EPA/540/AR-94/504
September 1994
Eco Logic International Gas-Phase Chemical Reduction
Process-The Thermal Desorption Unit
Applications Analysis Report
Risk Reduction Engineering 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 the auspices of the Superfund Innovative Technology
Evaluation (SITE) Program under Contract No. 68-C9-0033 to Foster Wheeler Enviresponse,
Inc. (FWEI). It has been subjected to the Agency's peer and administrative review, and it
has been approved for publication as an EPA document. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
-------
Foreword
The SITE Program was authorized in the 1986 Superfund Amendments and Reautho-
rization Act (SARA). The program is administered by EPA's Office of Research and
Development (ORD). The purpose of the program is to accelerate the development and use
of innovative cleanup technologies applicable to Superfund and other hazardous waste
sites. This is accomplished through technology demonstrations designed to provide perfor-
mance and cost data on selected technologies.
The SITE Program funded a field demonstration to evaluate the ECO LOGIC
Gas-Phase Chemical Reduction Process, developed by ELI Eco Logic International, Inc.,
Ontario, Canada. The ECO LOGIC Demonstration took place at the Middleground
Landfill in Bay City, Michigan, using landfill waste; it assessed the technology's ability to
treat hazardous wastes, based on performance and cost. Three reports contain the results of
the Demonstration: a Technology Evaluation Report (TER), which describes the field
activities and laboratory results; this Applications Analysis Report (AAR), which inter-
prets the data and discusses the applicability of the technology to a soil feedstock; and a
second, independent AAR, which interprets the data and discusses the applicability of the
reactor system to liquid feedstocks.
A limited number of copies of this report will be available at no charge from EPA's
Center for Environmental Research Information, 26 West Martin Luther King Drive,
Cincinnati, OH 45268 (513-569-7562). Requests should include the EPA document
number found on the report's front cover. When this supply is exhausted, additional copies
can be purchased from the National Technical Information Service, Ravensworth Bldg.,
Springfield, VA 22161, 703-487-4600. Reference copies will be available at EPA libraries
in their Hazardous Waste Collection. To inquire about the availability of other reports, call
the EPA Clearinghouse Hotline at 1-800-424-9346 or 202-382-3000 in Washington, DC.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
-------
Abstract
This report evaluates the capability of the ECO LOGIC Gas-Phase Chemical Reduc-
tion Process to detoxify organics in a solid matrix. The report presents data from the recent
EPA SITE Demonstration, which tested the ECO LOGIC Thermal Desorption Unit
(TDU), and evaluates the costs of operating the unit.
The ECO LOGIC Process thermally separates organics, then chemically reduces them
in a hydrogen atmosphere, converting them to a reformed gas that consists of light
hydrocarbons and water. ECO LOGIC designed the TDU to remove organic and metallic
contaminants from soil, sending the desorbed organics to the reactor system for further
treatment. The TDU produces two principal residual streams: treated soil and quench
water.
The SITE Program evaluated the ECO LOGIC Process at the Middleground Landfill
in Bay City, Michigan. There, the TDU/reactor system processed 1.1 tons of soils,
contaminated principally with polychlorinated biphenyls(PCBs). The test did not demon-
strate successful treatment of contaminated soils. Rather, it provided proof-of-concept data
to evaluate the system's strengths and weaknesses; it focused attention on the areas that
require additional engineering evaluation or change.
-------
Contents
Foreword iii
Abstract iv
Figures vii
Tables Vii i
Abbreviations ix
SI Conversion Factors xii
Acknowledgments xiii
1. Executive Summary 1
Introduction 1
The SITE Demonstration 1
Conclusions 1
Waste Applicability , 2
Costs 2
2. Introduction 4
The SITE Program 4
SITE Program Reports 4
Key Contacts 5
3. Technology Applications Analysis 6
Process Description 6
Test Conditions 8
Conclusions 9
Technology Evaluation 10
Organics Destruction 10
Air Emissions 11
Intermediate and Residual Stream Characterization 11
Equipment and Operating Considerations 16
Technology Applicability 16
Site Characteristics 16
Applicable Media 16
Safety Considerations 17
Staffing Issues 17
Regulatory Considerations 17
Clean Air Act 17
Clean Water Act 18
Comprehensive Environmental Response, Compensation, and Liability Act 18
Occupational Safety and Health Act 18
Resource Conservation and Recovery Act 18
-------
Contents (Continued)
Toxic Substances Control Act 18
State and Local Regulations 18
References 18
4. Economic Analysis 20
Introduction 20
Conclusions 20
Issues and Assumptions 20
Site-Specific Factors 20
Costs Excluded from the Estimate 21
Basis for Economic Analysis 21
Results of Economic Analysis 21
References 22
Appendices
A. Demonstration Sampling and Analysis 23
B. Vendor's Claims 27
Vi
-------
Figures
1. Gas-phase chemical reduction reactions
2. Reactor and TDU system schematic diagram
3. The ECO LOGIC Reactor
A-l. Sampling and monitoring stations 23
B-l. ECO LOGIC Process reactions 28
B-2. Commercial-scale process reactor 29
B-3. Commercial-scale process unit schematic 30
VII
-------
Tables
1. Summary Results of TDU Testing 3
2. MDNR Air Permit Conditions 12
3. Mass Distribution of Selected Streams 12
4. Component Partitioning 13
5. Reformed Gas Comparison to Other Fuels 14
6. Summary of TDU Operating Conditions 16
7. Economic Analysis forth ECO LOGIC TDU System 22
A-l. EPA Sample Locations 24
A-2. ECO LOGIC Process Control Monitoring Stations 24
A-3. Flue Gas Sampling and Analytical Methods 25
A-4. Solids Sampling and Analytical Methods 26
A-5. Liquids Sampling and Analytical Methods 26
B-l. Hamilton Harbor Performance Test Results 31
B-2. U.S. EPA SITE Program Results 32
B-3. Summary of Test Results from the Lab-Scale Thermal Desorption Mill 33
VIII
-------
Abbreviations
AAR Applications Analysis Report
AAS atomic absorption spectroscopy
ARARs applicable or relevant and appropriate requirements
ASTM American Society for Testing and Materials
CAA Clean Air Act
CB chlorobenzene
CEM continuous emission monitoring
CEMS continuous emissions monitoring system
CERCLA Comprehensive Environmental Response, Compensation, and Liability Act
CFR Code of Federal Regulations
CIMS Chemical lonization Mass Spectrometer
C O carbon monoxide
CO2 carbon dioxide
CP chlorophenols analysis
Cl chloride
C12 chlorine
CVAAS cold vapor atomic absorption spectroscopy
CWA Clean Water Act
DE destruction efficiency
DOT U.S. Department of Transportation
DRE destruction and removal efficiency
dscf dry standard cubic foot
dscm dry standard cubic meter
EER Energy and Environmental Research Corp.
ECO LOGIC ELI Eco Logic International, Inc.
