment is $182 per ton. At sites that do not re-
quire site preparation and residual shipping and
handling costs, the cost for treatment is $138 per
ton. The cost for leaching activities only is $70
per ton for treatment of 10,000 tons of soil.
58
-------
Section 5
Technology Status
The information in this section was provided by COGNIS,
Inc., which is the technology developer, and Doe Run
Company, which now owns the equipment and the license.
As noted in the text, the SITE program evaluation of the
COGNIS process occurred during four days of a much
longer full-scale cleanup operation that COGNIS was
contracted to complete at TCAAP. Below is an article from
Remediation by William E. Fristad, who worked in the
area of technology development, laboratory operations
and plant support for COGNIS, Inc., that describes the
overall cleanup effort by COGNIS, Inc., at TCAAP. This
article is reprinted with permission of John Wiley and
Sons, Inc.
The TERRAMET® process equipment and license to the
technology was sold to The Doe Run Company when
Henkel Chemical Company, the parent of COGNIS, INC.,
exited the remediation market. Upon completion of the
Twin Cities Army Ammunition Plant project the
equipment was demobilized from Minnesota to Doe Run's
Buick Resource Recycling Facility in Boss, Missouri.
The Doe Run Company has been producing lead for the
past 140 years. Seven years ago, Doe Run completed the
construction on a new facility that would recover and
recycle lead from scrap. Doe Run is committed to the
recycling of metals, particularly lead, at the Buick
Resource Recycling Facility. Each year the facility
recovers and recycles lead from over 175,000 tons of lead
bearing products, by-products, wastes and residues. Buick
has recovered and recycled lead from over 25 firing ranges
from National Guard units around the country. Soil and
other waste streams have been treated at the facility from
Department of Energy, Naval Facilities Engineering
Command, and Government Owned/Contractor Operated
(GOCO) facilities.
The Doe Run Company integrated the equipment into its
existing RCRA Part B permitted facility. The
TERRAMET process complements Doe Run's existing
capability to process lead contaminated sands and soils
through its pyrometallurgical circuit. The new process is
called TERRAMET Services. The purpose is to provide a
low cost permanent solution to metals contaminated soil
and other solid waste streams. Contaminated soil can be
shipped to the facility, the metals removed and reused to
form lead and lead alloy ingots, and the soil put to
beneficial reuse as part of the mining operation. Shipping
contaminant offsite to a permitted facility benefits those
who require fast reuse and transfer of clean parcels. If
required, TERRAMET equipment can still be mobilized to
projects. Special projects, such as the separation of lead
from depleted uranium at Lake City Army Ammunition
Plant in Independence, Missouri, will require the plant to
move onsite to the project.
TERRAMET Services offers flexibility in treating waste
on or offsite plus in choosing the treatment process at
Buick Resource Recycling to minimize cost.
Waste Classification - Determine if the soil or waste
stream should go straight to the lead smelter, thereby
classifying the waste as nonhazardous based on the
RCRA recycling exemption. This classification
greatly reduces transportation costs, storage
requirements, and turnaround time.
Soil Washing - If the metals can be removed
physically, the soil will only undergo soil washing.
The lead concentrate will go directly to the smelter.
The soil with lead below the state's required lead
concentration will be declared clean and used as cover
in the mining operation.
59
-------
Soil Washing/Soil Leaching - If the metals cannot be
removed by soil washing alone because the lead or
other metals have weathered or have been smeared
into the soil, then the soil will be washed to remove
particulate metal and leached to remove the remainder
required to reach state requirements. Again, the lead
concentrate would go to the smelter and the soil would
be reused.
Pyrometallurgical - Some metals and minerals are
required in the smelting process. The soil can go
directly to the smelter depending on the operation and
the feed material make up.
Selective Processing - Doe Run personnel can work
with generators to custom fit a treatment process for a
particular waste. There are instances, driven by
regulation or waste characterization, that require a
certain process step, end use, or remediation goal. Doe
Run will combine unit operations, segregate and
process "hot spots," or modify its system (in
compliance with the Part B permit) to accommodate
special needs.
