&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

-------