EPA U.S. Environmental Protection Agency
FID flame ionization detection
ft feet
FPD flame photometric detector
FWEI Foster Wheeler Enviresponse, Inc.
g gram
IX
-------
Abbreviations (Continued)
GC
GF
gr
gpm
H2
HCB
HCI
hr
HR
ICAP
in.
kg
kW
L
Ib
MASA
MDNR
m
mg
min
mo
MS
NAAQS
NDIR
NDUV
ng
NOx
NPDES
o2
ORD
OSWER
OSHA
PAHS
PCBs
PCDD
PCDF
PCE
PH
PICs
gas chromatography
graphite furnace
grains
gallons per minute
hydrogen
hexachlorobenzene
hydrogen chloride
hour
high resolution
inductively coupled argon plasma spectroscopy
inch
kilogram
kilowatt
liter
pound
Method of Air Sampling and Analysis
Michigan Department of Natural Resources
meter
milligram
minute
month
mass spectroscopy
National Ambient Air Quality Standards
nondispersive infrared
nondispersive ultraviolet
nanogram
nitrogen oxides
National Pollutant Discharge Elimination System
oxygen
Office of Research and Development
Office of Solid Waste and Emergency Response
Occupational Safety and Health Act
polycyclic aromatic hydrocarbons
polychlorinated biphenyls
polychlorinated dibenzo-p-dioxin
polychlorinated dibenzofuran
perchloroethylene
a measure of acidity/alkalinity
products of incomplete combustion
-------
Abbreviations (Continued)
PIR product of incomplete reduction
POHC principal organic hazardous constituent
BOIW publicly owned treatment works
ppb parts per billion
PPE personal protective equipment
ppm parts per million
ppmv Parts per million by volume
psig pounds per square inch gauge
QA quality assurance
Q| quality indicator
RCRA Resource Conservation and Recovery Act
RREL Risk Reduction Engineering Laboratory
SARA Superfund Amendments and Reauthorization Act
scf standard cubic feet
scfm standard cubic feet per minute
sec second
SITE Super-fund Innovative Technology Evaluation Program
SQ2 sulfur dioxide
SVOCs semivolatile organic compounds
TCLP Toxicity Characteristic Leaching Procedure
TDU thermal desorption unit
TER Technology Evaluation Report
THC total hydrocarbon
TSCA Toxic Substances Control Act
TSD treatment, storage, and disposal
VOCs volatile organic compounds
VOST volatile organic sampling train
fig microgram
-------
SI Conversion Factors
Multiply
Area:
Flow rate:
Length:
Mass;
Volume:
Temperature:
Concentration:
Pressure:
Heating value:
English (US)
Units by
1ft2
1m2
1 gal/min
1 gal/min
1 MOD
1ft
lin.
lib
lib
1ft3
1ft3
Igal
Igal
°F-32
1 gr/ft}
1 gr/gal
1 lb/ft3
lib/in.2
1 lb/in.2
Btu/lb
Btu/scf
Factor to get
0.0929
6.452
6.31xlO-5
0.063 1
43.81
0.3048
2.54
453.59
0.45359
28.316
0.0283 17
3.785
0.003785
0.55556
2.2884
0.0171
16.03
0.0703 1
6894.8
2326
37260
Metric (SI)
Units
m2
cm2
m3/s
L / S
L / S
m
cm
g
kg
L
m3
L
m3
o
C
g/m3
g/L
g/L
kg/cm2
Newton/m
Joules/kg
Joules/scm
XII
-------
Acknowledgments
Under EPA Contract 68-C9-0033, FWEI prepared this report for EPA's SITE Pro-
gram with the supervision and guidance of Gordon M. Evans, EPA SITE Program
Manager in the Risk Reduction Engineering Laboratory (RREL), Cincinnati, Ohio. The
FWEI Project Manager was Gerard W. Sudell; James P. Stumbar, Ph.D., provided the
technical review; Marilyn K. Avery edited the document.
Energy and Environmental Research Corporation (EER) provided sampling and
analytical support to EPA. Contracted to perform data analysis, data reduction, and
analytical review, EER provided the scientific data that form the basis for this report.
Kelvin Campbell and Craig McEwen of ELI Eco Logic International provided
continued assistance throughout the project, as did Edward Golson of the City of Bay City.
Sue Kaelber-Mattock supported the project for the Michigan Department of Natural
Resources.
XIII
-------
Section 1
Executive Summary
Introduction
This report summarizes the findings of the SITE Demonstra-
tion of the Gas-Phase Chemical Reduction Process and TDU
developed by ELI Eco Logic, International, Inc. (ECO LOGIC)
of Ontario, Canada.
Under the auspices of the SITE Program, and in cooperation
with the City of Bay City, Michigan; Environment Canada;
and Ontario Ministry of the Environment and Energy; EPA
conducted the demonstration of the ECO LOGIC Process at
Bay City's Middleground Landfill. The landfill accepted mu-
nicipal and industrial wastes for approximately 40 years. A
1991 remedial investigation indicated elevated levels in ground-
water of trichloroethene, PCBs, 1 ,2-dichloroethene methyl-.
ene chloride, toluene, and ethylbenzene. The groundwafer
contained lesser concentrations of benzidine, benzene, vinyl
chloride, chlorobenzenes, polycyclic aromatic hydrocarbons
(PAHs), lindane, dieldrin, chlordane, and DDT metabolites.
The patented ECO LOGIC Gas-Phase Chemical Reduction
Process treats organic hazardous waste in soil and liquid
media. During processing, the soil waste feed enters a TDU
designed to desorb and evaporate organics from solids. Treated
solids are cooled and stored for proper disposal. The evapo-
rated organics pass to the reactor where they are treated with
PCB-contaminated liquids and water in the gas-phase chemi-
cal reduction reactor to produce reformed gas.
The reaction products include hydrogen chloride from the
reduction of chlorinated organics, such as PCBs, and lighter
hydrocarbons, such as methane and ethylene, from the reduc-
tion of straight-chain and aromatic hydrocarbons. The absence
of free oxygen in the reactor inhibits dioxin formation. Water
acts as a hydrogen donor to enhance the reaction.
A scrubber treats the reformed gas to remove hydrogen chlo-
ride and particulats Of this gas, a portion recycles back into
the reactor; the rest is either stored or feeds a propane-fired
boiler prior to release to the atmosphere. The recycle stream
may be used as a fuel in other system support equipment, such
as the boiler that generates steam. The final combustion step in
the boiler met Resource Conservation and Recovery Act
(RCRA) requirements, making the reformed gas environmen-
tally acceptable for combustion.
The SITE Demonstration
The two-part demonstration took place in October and De-
cember 1992,using PCB-contaminated oil, water, and soil
extracted directly from the landfill. ECO LOGIC first per-
formed a series of shakedown tests to establish optimum
system performance. Two liquid tests investigated reactor
performance (Conditions 1 and 3); a soil test (Condition 2)
studied the complementary TDU. The TDU test consisted of
two runs, processing a total of 1.1 tons of soil contaminated
with 627 ppm of PCB. This report provides only the results
for the soil test; a second, independent AAR has published the
results of the two liquid tests.
EPA collected extensive samples around the major system
components and stored or logged important data on operating
and utility usage. Laboratory analyses provided information
on the principal process streams: desorbed soil, reactor grit,
scrubber residuals, reformed gas, and boiler stack emissions.
EPA evaluated these data against established program objec-
tives to determine the capability of the process to treat the
designated waste.
Conclusions
Based on the program objectives, the demonstration con-
firmed the feasibility of the gas-phase chemical reduction
process for treating PCBs and other chlorinated organic com-
pounds, producing a fuel gas from PCB-contaminated soil,
and providing environmentally acceptable air emissions.
The TDU did not perform to design specifications. EPA
categorized the TDU test data as a system proof-of-concept
rather than as a comprehensive evaluation of a fully devel-
oped unit. The test data indicated that the TDU, as presently
configured, achieved desorption efficiencies at the expense of
throughput. In addition, ECO LOGIC experienced material
handling problems with the TDU feed. The combination of
material handling problems and inadequate organics desorp-
tion showed a need for further development. The test data
have identified system strengths and targeted areas that re-
quire improvement.
Nevertheless, the demonstration did show that ECO LOGIC'S
TDU can desorb PCB contaminants. Subsequent treatment of
the desorbed gas in the reactor produced stack emissions that
generally met stringent regulatory levels. The reformed gas
-------
composition resembled coal-gas fuel. The scrubber liquor
required either disposal as a RCRA waste or recycling through
the system for additional treatment. Table 1 correlates the
program conclusions with the program objectives.
Waste Applicability
The SITE Program concluded that the ECO LOGIC Process,
from the reactor to the stack (independent of the TDU),
efficiently treated liquid wastes containing oily PCBs, other
organics, and water containing PCBs, other organics, and
metals. Stack emissions met stringent regulatory levels. An
independent AAR presents the results of the reactor tests.