The price of soil treatment at Buick is lower than the
field price because there is no mobilization, no per
diems, and the labor force required to treat the soil is
already in place manufacturing lead. Raw material
handling, health and safety activities, material handling,
and maintenance are all performed at a much lower rate
at a fixed facility than in the field.
American Society for Testing and Materials (ASTM).
1994. Methods Published Annually by ASTM.
Doe Run Contact
Lou Magdits, TERRAMET Manager/Raw Materials
Manager
The Doe Run Company
Buick Resource Recycling Facility
HwyKK
HC 1 Box 1395
Boss, MO 65440
(573) 626-3476
(573) 626-3405 Fax
hnagdits@misn.com
60
-------
Reprinted by permission of John S. Wiley & Son, Inc. from Remediation,
Autumn 1995, pp. 61-72, copyright 1995.
Case Study: Using Soil
Washing/Leaching for
the Removal of Heavy
Metal at the Twin Cities
Army Ammunition Plant
William E. Fristad
Dr. William E. Fristad is
Director ofTERRAMET*
Technology for COGNIS,
Inc., of Santa Rosa,
California. Over the past
four years he has been
responsible for technol-
ogy development,
laboratory operations,
and plant support. His
background is in organic
and inorganic chemistry,
and he has worked in
those fields for COGNIS
and its parent company,
the Henkel Group, in
both the United States
and Germany.
COGNIS TERRAMET® soil leaching and Bescorp soil washing systems
have been successfully combined to remediate an ammunition test burn
area at the Twin Cities Army Ammunition Plant (TCAAP), New Brighton,
Minnesota. This cleanup is the first in the country to successfully combine
these two technologies, and it offers a permanent solution to heavy metal
remediation. Over 20,000 tons of soil were treated in the project. The
cleaned soil remained on-site, and the heavy metal contaminants were
removed, recovered, and recycled. Eight heavy metals were removed from
the contaminated soil achieving the very stringent cleanup criteria of
-------
WEUUAM E. FRISTAD
The health concerns
were VOCs in the
drinking water
and heavy metals
in the soil.
heavy metals removal, recovery, and recycle process. COGNIS has now
completed a 20,000 ton cleanup of a metal-contaminated site where the soil
was cleaned to RCRA-mandated standards and returned to the site, while
all contaminant metals were recovered and recycled through a lead
smelter.
The Twin Cities Army Ammunition Plant (TCAAP), the Army, and the
Minnesota Pollution Control Agency (MPCA) saw the value in a permanent
treatment remedy which would remove the contaminants, and thereby
limit the human and environmental health threat, long-term liability, and
restrictions on real estate reuse. All parties were adamant in seeking an
alternative that minimized or eliminated material requiring permanent
off-site landfilling. This led to the selection of COGNIS to prime the
soil washing/TERRAMET® soil leaching process to remediate Site F, an
ammunition open burning/open detonation area.
A treatment train of soil washing and TERRAMET® soil leaching was
developed and assembled in less than one year from treatability study to
full-scale operation on-site by a team of COGNIS, Inc., and Brice
Environmental Services Corp. (Bescorp). The physical separation and
chemical leaching stages allow both participate and ionic metal contami-
nants to be removed, recovered, and recycled in a single integrated plant.
Live and spent small arms ordnance was also washed and removed for
proper disposal within the soil treatment plant.
TCAAP HISTORY
The Twin Cities Army Ammunition Plant, New Brighton, is an industrial
complex covering 2,370 acres in the greater metropolitan Twin Cities of
Minneapolis and St. Paul, Minnesota. It was built in 1941 and operated as
a government-owned, contractor-operated (GOCO) manufacturing facility
by Federal Cartridge Company. In the early years it produced primarily .30
and .50 caliber ammunition and 105-mm and 155-mm projectile casings.
During the Vietnam era, TCAAP also produced 7.62-mm and 5.56-mm
cartridges.
In 1982, TCAAP was included on the National Priorities List. The health
concerns were VOCs in the drinking water and heavy metals in the soil.
The practice throughout the production life of the plant was to bury
material on-site in specific disposal areas. Off-spec ordnance material and
ordnance test components were burned or detonated at Site F, the area
being remediated in this project. Through the remedial investigation
process, 14 additional sites within the TCAAP facility have been identified
as contaminated areas.