The reactor did not directly process soil. Instead, ECO LOGIC
provided a complementary front-end TDU to treat soils. The
principal TDU residual streams were the quench water and the
treated soil. PCB concentrations ranged from 8.26 to 29.2
ppm in the treated soil, resulting from inadequate desorption
of semivolatile organic compounds (SVOCs). Based on this
result and others discussed in Section 3, the TDU requires
further development to successfully process contaminated
soils.
Costs
The twelve categories established for the SITE Program formed
the basis for the cost analysis. Costs relate to the TDU, as
operated at the Middleground Landfill: Based on the eco-
nomic analysis, the estimated cost (1994 U.S. dollars) for
treating solid wastes similar to those at the Bay City site range
from $630/ ton (at a 60% utilization factor) t o $500/ton (80%
utilization factor). Important elements affecting costs are fuel
(67%). equipment (11%). and labor (9%). The rest of the cost
components comprised 13%.
-------
Table 1.
Summary Results of TDU Testing
Validate key cost assumptions
Characterize effluents and residuals
Determine suitability of reformed gases
for reuse/resale
Demonstrate system reliability
Develop mass balances
Characterize scale-up parameters
Validate CIMS
Document system operating conditions
Results
Objective
Demonstrate ORE for PCBs: 99.9999%
Demonstrate DE for HCB: 99.99%
Ensure no formation of PCDD/PCDF
Characterize PIC emissions
Characterize HCI emissions
Document MDNR air permit compliance
Characterize criteria air pollutants
Not Range
Met met achieved
X 99.9999%
X 72.13(099.99%
X PCDD DE:
42.5% to 99.45%
PCDF DE:
54.6% to 96.1 2%
X
X 0.66 mg/dscm;
1 50 mg/hr;
99.96% removal
X
X
Comments
Requirements met.
Inefficient desorption
from soil in Run 1 .
No net PCDD/PCDF
formation..
Emissions characterized.
HCI emissions
acceptable.
Air permit compliance
documented.
Easily met permit
conditions.
X
X
Throughput reliability:
4 to 21.2% of
design.
System availability:
24%
Cost elements identified.
Organics destroyed;
metals partitioned to
quench water and
scrubber water; after
further treatment,
scrubber liquor may be
suitable for POTW.
Closely matched
composition of other
commercial fuel gases.
Process reliability
requires improvement.
Generally good closures,
except for certain
metals.
Characterized
May reflect data trends
useful for process
control.
Data available for
commercial scaleup.
-------
Section 2
Introduction
The SITE Program
In 1986 EPA's Office of Solid Waste and Emergency Re-
sponse (OSWER) and the Office of Research and Develop-
ment (ORD) established the SITE Program to promote the
development and use of innovative technologies to clean up
Superfund sites across the country. Now in its eighth year,
SITE is helping to commercialize the treatment technologies
necessary to meet new federal and state cleanup standards
aimed at permanent remedies, rather than short-term correc-
tions. The SITEProgram includes four major elements: the
Demonstration Program, the Emerging Technologies Pro-
gram, the Measurement and Monitoring Technologies Pro-
gram, and the Technology Transfer Program.
The major focus has been on the Demonstration Program,
designed to provide engineering and cost data on selected
technologies. EPA and the technology developers that partici-
pate in the program share the cost of the demonstration.
Developers are responsible for demonstrating their innovative
systems, usually at Superfund sites selected by EPA. EPA is
responsible for sampling, analyzing, and evaluating test re-
sults. The outcome is an assessment of the technology's
performance, reliability, and cost. This information, used in
conjunction with other data, enables EPA and state decision
makers to select the most appropriate technologies for Super-
fund cleanups.
Innovative technology developers apply to participate in the
Demonstration Program by responding to EPA's annual so-
licitation. EPA will consider a proposal at any time from a
developer who has scheduled a treatment project on Super-
fund waste. To qualify for the program, a new technology
must have a pilot- or full-scale unit and offer some advantage
over existing technologies. Mobile technologies are particu-
larly interesting.
Once a proposal has been accepted, EPA and the developer
work with the EPA regional offices and state agencies to
identify a site containing wastes suitable for testing the tech-
nology. EPA prepares a detailed sampling and analysis plan
designed to thoroughly evaluate the technology by providing
analysts with reliable data. A demonstration may last any-
where from a few days to several months, depending on the
process and the quantity of waste needed to assess the tech-
nology. Ultimately, the Demonstration Program rates the
technology's overall applicability to Superfund problems.
The second major element of the SITE Program is the Emerg-
ing Technologies Program, which fosters the further investi-
gation and development of treatment technologies that are still
at laboratory scale. Successful validation of these technologies
could lead to the development of systems viable for field
demonstration. A third component of the SITE Program, the
Measurement and Monitoring Technologies Program, pro-
vides assistance in the development and demonstration of
innovative technologies that will better characterize Super-
fund sites. The Technology Transfer component ensures effec-
tive distribution of the results of the demonstration projects.
SITE Program Reports
Two documents incorporate the results of each SITE Demon-
stration: the TER and the AAR. The TER contains a compre-
hensive description of the demonstration and its results. This
report assists engineers who are performing a detailed evalua-
tion of the technology for a specific site and waste. The
technical evaluations provide a detailed understanding of the
technology performance during the demonstration and assess
the advantages, risks, and costs for a given application.
The AAR estimates Superfund applications and technology
costs, based on available data. It compiles design and test data,
summarizes them, explores other laboratory and field applica-
tions, and discusses the advantages, disadvantages, and limita-
tions of the technology. The AAR attempts to synthesize
available information and draw reasonable conclusions for the
technology's use. The report discusses factors such as site and
waste characteristics that have a major effect on costs and
performance. Pilot- and full-scale operations data provide the
bases for estimating technology costs for different applica-
tions.
The amount of available data needed to evaluate an innovative
technology varies widely. Data may be limited to laboratory
tests on synthetic waste or may extend to performance data on
actual wastes treated in the field at the pilot or full scale. In
addition, conclusions regarding Superfund applications drawn
from a single field demonstration have limitations. A success-
ful field demonstration does not necessarily ensure that a
-------
technology will become widely applicable or attain full de-
velopment at the commercial scale. The AAR can assist
remedial managers in planning Superfund cleanups; it repre-
sents an important tool in the development and commercial-
ization of the technology.
Key Contacts
The sources listed below can provide additional information
concerning the SITE Demonstration, the site, or the ECO
LOGIC Gas-Phase Chemical Reduction Process.
Gordon M. Evans
SITE Project Manager
U.S. Environmental Protection Agency
26 W Martin Luther King Drive (MS-215)
Cincinnati, OH 45268
Phone: 513-569-7684
Fax: 513-569-7620
James Nash
ELI Eco Logic International, Inc.
143 Dennis St.
Rockwood, Ontario NOB 2KO
Canada
Phone: 519-856-9591
Fax: 519-856-9235
-------
Section 3
Technology Applications Analysis
This AAR assesses the capability of the ECO LOGIC TDU to
treat soils contaminated with PCBs and other hazardous sub-
stances. EPA has based the assessment on the results of the
SITE Demonstration and on data supplied by the technology
developer. The report contains a summary of relevant material
from the more detailed TER. Since the results of the demon-
stration, provided in the TER, are of known quality, the report
bases its conclusions on them.
Appendix A describes the demonstration sampling and ana-
lytical locations/methods; Appendix B, ECO LOGIC'S claims
for the technology.
Process Description
The patented ECO LOGIC Gas-Phase Chemical Reduction
Process (Reactor System) treats organic hazardous waste in a
hydrogen-rich atmosphere at approximately 900°C (1,650°F)
and ambient pressure, producing a reformed gas. Water acts
as a hydrogen donor to enhance the reaction. The reaction
products include hydrogen chloride, from the reduction of
chlorinated organics, such as PCBs, and lighter hydrocarbons,
such as methane and ethylene from the reduction of straight-
chain and aromatic hydrocarbons. A scrubber treats the re-
formed gas to remove hydrogen chloride and particulates. Of
this gas, a portion recycles back into the reactor; the rest is
either compressed for storage or feeds a propane-fired boiler
prior to release to the atmosphere. The absence of free oxygen
in the reactor inhibits dioxin formation.
Figure 1 shows some of the reactions that lead to the major
intermediate and final products. Through hydrogenation, the
first five reactions remove chlorine from PCBs and reduce the
higher molecular weight hydrocarbons to simpler, more satu-
rated compounds. The final reaction regenerates hydrogen.
Figure 2 illustrates the process in a schematic diagram of the
field demonstration unit. The demonstration-scale reactor (Fig-
ure 3) is mounted on a drop-deck trailer. The trailer carries a
scrubber system, a recirculation gas system, and an electrical
control center. A second trailer holds a propane boiler, a waste
preheating vessel, and a waste storage tank.
ECO LOGIC designed the TDU/reactor process to treat 4
tons/day of waste oil, 10 tons/day of wastewater, and 25 tons/
day of soil, depending on the nature of the contaminants, their
degree of chlorination, and their water content.
The TDU processes soil by desorbing organics at 500-600°C
(930-1,1 10°F) into a hydrogen-rich carrier gas and by dissolv-
ing volatile metals in a molten metal bath. Some of each
volatile metal will pass to the reactor with the carrier gas, an
additional portion will be dissolved into the bath, and the
remainder will stay in the treated soil. Nonvolatile metals
remain in the treated soil; quench water cools the treated soil
prior to disposal. The hydrogen-rich carrier gas conveys the
desorbed organics to the reactor (Figure 3). where they are
treated in a gas-phase chemical reduction reaction to produce
reformed gas.
Hydrogen and molten tin are used because the two elements
do not react, unlike other combinations such as tin and oxy-
gen. Tin offers favorable properties: high thermal conductiv-
ity, high density, low vapor pressure, and high surface ten-
sion. Tin's high thermal conductivity makes it a good heat
transfer fluid, efficiently raising the soil temperature. Its high
density causes the soil to float on the bath's surface, prevent-
ing its mixing with the tin. Low vapor pressure prevents
evaporation of the tin. Because it has a high surface tension
and is nonwetting, molten tin does not soak into the pores
between soil particles. Hence, the treated soil is easily sepa-
rated. In addition, molten tin is a good solvent for heavy
metals such as lead, arsenic, and cadmium. If these metals are
present in the elemental state, they dissolve into the bath. If
they are present as oxides or as other compounds, the hot
hydrogen atmosphere can convert them to an elemental state.
During the tests, a hopper with a screw feeder dropped waste
soil onto the tin bath. The screw feeder provided a gas seal
between the outside air and the hydrogen atmosphere inside
the TDU. The auger's variable speed drive controlled the feed
rate. Once inside the TDU, the soil floated on top of the
molten tin and heated to 600°C (1,110°F) Organic materials
vaporized from the soil and flowed to the reactor with the
hydrogen carrier gas. A paddlewheel removed treated soil
from the end of the tin bath and fed it into a quench tank. A
drag conveyor removed the soil from the quench tank.
For the demonstration, a heat exchanger evaporated contami-
nated aqueous feedstock to form steam and a concentrated
heated liquor. Atomizing nozzles sprayed the heated liquor,
with associated particulates, into the reactor. A separate pump
sent PCB-rich oils directly to the reactor through other atom-
izing nozzles. Compressed hydrogen-rich recirculation gas
passed through a gas-fired heat exchanger and entered the top
-------
Cl
Cl
+ 5H2
Cl
Cl
PCB
Benzene
4HCI
Hydrogen
Chloride
Polycyclic Aromatic
Hydrocarbons
f\J\
(Phenanthrene)
3H2
Benzene
Ethylene
Benzene
+ 9H2
->• 9CH4
Methane
C2H4
+ 2H2
2CH4
CnH(2n+2)
Straight-Chain
Hydrocarbons
(n-1)H2
n CH4
Methane
CH4
+ H20
CO
3H2
Figure 1. Gas-phase chemical reduction reactions.
7
-------
Recirculating Gas
)00°C^
k
ictor
0°C
a*
Scr
jbt
er
35°C ^
Gas
Booster
Scrubber
Water
Make-up
Hydrocarbon
Gas
-Propane
Condensate
Figure 2. Reactor and TDU system schematic diagram.
of the reactor tangentially. The tangentia 1 entry swirled the
fluids to provide effective mixing. As indicated in Figure 3,
the swirling mixture traveled downward in th annulus formed
by the reactor wall and the central ceramic-coated steel tube,
past electrically heated bars. These bars heated the mixture to
900<€ (1,650'F) At the bottom of the reactor the mixture
entered the tube, reversed direction, and flowed upward to the
outlet of the reactor. The reduction reactions occurred as the
gases traveled from the reactor inlets to the scrubber inlet.
After quenching, the gases flowed through a scrubber where
contact with water removed hydrogen chloride and fin e par-
ticulates. A large water-sealed vent, acting as an emergency
pressure relief duct, passed scrubber water to a tank below. A
pump recirculated the scrubber water in a loop through an
evaporative cooler to reduce its temperature to 35°C (95°F).
Caustic and make-up water, added to the scrubber liquor,
maintained hydrogen chloride removal efficiency. The scrub-
ber produced two effluent streams: sludge and decant water.
The reformed gas exiting the scrubber contained excess hy-
drogen, lighter hydrocarbon reduction products such as meth-
ane and ethylene, and a small amount of water vapor. A
portion of this hydrogen-rich gas was reheated to 500°C
(930°F) and recirculated back into the reactor; the rest of the
gas served as supplementary fuel for a propane-fired boiler.
The boiler produced steam used in the heat exchanger and
burned the reformed gas, which was the only air emission
from the process.
When treating wastes containing highly concentratd organ-
its, the process generates excess reformed gas. The system
can compress the reformed gas and store it for later use as fuel
in other parts of the process.
Test Conditions
In preparation for the SITE Demonstratiqn ECO LOGIC first
adjusted the system to obtain peak performance, then per-
formed a tracer material pretest to adjust sampling equipment
and trains. Two test runs (Conditions 1 and 3) followed over
the next 17 days; they introduced PCB-contaminated liquids
directly to the reactor. During two additional test runs (Condi-
tion 2) over an additional nine-day period the TDU processed
PCB-contaminated soil, sending desorbed gases to the reactor
for further treatment. The two runs treatd 1 .1 tons of soil
contaminated with 627 ppm of PCB and 14,693 ppm of
hexachlorobenzene (HCB) (tracer).
The ECO LOGIC SITE Demonstration objectives were as
follows:
-------
*™™,™™j
-1
To Scrubber
Waste Injection Ports
Reactor Steel Wall
Fibreboard Insulation
Refactory Lining
Electric Heating Elements
Ceramic-coated Central
Steel Tube
To Grit Box
Figure 3. The ECO LOGIC Reactor.
Demonstrate at least 99.9999% destruction and removal
efficiency (ORE) fo PCBs.
Demonstrate at least 99.99% destruction efficiency (DE)
for HCB added to the soil feed as a tracer.
Ensure that no dioxins and furans were formed.
Characterize product of incomplete combustion (PIC)
emissions.
Characterize hydrogen chloride emissions.
Document compliance with Michigan Department of Natu-
ral Resources (MDNR) air permit conditions.
Characterize criteria air pollutant emissions.
Validate key cost assumptions used in process economic
analyses.
Characterize effluents and residual streams relative to
disposal requirements.
Determine the suitability of the reformed gases for reuse./
resale.
. Demonstrate system reliability.
. Develop a system mass balance, including metals.
. Characterize critical process scale-up parameters.
. Validate the ECO LOGIC Chemical lonization Mass
Spectrometer (CIMS).
. Document system operation during all test runs.
Conclusions
Based on the program objectives, EPA found that the demon-
stration confirmed the feasibility of the gas-phase chemical
reduction process for treating PCBs and other chlorinated
organic compounds, producing a low Btu fuel gas frm PCB-
contaminated soil and providing environmentally acceptable
air emissions.