TCAAP has a history of applying innovative environmental solutions
to their contaminated areas. TCAAP entered into the first Federal Facilities
Agreement (FFA) in history during 1987. The objectives were to investigate,
monitor, and exchange information concerning historical pollution and to
provide a forum for setting future actions. Under the FFA, TCAAP
implemented one of the first applications of in-situ volatilization (ISV) for
a TCE plume in 1986. In 1987 TCAAP signed a contract to have soil
contaminated with PCBs burned on-site. Starting in 1993 TCAAP began
62
REMEDIATION/AUTUMN 1995
-------
Standard soil
washing is the use of
physical techniques
to wash the surfaces
of soil particles and
to separate soil
particles into clean
and contaminated
fractions.
CASE STUDY: USING Son- WASHING/LEACHING FOR THE REMOVAL OF HEAVY METAL
demonstrating the TERRAMET® technology for metals removal and recov-
ery. An ongoing project elsewhere on the plant is the biodegradation of
creosote-soaked utility poles. The property is currently under modified
caretaker status and will eventually be turned over for redevelopment.
STATE OF SOIL WASHING/LEACHING TECHNOLOGY
Soil washing has been a proven technology in Europe for over a
decade. In 1992, when the laboratory phase of this project began, no full-
scale soil washing job had yet been performed in the United States. In the
meantime several full-scale soil washing projects have been conducted
•which effected significant volume reduction. Soil leaching with dissolution
and recovery of metals had not yet been successfully demonstrated and
documented full-scale anywhere in the world. The TCAAP project was the
first to implement soil •washing and soil leaching -whereby the contaminants
were removed and recycled and the cleaned soil was returned on-site. Acid
leaching with acetic acid2 and hydrochloric acid3 leachants had been
investigated and described in bench-scale experiments for a number of
years.4 Previous unsuccessful attempts to leach lead with chelating agents
(EDTA) have been documented at the Lee's Farm (Woodville, Wisconsin)
and ILCO (Leeds, Alabama) sites.5 Other examples of soil washing with
\vater, surfactants, and EDTA solutions as a volume reduction method
before stabilization have also been studied,6 while the commercial
drawbacks of EDTA leaching were also described.7 The U.S. Bureau of
Mines had developed a bench-scale process by 1992; however, that
process had not yet been scaled up beyond the laboratory.8 Since then, the
Bureau of Mines has modified their leaching process and performed pilot-
scale runs on lead-contaminated soil at firing ranges.
Standard soil washing, as defined here, is the use of physical
techniques to wash the surfaces of soil particles and to separate soil
particles into clean and contaminated fractions. The goal is normally to
generate clean gravel and sand fractions, while concentrating the contami-
nants into the fines (silt and clay) fraction. Physical separation techniques
based on magnetic, density, or flotation principles can also be utilized to
selectively remove fractions for specific further treatment. We define "soil
leaching" as the use of chemical leaching solutions to dissolve metal
contaminants, remove the metals from the soil constituents, and recover
the dissolved metals from solution.
FROM BENCH-SCALE TO FULL-SCALE
Federal Cartridge Company commissioned a treatability study on soil
washing/leaching in the fall of 1992. COGNIS recognized that most soils
contaminated with metals were going to contain small pieces of metal
which would show up in a sand-sized soil fraction as well as metallic
compounds which would be adsorbed into soil fines. Therefore, in the
TERRAMET® treatment scheme, sand and fines would need to be leached
separately. COGNIS sought a team member who could perform the task
of separating the soil into sand and fines fractions and provide physical
separation of any metal pieces. Bescorp was chosen because of their
REMEDIATION/AUTUMN 1995
63
-------
WILLIAM E. FRJSTAD
Exhibit 1. Leaching-Metal Recovery Concept
TERRAMET® Soil Remediation Systems
Metal-loaded
leachant
Recovered
metals
Regenerated
leachant
Leachant
make-up
experience with density separation of placer gold deposits and battery
debris. When the samples arrived from TCAAP, COGNIS and Bescorp
simultaneously processed the soil and tested combinations of density
separation, leachant, and residence time.
The COGNIS TERRAMET® soil remediation concept uses a leaching-
metal recovery process to leach metals from contaminated soil with one
of several proprietary aqueous acidic leaching solutions. The leachant is
specifically matched with the soil and type of metallic contaminants by
incorporating oxidants, reductants, or complexing agents. By proper
additive control the system can solubilize most chemical forms of lead
contamination including lead sulfate, lead dioxide, and lead metal. The
leaching-metal recovery concept is illustrated in Exhibit 1. Lead-contami-
nated soil is contacted with the leaching solution to dissolve the contami-
nant. The metal-loaded leachant then flows to a patented electrochemical
metal recovery unit where the dissolved lead is removed. The lead-free
leachant is then recycled within the plant for additional leaching. The
approach recycles the leachant solution, and no liquid waste streams are
generated during full-scale operation of the TERRAMET® plant. Leachant
make-up chemicals are added to replenish those which exit the plant in
treated, neutralized, and dewatered-soil. The neutralized leachant residuals
in treated soil have been determined to be safe by the MPCA.
In the treatability study, a composite soil sample from the ammunition
test burn area at TCAAP was studied. At Site F both metallic lead fragments
as well as ionic lead species were found. Because of the high density of
lead relative to most other native soil constituents, the incorporation of
density separation into the physical stages of soil washing was a logical
pretreatment to leaching. Soil •was first attrited, screened, and rinsed to
remove and clean the oversize ordnance and rock. The unclersi/e soil
material was then separated into sand and fines fractions. The sand fraction
was then subjected to density separation using standard mineral process-
ing equipment (jigs, tables, panning) to remove heavy lead and other metal
<54
REMEDIATION/AUTUMN 1995
-------
CASE STUDY: USING SOIL WASHING/LEACHING FOR THE REMOVAL OF HEAVY METAL
Ex
Mbit 2. Density Separation of Lead Particles
.{Mesh) „
+8
+30
+50
+100
+200
-200
Lead Conc^trationbeforeanci a
'Before \
2050
4190
2410
4060
2900
4360
1 After '
996
389
494
842
2960
2490
*" s *
Removed '
51
91
80
79
0
43
;! > * « A «( ** •* ^'» '
fter Density Separatto
Trials
Before ^
NA
1436
1025
901
940
679
TrtalS
NA
168
187
138
416
739
t (ppm)
' v
-VoPb
Removed '
NA
88
82
85
56
0
fragments. The results of density separation lowered the lead concentra-
tion in the sand significantly (Exhibit 2).
Once the larger metal pieces were removed, the remaining fine lead
particles and ionic lead species adsorbed into the soil particles could be
removed cost-effectively by chemical leaching. Exhibits 3 and 4 show the
results of leaching the fines and density pretreated sand fraction. Leaching
the fines was very effective and gave residual lead concentrations of <20
ppm with leachant #2. A number of leachants were tested, and leachant
#2 was eventually chosen for the full-scale remediation. The results on the
sand •were equally satisfactory. A residual lead concentration of <70 ppm
•was also achieved with leachant #2. It was important to show in the
treatability study that lead could be leached from both the fines and the
sand fractions.
After the small bench-scale experiments proved the success of the
multiple leaching concept, additional larger-scale continuous leaching
experiments verified the leaching results. The continuous-scale apparatus
more closely approximated full-scale treatment. It employed an agitated
leaching vessel in which a soil slurry •was continuously leached. The metal
loaded leachant was continuously pumped to the metal recovery unit
where the lead was removed from the leachant and lead recovered as a
solid lead product. The lead-depleted leachant was then returned to the
leaching vessel for continued leaching. After leaching was complete, the
soil-leachant slurry was dewatered and neutralized. Thus, the entire
leaching, clarification, and metal recovery process operated continuously
on the batch of soil in the leaching vessel. Exhibit 5 illustrates typical data
on TCAAP soil. Routinely <100 ppm residual lead and TCLP passage was
REMEDIATION/AUTUMN 1995
65
-------
WmutAM E. FRISTAD
Exhibit 3. Leaching* of TCAAP -200 Mesh Fines
*f
, L '"'. ' <
" i
ii4,!,'j*( *jfi»*
1 Leaciiarjt
2
2
„ y f
CumtilattVe,% Pb Leached
•v / i
teach
1
59
64
2
90
92
ng Contact*
3
96
97
4
97
98
> i
5 '
97
98
Initial**
,m
. (ppm>
575
608
Final***
IPb]-
(ppm)
16
11
* Data are from five consecutive contacts of soil samples with leachant.