-------
The TDU did not perform to design specifications. EPA
categorized the TDU test data as a system proof-of-concept
rather than as a comprehensive evaluation of a fully devel-
oped unit. The test data indicated that the TDU, as presently
configured, achieved desorption efficiencies at the expense of
throughput. In addition ECO LOGIC experienced material
handling problems with the TDU feed. The combination of
material handling problems and inadequate organis desorp-
tion indicated a need for further development. The test data
have identified system strengths and targeted areas that re-
quite improvement.
Nevertheless, the demonstration did show that ECO LOGIC'S
TDU can desorb PCB contaminants. Treatment of the des-
orbed gas in the reactor produced stack emissions that gener-
ally met stringent regulatory levels. The reformed gas compo-
sition resembled coal-gas fuel. The scrubber liquor required
either disposal as a RCRA waste or recycling through the
system for additional treatment. Table 1 (Executive Sum-
mary) correlates the program conclusions with the program
objectives.
Technology Evaluation
The demonstrated ECO LOGIC TDU/Reactor Gas-Phase
Chemical Reduction Process is a pilot- or smdl commercial-
scale, trailer-mounted system, capable of treating wastewater,
waste oil, and solids such as soils. The TDU demonstration
(Condition 2) consisted of initial shakedown runs, a blank run
to determine train capacities, and two runs on soils to test the
unit's desorption capabilities. An independent AAR discusses
the demonstration of the stand-alore ECO LOGIC Reactor
System (Conditions 1 and 3).
A liquid pool of waste within the Middleground Landfill
provided both the liquid feedstock for the reactor process tests
(Conditions 1 and 3) and the contaminated soil for the TDU
tests. The Condition 2 runs treated 963 kg of soil contami-
nated with 627 ppm of PCBs and 14,693 ppm of HCB
(Condition 2)-a tracer material used to determin e DEs.
In an earlier Hamilton Harbor tept ECO LOGIC fed sediment
directly into the reactor system. However, the feed rates did
not meet their expectations. For the SITE Program, trje
installed the TDU as a front-end to the reactor ECO LOGIC
designed it to desorb organics from soil, feed the desorbed
organics to the reactor for further destruction, and then quench
the treated soil prior to disposal. They employed a molten
metal bath to dissolve unvaporized metals and to stabilize the
metals for later disposal.
The early stages of the TDU test revealed that the system
needed further engineering to improve materials handling and
to correct design deficiencies. The first run processed 2.12 kg/
min of contaminated soil. This was well below the 1 Okg/min
test target. Treated soil that was recovered from the TDU
quench tank showed inadequate contaminant removal.
ECO LOGIC further reduced the feed rate to 0. 4 kg/min to
improve removal efficiency. Although the test results indicate
that this was successful, the reduced throughput negatively
affected the economic viability of the TDU configuration.
ECO LOGIC must pursue further development and testing.
organics Destruction
To determine the efficiency of organics destruction, EPA
evaluated DREs, DES, benzene ring destruction, and forma-
tion of dioxins, furans, and PICs.
ORE
DRE compares the mass flow rate of selected f eedstoc k com-
......ids-iri this case PCBs-to their mass flow rate in the
)iler stack gas
(%) = (1
'stack'
Whenever possible, the evaluation based DRE calculations on
actual detected values. When the value was below the detec-
tion limit for the method, input stream values were set at zero,
while output streams were set at the detection limit value-the
most conservative approach.
The ECO LOGE TDU/Reactor Process achieved 99.9999%
DRE at the boiler stack. A low valu e (99.99%), attained by
Run 2, appears anomalous, since the DRE calculated upstream
of the boiler achieved the 99.9999% Toxic Substances Control
Act (TSCA) criterion. Therefore, it was removed from consid-
eration. These results show the system's potential to meet
established TSCA DRE requirements, qualifying it as a pos-
sible PCB destructor. However, data at higher soil throughput
are needed to ensure that adequate processing can be achieved
while still maintaining at least 99.9999 % DREs for PCBs.
addition, other TSCA requirements affecting residuals, stack,
and particulate emissions must also be met. The stand-alone
Reactor System tests achieved a 99.9999% DRE, supporting
the conclusion that a further-developed TDU/reactor system
could achieve TSCA destruction efficiencies.
DE
DE is a measure of the system's ability to achieve organics
destruction as measured around the system and all output
streams
DE (%) = (1 - Mass /Mass )xloo
nulniit inmil'
In
output
input'
An HCB additive acted as a tracer in the feedstock for DE
calculation. The program established 99.99% DE as an objec-
tive in order to compare the test results to RCRA incinerator
criteria. The system achieved a 72% DE for HCB in the soil
fed to the TDU for Run 1. After reducing the soil throughput,
the DE improved to 99.99% (test objective for Run 2). The DE
results are encouraging; they indicate that the desorption con-
cept is workable. Once ECO LOGIC resolves the throughput
deficiency, contaminant removal from the desorbed soils should
meet the 99.99% DE target.
Dioxins and Furans
The ECO LOGIC Process reduces organics in a high-tempera-
ture hydrogen environment, as opposed to combustion by
10
-------
incineration in an oxygen environment. The absence of oxy-
gen inhibits the formation of polychlorinated dibenzo-p-di-
oxin/polychlorinated dibenzofuran (PCDD/PCDF). Although
verifying the reduction mechanisms inside the reactor was not
an objective of the demonstration, the test confirmed a net
destruction at the stack of trace PCDD/PCDF in the feedstock.
Stack emissions registered from 0.04 to 0.14 ng/dscm for both
dioxin and furan-results well within incineration regulatory
guidelines. The low PCDD/PCDF stack concentrations sup-
port the conclusion that the system can effect a net PCDD/
PCDF destruction, resulting in PCDD/PCDF stack emission
concentrations significantly lower than current limits.
PICS
"PICs" is an incineration term not directly applicable to the
ECO LOGIC Process. The term describes a combustion
system's ability to degrade feedstock organics. hi a combus-
tion system the final gaseous products are ideally water,
carbon dioxide, and hydrogen chloride; other organic com-
pounds are PICs. The ECO LOGIC Process products are more
appropriately termed products of incomplete reduction (PIRs).
The process generates both PIRs during the gas-phase reac-
tions and PICs during the reformed gas combustion phase.
Both terms are used here to facilitate comparisons between
the emissions from combustion devices and those from the
ECO LOGIC Process.
Incineration processes often select total hydrocarbons (THC),
carbon monoxide, total PAHs, and benzene as indicators of
PIC/PIR formation. For the ECO LOGIC TDU test, the three
indicators, THC, carbon monoxide, and total PAHs, were
much lower than regulatory guidelines and well within the
MDNR permit conditions. The THC average was 0.65 ppmv,
carbon monoxide ranged from 8.2 to 13 ppmv, and average
total PAH was 3.35 m/dscm (all corrected to 7% oxygen, dry
basis). Benzene emissions, ranging from 2 to 6 |ig/dscm, met
MDNR permit conditions. The remedial manager can expect
that the ECO LOGIC TIN/Reactor Process will meet antici-
pated permit limits for THC, carbon monoxide, and PAH
emissions at other sites.
Air Emissions
EPA evaluated emissions of criteria air pollutants and hydro-
gen chloride, as well as compliance with the MDNR air
permit.
Criteria Air Pollutants
During the tests, continuous emission monitors (CEMs) mea-
sured the concentrations of the criteria air pollutants at the
stack: nitrogen oxides, sulfur dioxide, particulates, THC, and
carbon monoxide. Each of these pollutant emission concentra-
tions was low, well under the level established in the MDNR
air permit, nitrogen oxides averages ranged from 62.9 to 69.7
ppmv; sulfur dioxide, from 1.7 to 2.2 ppmv; and particulates,
from 0.13 to 0.43 mg/dscm (all corrected to 7% oxygen, dry
basis), oxygen averaged 7.9% and carbon dioxide, 8.9%. The
system can be expected to achieve similar results at other
sites.
The demonstration-scale boiler operated between high and
low fire, depending on the system's steam requirements. The
test analyses showed out-of-range spike concentrations of
THC and carbon monoxide (indicators of combustion effi-
ciency) during low-fire operation, most notably in Condition
1, Run 1 (one of the liquid runs), when the boiler was cycling
between high and low fire. Future users must be alert to the
potential for decreased combustion efficiency and increased
emissions of criteria air pollutants during low-fire operation.
The boiler should be operated at firing rates and air/fuel ratios
that prevent these spikes. Since the DREs were adequate in
the scrubbed reformed gas, reduced combustion efficiency in
the boiler will not affect the ability of the reactor process to
destroy hazardous organics.
Hydrogen Chloride
The ECO LOGIC TDU/Reactor System reduced stack hydro-
gen chloride emissions to below the MDNR-permitted levels.
RCRA emission limits set incinerator hydrogen chloride emis-
sions at 4 Ib/hr (or less), or 99% removal. The reactor system
easily achieved this; average stack concentrations were 0.68
mg/dscm at 153 mg/hr. Removal efficiencies reached 99.98%.
Most of the chlorine in the feedstock accumulated in the
scrubber effluent.
MDNR Permit Compliance
Table 2 compares the test results to the conditions imposed by
the MDNR air permit. PCB concentrations exceeded permit
limits for Run 2. However, PCB mass emissions met the
permit levels. ECO LOGIC'S dissatisfaction with the TDUs
desorption efficiency has already been noted.
In contrast to the reactor system tests-in which benzene
stack concentrations exceeded permit limits-the TDU tests
met MDNR limits. However, Condition 1 and 3 test results
suggest that future users should carefully monitor/control
benzene emissions; ECO LOGIC'S scale-up designs should
address these areas.
Intermediate and Residual Stream
Characterization
Intermediate and residual stream evaluations provided pro-
cess mass balance data; intermediate process, major effluent,
and miscellaneous stream characterizations; and confirmation
of adherence to TSCA permit conditions. Table 3 presents the
mass distribution of the waste feed and effluent streams as
fractions of the total waste feed. Although the excavated site
soil comprised only three-eights of the waste feed, it con-
tained approximately 99% of the contaminants. The three
major effluent streams were the stack gas, scrubber decant,
and treated soil. Most of the material in these streams entered
the process as combustion air and process water. Boiler com-
bustion air contributed most of the mass to the stack gas
stream; scrubber water, to the scrubber decant stream.
Table 4 shows the concentrations of the major contaminants
in the intermediate and effluent streams. These data indicate
the tendency of contaminants to concentrate in the various
intermediate and residual streams.
11
-------
Table 2.
MDNR Air Permit Conditions
Parameter
Unit
Permit .limit
Condition 2 average
HCI (7% O2, dry basis)
THC as methane
(7% O2 dry basis)
CO (7% O2, dry basis)
PCBs (dry basis)
Benzene (dry basis)
Chlorobenzenes as
1 ,2,4-trichlorobenzene (dry basis)
Opacity
Scrubber inlet temperature
Scrubber solution
On-line mass spectrometer
Reactor temperature
Reactor pressure
System oxygen
Gas booster dP
Recirculation flow rate
mg/dscm
Ib/hr
ppmv
Ib/hr
ppmv
Ib/hr
mg/dscm
Ib/hr
pgldscm
Ib/hr
ng/dscm
Ib/hr
%
°C
PH
Yes/No
in. H2O
%
in. H20
cfm
5.2
0.027
200
0.19
100
0.15
0.09
0.00048
20
0.00009
1
0.000002
0
>35
>8
Yes
>850
-------
Table 4.
Component Partitioning
Streams Coobl
Component
Total PCBs (mono-deca)
Total PAHs
Total PCDD/PCDF
Total chlorobenzenes
Total chlorophenols
Benzene
HCB
Component
Total PCBs (monodeca)
Total PAHs
Total PCDD/PCDF
Total chlorobenzenes
Total chlorophenols
Benzene
HCB
N/A Not applicable
ND Not detected
SAT Saturated
J Comnoundfs^ detected
Run
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
Run
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
R1
R2
at nonnentr?
SS3
Site
soil
538.000
716,000
24,900
ND
10.1
0.41
16.200
21 ,900
ND
ND
42'
ND
6,300
13,000
ss13
Scrubber
decantS
46.1
46.1
4,460
4,460
0.83
0.83
ND
ND
ND
ND
4,900
4,900
ND
ND
itions helow the nuantit;
SS10
Treated
soil
29,200
8,260
ND
2,400
6.34
6.04
ND
1,100
ND
841
ND
5,1 002
4,800,000
ND
Streams
ss22
Scrubber
liquor
26.2
26.8
4,120
3,700
0.001
0.001
ND
ND
ND
ND
340
2102
ND
ND
^tive limit
SS19
TDU exit
gas5
14,700
5,690
N/A
N/A
16.7
2.06
394,000
77,400
15,800
3,660
SAT
SAT
74,369
9,792
(PPb)
ss14
Reformed
gas5
1.26
1.67
N/A
N/A
0.0003
0.00094
ND
ND
ND
ND
SAT
SAT
ND
ND
SS11
Reactor
grit3
4,400
4,400
354,000
354,000
0.054
0.504
21 ,700
21,700
28,500
28,500
190
190
1
3,100,
3,100
SS24
Quench
water
3,220
121
51,100
414
0.185
0.044
ND
804
ND
15.36
1
240,
260
ND
790
ss12
Scrubber
sludge3
16,400
16,400
36,300,000
36,300,000
331
331
147,000
147,000
194,000
194,000
2,400
2,400
1
18,000,
18.000
SS16
Stack5
1
0.08
1.37
ND
ND
0.00014
0.00004
ND
ND
ND
ND
1
2
6
1
0.63
ND
Compound detected at concentrations above linear range for analysis.
From composite sample taken over two runs.
Naphthalene detected at concentrations below the quantitative limit.
Concentration given as ng/dscm.
13
-------
scrubber residuals (consisting of sludge, decant, and liquor),
and stack gas emissions. The stack gas emissions have already
been described.
The TDU'S capability to effectively desorb organics and
dissolve volatile metals from soil affects the process's versa-
tility. Without the TDU, the ECO LOGIC Process can treat
soil fed directly to the reactor, but feedstock size restrictions
(less than 1/4 in) would limit its application.
Intermediate Process Stream
Table 5 compares the reformed gas composition to several
commercially available fuels. The scrubbed reformed gas is
similar to blue water gas; its quality could be adequate to bum
in suitable combustion equipment during commercial-scale
operations. Use of the reformed gas in cogeneration or other
equipment to support the remedial operation could improve
the economics of large-scale applications. Although the re-
formed gas was of commercial quality, it would be a specialty
fuel requiring burners tailored to its properties. Compressing
and storing the reformed gas for resale or future use would
probably be uneconomical. Unlike propane, the compressed
reformed gas needs cryogenic temperatures to liquefy. There-
fore, storage as a gas would require excessively large tanks.
Benzene was the most prevalent PIR in the reformed gas.