** Based on the total Pb detected in leachant plus Pb retained in soil as
determined by nitric acid digestion.
*** Based on EPA acid digestion of treated soil.
Exhibit 4. Leaching* of TCAAP Density-Pretreated -200 Mesh Sand
Leachant
2
2
Cumulative % Pb Leached
Leaching Contact #
1
49
75
2
71
87
3
81
90
4
85
92
5
88
93
Initial**
[Pb] ,
(ppm) -
585
190
Final***
[Pb]
(ppm)
69
14
* Data are from five consecutive contacts of soil samples with leachant.
** Based on the total Pb detected in leachant plus Pb retained in soil as
determined by nitric acid digestion.
**.* Based on EPA acid digestion of treated soil.
observed. The lead concentrations shown under the influent and effluent
columns were the concentrations of lead in the leachant entering and
exiting the metal recovery unit.
A flow chart for the full-scale COGNIS-Bescorp remediation process
which resulted from the treatability study is shown in Exhibit 6. Bescorp
•was responsible for the physical separation steps, and COGNIS was
responsible for the chemical leaching and metal recovery technology. The
soil is fed into the trommel where attrition breaks the soil material down
into its constituent particles of rock, gravel, sand, silt, and clay. The clean
66
REMEDIATION/AUTUMN 1995
-------
CASE STUDY: USING Son, WASHING/LEACHING FOR THE REMOVAL OP HEAVY METAL
Exhibit 5. Continuous-Scale Leaching Experiment TCAAP Soil
(Density Pretreated)
Soil (Avg)
Replicate 1
Replicate 2
Replicate 3
Leachate"' 'I
Sample 1
Sample 2
Sample 3
Sample 4
Sample 5
Sample 6
Sample 7
250 - 350
£ Influent
>Oig/mL)
15.4
10.6
4.6
2.5
1.4
0.9
0.5
31.1
32.6
28.2
33.2
v Effluent
2.5
2.1
1.3
0.7
<0.5
<0.5
<0.5
oversize rock, gravel, and ordnance are rinsed and removed by a 1/4-inch
screen. The oversize is sorted on a conveyer belt, and ordnance is removed
for proper disposal. The clean oversize rock and gravel exits the plant. The
sand and silt/clay fines are separated in a patented vertical separation
chamber which separates the sand from the fines by hydraulic settling
forces. The metal-containing fines are swept into a clarifier where they are
flocced and allowed to settle. The settled fines are pumped through a series
of leaching clarifiers. The leached fines are dewatered and neutralized
before being combined with the clean oversize and sand fractions. The
coarse-sand fraction is run through a mineral jig while the finer sand is run
down a spiral to remove the bulk of the metallic lead and copper particles,
flecks, and dust. The pretreated (metal particle free) sand fraction is then
subjected to counter-current leaching. The leached sand is dewatered and
neutralized before being combined with the fines and oversize. The metal
REMEDIATION/AUTUMN 1995
67
-------
WnxiAM E. FRISTAD
Exhibit 6. COGNIS/Bescorp Full-Scale Process Diagram: Soil Washing/Soil Leaching Process
Feeder
o o o o
I
Oversize
Trommel
Ordnance Removal
O O O O
loaded leachant is fed into the metal recovery units where the dissolved
contaminant metals are electrochemically reduced out of the leachant
solution and recovered in solid metallic form for recycle.
FULL-SCALE RESULTS
Remedial operations at the site began with clearing the site: burning
of the vegetation and removing trees, small buildings, and underground
utilities. All of these materials were decontaminated before removal from
the site. The excavation of the site proceeded in six-inch lifts across the
68
REMEDIATION/AUTUMN 1995
-------
CASE STUDY: USING SOIL WASHING/LEACHING FOR THE REMOVAL OF HEAVY METAL
Feed soil was loaded
into the process
equipment by a
front-end loader.
surface of the site. Much of the contamination was in the top one to two
feet of the three-acre site. Disposal trenches were also located during the
excavation, and 16 such burial trenches down to a depth of 15 feet were
eventually characterized, excavated, and treated. The extent of all excava-
tion was directed by X-ray fluorescence (XRF) spectroscopy, and decisions
to stop excavation were confirmed by laboratory analysis. Use of XRF in
defining excavation limits helped to contain soil treatment costs by
minimizing unnecessary over-excavation.