Benzene concentrations exceeded the measuring capability of
the sampling train. However, as shown in Table 4, the com-
bustion step in the boiler destroys most of the residual ben-
zene. PAH emissions from the boiler stack also were low. The
reformed gas is generated from a hazardous waste, presenting
Table 5. Reformed Gas Comparison to Other Fuels
Composition, percent by volume
Gaseous
fuels
ECO LOGIC reformed
gas (average)
Blast furnace gas
Blue water gas
Carbon water gas
Coal gas
Coke-oven gas
Natural gas
(15.8%C2H6)
Producer gas
64.8
1.0
47.3
40.5
54.6
46.5
14.0
N2
21.6
60.0
8.3
2.9
4.4
8.1
0.8
50.9
02 CH4
0.015 7.1
0.7 1.3
0.5 '10.2
0.2 24.2
0.8 32.1
83.4
0.6 3.0
CO
3.3
27.5
37.0
34.0
10.9
6.3
28.0
C02
2.5
11.5
5.4
3.0
3.0
2.2
4.5
C2H4 CeHe
0.05
6.1 2.8
1.5 1.3
3.5 0.5
-
_
Gaseous
fuels
ECO LOGIC reformed gas (average)
Blast furnace gas
Blue water gas
Carbon water gas
Coal gas
Coke-oven gas
Natural gas (1 5.8% C2H6)
Producer gas
Molecular
wt. of fuel
Ib mass/
Ib mole
10.6
29.6
16.4
18.3
12.1
13.7
18.3
24.7
Higher heating
value
Btu/lbm
10,580
1,170
6,550
1 1 ,350
16,500
17,100
24,100
2,470
HHV
Btu/scf
286
89
277
535
514
603
1,136
157
Sp. gr.
air = 1 .0
0.40
1.02
0.57
0.63
0.42
0.47
0.63
0.85
MW Molecular weight
HHV Higher heating value
sp, gr. Specific gravity compared to air at 6Q°F
14
-------
a further difficulty in its utilization as a fuel outside of the
process. However, the results of the demonstration show that
burning the reformed gas in combustion equipment would
adequately destroy any residual hazardous organics.
Major Residual Streams
Treated soil-Since the TDU did not perform to design
specifications, the treated soil did not meet TSCA nonhazard-
ous disposal requirements. Table 4 indicates an order of
magnitude improvement in the residual PCB concentrations
between Runs 1 and 2. The Run 2 throughput was signifi-
cantly lower than Run 1; the PCB results reflect the improved
desorption resulting from increased residence time. This trend
indicates that after appropriate process modifications, the
treated soil might meet regulatory standards for commercial
disposal.
Tin bath-The molten metal bath heats the soil to volatilize
the organics; it absorbs metals and nonvolatile materials.
The program limited analytical efforts to metals analysis. The
results of these analyses ranged from nondetectable levels to
380 ppm phosphorus. Eventually the tin bath will need to be
reclaimed or replaced. For future applications, the remedial
manager should consider the economics of this procedure.
Quench water-ECO LOGIC designed the quench bath prin-
cipally to cool the hot desorbed soils; it also operates as a soil
scrubber to remove some of the contaminants not fully volatil-
ized or absorbed in the molten metal bath.
The quench water contained 0.260 ppm benzene, 0.185 ppb
PCDD/PCDF, 3.2 ppm PCBs, and 5 1.1 ppm total PAHs. It is
likely that the composition of this stream will change once the
TDU is fully developed and operational.
Reactor Grit-The first test run revealed that the reactor grit
volume was small enough for exclusion as an effluent stream.
Any accumulation can be either recycled or stored for permit-
ted disposal after the treatment program.
Considering only PCB congeners that have three or more
chlorine atoms (as defined by TSCA), the PCB concentration
detected in the grit was 3.63 ppm. A congener consists of all
PCB compounds having the same number of chlorine atoms
but arranged in different positions for any individual congener
compound. If monochlorobiphenyls, dichlorobiphenyls, and
nondetected congeners (assumed to be present at the detection
level) are included, the grit could contain a maximum 4.4 ppm
PCB concentration. The grit also contained 354 ppm PAH,
0.19 ppm benzene, 3.1 ppm HCB, 28.5 ppm total
chlorophenols, and 0.504 ppb dioxin/furan. These concentra-
tions could affect DE if the grit is considered a process output
rather than a recycled stream. However, at the commercial
scale, ECO LOGIC plans to recirculate this stream through
the reactor.
Scrubber Residuals-Trie scrubber is a critical component in
the gas-phase chemical reduction process. The scrubber effec-
tively removes a variety of organic and metallic compounds,
particulates, and chlorides. It is a key element in achieving
DREs. Table 4 shows elevated levels of hazardous organic
compounds in the scrubber sludge-mainly PAHs, with lesser
concentrations of benzene, HCB, other chlorobenzenes, PCBs,
and PCDD/PCDF. If this sludge is not recycled through the
process, it must be treated as a TSCA and RCRA hazardous
waste.
Based on detected PCB congeners, the PCB concentrations in
the scrubber decant (46.1 ppb total) and scrubber liquor
streams (26.8 ppb total) did not meet the TSCA criterion of
less than 3 ppb per PCB congener in liquid residuals. For the
demonstration, these streams were combined in a storage
tank. Subsequent sampling by TSCA personnel confirmed
that the stored liquids met the 3 ppb TSCA criterion.
If monochlorobiphenyls, dichlorobiphenyls, and nondetected
congeners (assumed to be present at the detection level) are
included, the scrubber decant could contain maximum PCB
concentrations up to 46 ppb; the scrubber liquor could contain
maximum PCB concentrations up to 26.8. The scrubber de-
cant also contained 4.46 ppm PAHs and 4.9 ppm benzene.
The scrubber liquor contained up to 4.1 ppm PAHs and 0.340
ppm benzene. If these streams are not recycled through the
process, they will require further treatment as a RCRA waste.
The scrubber residuals did not contain detectable levels of
chlorobenzenes or chlorophenols. Chlorobenzenes and
chlorophenols appeared only in the scrubber sludge.
Miscellaneous Streams
The demonstration team collected water that came in contact
with the processing equipment, such as wash and rinse water
from equipment decontamination, and stored it apart from
other wastes, disposing of it as a hazardous waste. The treat-
ment/disposal of this wash/rinse water is site-specific.
TSCA Permit Conditions
The program required that ECO LOGIC obtain a TSCA
research and development permit. The permit conditions ad-
dressed PCB throughput and PCB concentrations in the efflu-
ent streams. TSCA established maximum PCB levels of 2
ppm per congener in soil and 3 ppb per congener in water
streams. TSCA evaluated the combined scrubber liquid re-
sidual streams based on samples from the storage tanks. These
samples met the criterion that allowed disposal in a commer-
cial treatment system. However, the local publicly owned
treatment works (POTW) imposed stricter PCB effluent con-
centrations than those permitted by TSCA, requiring disposal
of the liquid residuals through a RCRA-permitted facility.
Since POTWs set their acceptance requirements based on
their effluent requirements, acceptance/rejection of the scrub-
ber liquid streams will be site-specific. In order for the ECO
LOGIC system to process PCB materials, a TSCA permit will
be required. The remedial manager should formulate a sched-
ule that includes obtaining a TSCA permit and addressing any
process and operating constraints that the permit may impose.
15
-------
Equipment and Operating Considerations
The remedial manager considering the use of the ECO LOGIC
TDU should understand the functions of the major equipment
components and potential operating problems associated with
them.
System Components
The principal components of the TDU system are its material
handling systems and its proprietary internal components.
More data on the proprietary elements will be available to
future users once ECO LOGIC develops and commercializes
the TDU.
Material Handling System-For the demonstration, the
material handling systems consisted of a feed conveyor, a
feed hopper, a feed screw, and a treated soil drag chain.
Blockages in the TDU feed hopper impeded operations. ECO
LOGIC resorted to both forcing feed through the hopper and
removing the feed screw to free blocked material. Inefficient
feed operations resulted in significantly reduced feed rates.
Material handling considerations (as with almost any system)
can affect system performance and costs. Adding various soil
pretreatment steps (size reduction, classification, etc.) might
prevent these material handling problems.
TDU Internal Components—The TDU design is propri-
etary. The reduced demonstration throughput resulted from
feed hopper restrictions and internal TDU operations. ECO
LOGIC is currently modifying both systems to improve
throughput.
System Reliability
The program evaluated system reliability during processing.
The TDU test was designed to treat 19 tons of material-only
1.1 tons were processed. The reliability has been expressed in
terms of percent of rated capacity-actual throughput (2.12
kg/min) compared to planned throughput (10 kg/min). This
translates to a 21% reliability.