COGNIS and Bescorp built an integrated mobile trailer-mounted field-
scale soil treatment plant and drove it on six flatbed trailers to the TCAAP
site. All soil processing was conducted on a preexisting concrete pad
located approximately 100 yards from the excavation at Site F. The
treatment pad provided secondary containment for a small feed soil
stockpile, the processing equipment, and the treated soil stockpile. During
all operations, rainwater which collected on the treatment pad was
collected and either used as process water or treated before disposal to the
sanitary sewer system.
Feed soil was loaded into the process equipment by a front-end loader.
After-treatment soil was sampled approximately every 10 tons and samples
composited into 30- or 60-ton batches for laboratory analysis before
loading into storage bunkers. Samples were analyzed for all eight project
metals (EPA 3050 digestion for total metals content). Analytical results were
returned within 24 hours, clean soil was returned to Site F, and soil which
failed to meet cleanup criteria was reprocessed.
The first week of soil processing constituted an acceptance period for the
MPCA. During this period all the output of the treatment plant was carefully
monitored to ensure that all the regulatory goals were being achieved and that
the plant operated in a safe manner. The analytical data for the acceptance
period runs are summarized in Exhibit 7. Each run represented 60 tons of
processed soil. The treatment plant was operated for several weeks in the fall
of 1993 before shutting down for the winter. In the summer of 1994 treatment
was resumed, and Exhibit 8 summarizes the treatment results for both 1993
and 1994 at which point over 12,000 tons of soil had been treated. By the end
of the job in June 1995, over 20,000 tons of soil had been treated. Final 1995
data were not available at the time of this writing.
As a result of the soil treatment, over 20 tons of lead were recovered
and recycled through the local secondary lead smelter where the lead had
originally been purchased by TCAAP to manufacture the ordnance.
Ordnance removed from the oversize fraction was also separated into live
ordnance, spent ordnance, and inert projectiles. Approximately 250 tons
of ordnance materials were recovered for demilitarization.
An important part of the COGNIS/Bescorp process is that no process
water is discharged during the operation; all process water/leachant is
recycled within the plant. Process water at the end of each treatment season
was neutralized, clarified, and analyzed for all the project metals, and after
meeting the Metropolitan Waste Water Commission's criteria, it was
discharged to the sanitary sewer system. The sanitary sewer discharge
criteria were met each time.
REMEDIATION/AUTUMN 1995
69
-------
WILLIAM E. FRISTAD
Exhibit 7. TCAAP Acceptance Period Results
I (, *r
^
1 h I fw,f f 4 ^*
1 1*
" l""' Metal !J'
' it.
Antimony
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
( JA v A
RCRA
Cleanup
Goat
4.0
4.0
100
80
175
0.3
45
5.0
(i PJ * ^r" *
Acceptance Period Run Number (ppm)
l1 "* * ! A
1
<1
0.8
3.5
23.7
19
<0.02
8.9
<0.1
,2" '
<1
1.8
0.6
12.6
60
0.04
6.2
<0.1
' 3 ,
<1
<0.02
2.6
9.8
30
0.03
4.6
<0.1
4'
<1
3.0
6
12.7
60
<0.02
5.4
<0.1
5 V
<1
0.2
5
16.1
60
<0.02
6
<0.1
Exhibit 8. TCAAP 1993-1994 Results
Metals
of
Interest
Antimony
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
RCRA
Cleanup
~ Goat
4
4
100
80
175
0.3
45
5
Avg. Residual Metal in
Treated Soil (ppm)
1993
<0.5
0.8
6
39
173
0.1
9
<0.05
1994 '
0.9
0.2
6
65
90
0.3
7
<0.1
70
REMEDIATION/AUTUMN 1995
-------
Facilities should
make every effort to
debrief employees on
disposal practices
during their
employment.
CASE STUDY: USING Son, WASHING/LEACHING FOR THE REMOVAL OF HEAVY METAL
Minimization of hazardous waste was the primary goal at the outset of
this project. By the end of the 1994 treatment season, no soil or process
material left the TCAAP facility as hazardous waste for landfill. All
contaminant metals were recycled through a secondary lead smelter, all
process water was discharged to the sanitary sewer, and all soil remained
on the TCAAP facility.
At the close of the project, the process equipment was decontaminated,
and the treated soil was backfilled over the site, covered with topsoil, and
revegetated with prairie grasses.
LESSONS LEARNED
As COGNIS looks at its next opportunities—and as generators of metal-
contaminated soil ready themselves for soil treatment—a few lessons can
be shared from TCAAP.
Know the Site
Site characterization is difficult and expensive, particularly when the
site involves disposal areas. The best way to characterize a site is to
excavate the soil before treatment and final treatment pricing. At that point
the true nature of the site will be apparent, and both the client and the
technology vendor will better know what lies ahead. A considerable
amount of time and effort was spent characterizing Site F, and despite
millions of dollars spent, 15 of the 16 burial trenches were only identified
during the excavation/treatment. Considerable client money could have
been saved by taking fewer soil borings and excavating the site before
committing to final remedial contract terms.
Performing under a Fixed-Price Contract
Fixed-price contracts are dangerous when the problem is difficult to
define. By their very nature, disposal areas will tend to include materials
which nobody wanted to see again. The unearthing of these surprises can
lead to undesired change orders, require on-the-fly plant modifications,
and possibly lead to unfulfilled goals. Changed conditions should be
anticipated in the contracting mechanism. Productivity can remain high if
the contract is structured so that remediation can continue while details of
changed conditions are worked out.
Chemical Company versus Remedial Contractor Perspective
COGNIS is a company made up of people who have been commer-
cializing chemical technologies for over 25 years. This background helped
COGNIS continuously improve the process throughout the job duration,
so that the level of metal removal increased steadily over the course of the
project. Chemical companies measure their success in terms of productiv-
ity, not billability.
Don't Lose Valuable Historical Information
Facilities should make every effort to debrief employees on disposal
practices during their employment. Thirty years later, a considerable
REMEDIATION/AUTUMN 1995
71
-------
WILLIAM E. FRISTAD
amount of detective work must be done to try to second-guess likely
actions. Somehow, the employer should provide immunity from prosecu-
tion to encourage the full disclosure of information. Once that source of
history is gone, excavation efforts will be inefficient and processing costs
will be higher than necessary because of unexpected surprises in the feed
soil.
The COGNIS/Bescorp soil washing/TERRAMET® soil leaching process
was an example of the client, the regulator, and the technology vendor all
aiming for a common goal—completing metal-contaminated soil treatment
with no hazardous waste generation for off-site disposal. Under these
conditions, innovative technologies can succeed on their first full-scale
demonstration. The TERRAMET® process was shown to be uniquely
applicable on projects where the land will have high reuse possibilities or
where human exposure pathways are high, the treatment goals are very
strict, and a permanent solution with minimal liability is desired. •
NOTES
1. "Cleaning Up the Nation's Waste Sites: Markets and Technology Trends," U.S. EPA,
Office of Solid Waste and Emergency Responses, EPA Doc. No. EPA/542R/R-92/012,
Washington, DC, 1993.
2. Dumont, E., EP 291,746 (1988).
3. Deconta A.G., EP 72,885 (1982); Wallbank, P., Barckman, C, Piske, G., DE 3,703,922
(1987); Siemans A.G., EP 278,328 (1988); Weher, M., EP 402,737 (1990).
4. Raghavan, R., Coles, E., Dietz, D. "Cleaning Excavated Soil Using Extraction Agents: A
State-of-the-Art Review," U.S. EPA, Res. Dev. [Rep.] EPA-600/2-89/034, 1988.
5, Basu, T.K., Selvakumar, A., Gaire, R., Royer, M.D. "Selection of Control Technologies
for Remediation of Lead Battery Recycling Sites," Report by Foster Wheeler Envirosponse,
Inc., Contract No. 68-C9-0033, RREL, EPA, Cincinnati, Ohio; Rayford, R., Evangelista, R.,
Unger, R., "Lead Extraction Process," Draft Report to EPA, Contract No. 68-03-8255, U.S.
EPA, Emergency Response Branch, Edison, New Jersey, 1986; U.S. EPA, "Engineering
Treatment Forum Technical Assistance Report: Lead Battery Reclamation Sites," EPA/600/
X-88/242, Haz. Waste Res. Eng. Lab., Cincinnati, Ohio, 1988.
6. Earth, E.F., Traver, R.P., "Treatment of Lead Battery Contaminated Soil Utilizing Soil
Washing and Solidification/Stabilization Technology," Proc. 1990 EPA/A&WMA Int. Symp.,
Haz. Waste Treatment: Treatment of Contaminated Soils, Cincinnati, Ohio, 1990.
7. Schmidt, B.W., "Assessment of Treatment Techniques at Superfund Battery Sites," Int.
Symp. Haz. Waste Treatment: Treatment of Contaminated Soils, Cincinnati, Ohio, 1989.
8. Id.
72
REMEDIATION/AUTUMN 1995
-------
Section 6
References
Brinkmann Instruments. 1996. Telephone Conversation
Regarding Retsch/BrinkmannZM-1 Grinder. Between
Joseph Plurad and Pinaki Banerjee, PRC Environmen-
tal Management, Inc. (PRC). March 18.
COGNIS, Inc. (COGNIS). 1996. Letter from William E.
Fristad, COGNIS to Pinaki Banerjee, PRC. Dated April
5.
Evans, G.E. 1990. "Estimating Innovative Technology
Costs for the Superfund Innovative Technology Evalu-
ation (SITE) Program." Journal of Air Waste Manage-
ment Association. Volume 40, No. 7. July.
Federal Cartridge Company. 1994. David Fuller. Fac-
simile Regarding Electrical Usage by Soil-Treatment
Activities at the Twin Cities Army Ammunition Plant
(TCAAP). Sent to Jeff Swano, PRC. December 13.
General Testing Corporation. 1993. Pricing Policy. Au-
gust 1.
Gopher Smelting. 1995. Telephone Conversation Regard-
ing TCAAP Lead Disposal Costs. Between John Tap-
per, Plant Manager, and Jeff Swano, PRC. April 18.
R.S. Means Company, Inc. 1994. Means Heavy Construc-
tion Cost Data. 8th Annual Edition.
U.S. Environmental Protection Agency (EPA). 1988.
CERCLA Compliance with Other Laws Manual: Interim
Final. EPA/540/G-89/006. August.
EPA. 199la. Test Methods for Evaluating Solid Waste,
Volumes IA-IC: Laboratory Manual, Physical/Chemi-
cal Methods; and Volume II: Field Manual, Physical/
Chemical Methods SW-846, Third Edition. Office of
Solid Waste. EPA Document Control No. 955-001-
00000-1. With Subsequent Technical Additions. Revi-
sion 1. July.
EPA. 1991b. Technical Support Documents on Lead.
ECAO-CIN-757. January.
EPA. 1994a. "Applications Analysis Report of the
BESCORP Soil Washing System for Lead Battery Site
Treatment." November.
EPA. 1994b. "Guidance Manual for the Integrated Expo-
sure Uptake Biokinetic Model for Lead in Children."
EPA/540/R-93/081. February.
EPA. 1994c. Revised Interim Soil Lead Guidance for
CERCLA Sites and RCRA Corrective Action Facilities.
OSWER Directive Number 9355.4-12. July.
Wenck Associates, Inc. (Wenck). 1995a. TCAAP Site F
Soil Washing/Soil Leaching Technology Evaluation
Draft Report. September.
Wenck. 1995b. Personal Conversations Between Bill
Johnsen and Matt Bower, Wenck, and Pinaki Banerjee,
PRC.
Wenck. 1995c. Letter Regarding TCAAP Site F Data
Request. From William P. Johnsen, CPG. To Pinaki
Banerjee, PRC. April 7.
73
-------
-------
-------
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand corner.
If you do not wish to receive these reports CHECK HERE D;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use
$300
EPA/540/R-96/535
------- |