In addition, the test plan had specified three replicate runs,
but only two runs were completed because of TSCA permit
restrictions. System availability-the number of planned test
days compared to the actual test days for the entire system
(TDU and TDU/reactor)-measured 24%.
Scale-Up Parameters
One program objective sought to identify the critical process
scale-up parameters. Knowing these parameters assists future
users in evaluating a proposed commercial-size operation.
The TDU's molten bath temperature and contaminant resi-
dence time are the critical parameters for efficient desotption.
CIMS Validation
The CIMS is the primary process control unit for the TDU/
reactor system. It records and stores data. It measures se-
lected compounds and their desorption products to maximize
organic destruction. Demonstration results show that the
CIMS may reflect data trends useful for process control but is
not, at this stage of its development, a reliable source of
quantitative data. Further testing will determine whether the
CIMS can provide adequate process control.
System Operating Conditions
Automatic computer data and manual logs documented pro-
cess operating conditions and the status of the operating
components. These data clarified process results and docu-
mented compliance with permit conditions. Table 6 lists the
averages for several key system parameters; the TER contains
further details.
Technology Applicability
This section discusses the applicability of the technology
relative to the site, waste media, safety, and staffing.
S/fe Characteristics
The ECO LOGIC system requires a fairly level area, approxi-
mately 120 ft x 180 ft, for the processing and auxiliary
equipment. Utility tanks require level surfaces or supports.
Except for process gas tank support pads, no additional sur-
face support is needed. The reactor system equipment sits on
two mobile trailers; a separate trailer transports the TDU,
solid feed hopper, and quench system.
Cold-weather operations may inhibit efficient destruction be-
cause of the incremental amount of energy required to heat the
reactor and the TDU molten metal bath. In addition, feedstock
liquids would require melting prior to treatment; liquid residu-
als could freeze in the unheated storage tanks. Winterization,
including heat-tracing, is necessary to provide adequatefeed-
stock and to ensure uninterrupted processing.
Applicable Media
Initially, ECO LOGIC designed the reactor system to process
liquids, with soil processing limited to about 30% solids.'
ECO LOGIC added the TDU to gain greater feedstock pro-
cessing capabilities.
Table 6.
Summary of TDU Operating Conditions
System
Component
Bath
Air Space
Combustion Gas
Exhaust Gas
Pressure
Soil
HCB
Parameter
Temperature (°C)
Temperature (°C)
Temperature (°C)
Temperature (°C)
in. H20
Feed Rate (kg/min)
Feed Rate
Run
1
616
614
662
396
2.0
2.12
16.6
Run
2
632
610
653
500
2.5
0.40
9.67
16
-------
Without the TDU, the reactor system can process soil fed
directly to the reactor. ECO LOGIC reports that they can
process soil sized 1/4 inch or less. However, the demonstra-
tion did not test this approach. ECO LOGIC decided, instead,
to test the soil treatment capabilities of the TDU, designed to
process 25 tons/day. The demonstration TDU throughput
reached about 1.1 tons. Data from the TDU Demonstration are
contained in the TER.
The reactor system is best suited for processing liquids and
TDU off-gases/water vapor. The waste's organic content lim-
its the demonstration-scale system's feed rate because of
reformed gas generation. Currently, ECO LOGIC plans to
improve throughput by storing excess reformed gas after
compressing it. Future users should consider the implications,
logistics, and costs of this approach.
Safety Considerations
The principal safety considerations for the ECO LOGIC Pro-
cess concern personnel, chemical use, equipment integrity,
and process control.
Personnel Safety
The components of personnel safety requiring attention are
those associated with Construction Safety Standards [29 CFR
19261 addressing such topics as slips, trips, and falls; confined
space entry; contingency planning; etc. The regulations in 29
CFR 1910.120 address personal protective equipment (PPE).
High voltage electrical equipment standards are also a con-
cern.
Chemical Use
Equipment Integrity
Verification of system component integrity is essential to
process safety. The remedial contractor should undertake
pressure testing, hydrostatic testing, and metal embrittlement
evaluations. The results should be certified before processing
hazardous materials. Hydrogen is more difficult to contain
than other gases because of its small molecular size. There-
fore, interfaces of equipment, instruments, and piping must be
leak-free. To provide an additional safeguard, ECO LOGIC
maintains the system under slight positive pressure, prevent-
ing inmtrattion of oxygen. As a safety backup, ECO LOGIC
monitors internal oxygen levels and maintains gas feeds (pro-
pane and hydrogen) at low pressure to prevent pipeline breaks.
Process Safety System
ECO LOGIC designed a safety system to immediately react,
should any system upset occur. The control system initiates
system shutdown in response to high oxygen content, high
pressure drop across the circulating fan, scrubber pump fail-
ure, ground faults, boiler failure, high hydrocarbon emissions,
or power failure. However, these shutdown systems were not
needed during the demonstration.
Whenever process conditions require a system shutdown, the
system program stops the waste input streams and replaces
them with clean steam to prevent any negative pressure in the
reactor. The program also stops hydrogen flow and introduces
a nitrogen purge. The TDU waste feed stops, but the drag
chain continues to operate. The hydrogen flow above the
molten bath halts; a nitrogen purge replaces it. Reformed gas
flow to the boiler stops. Either an operator or an automatic
computerized process controller initiates these events.
The chemical hazards of the ECO LOGIC Process accompany Staffing ISSUCS
the use of propane, liquefied nitrogen/oxygen, hydrogen, in-
dustrial chemicals, and hazardous feed material. In addition,
the process generates methane. Standardized industrial proce-
dures provide guidance for storing, transporting, and handling
these materials.
There should be no undue concern associated with hydrogen
use in the process. Well established and proven procedures are
available for safe hydrogen storage and use. Hydrogen is no
more nor less dangerous than gasoline or methane. As with
these substances, hydrogen must be handled with due regard
for its unique properties.
The electrical, petroleum refining, chemical, petrochemical,
and synthetic fuel industries have safely used hydrogen in
large quantities for decades. Through much of the last century
Europe successfully used hydrogen-enriched gases (coal gas,
town gas, producer gas) to satisfy residential fuel needs.2 The
Northeast United States used coal gas until the late 1950s.
For the demonstration, ECO LOGIC developed a Hydrogen
Safety Procedure based on the Canadian National Research
Council's Safety Guide for Hydrogen.2 Ultimately, remedial
managers must assure themselves that the flammable gases
used in the ECO LOGIC Process are handled, stored, and used
in accordance with industry standards and guidelines.
The CMS system facilitates monitoring and remote adjust-
ment of process parameters. This reduces labor requirements
for monitoring and maintenance personnel. The monitoring
personnel must be capable of evaluating system problems and
directing maintenance personnel in problem resolution. Since
operations can be controlled remotely, only those personnel
needing to manually adjust or maintain the system compo-
nents require PPE. Since the system will be processing haz-
ardous substances, the medical monitoring, training, and per-
sonal protection requirements of 29 CFR 1910.120 will re-
main in effect.
Regulatory Considerations
Several pieces of federal legislation and any state or local laws
present compliance considerations in operating the ECO
LOGIC System.
Clean Air Act
The Clean Air Act (CAA) establishes primary and secondary
ambient air quality standards to protect public health; it also
sets emission limits for hazardous air pollutants. Each state
administers its own permitting requirements as part of the
State Implementation Plan developed to bring the state into
17
-------
compliance with National Ambient Air Quality Standards
(NAAQS). These standards apply to the ECO LOGIC Process
because of its potential emissions. The process will probably
require an air permit to operate at any site, whether or not the
state has attained its NAAQS. Even if the area is in attain-
ment, Prevention of significant deterioration regulations may
further curtail emissions. Regulatory requirements must be
determined on a site-by-site basis.
Clean Water Act
The Clean Water Act (CWA) regulates direct discharges to
surface water through the National Pollutant Discharge Elimi-
nation System (NPDES). These regulations require that waste-
water point-source discharges meet established water quality
standards. The ECO LOGIC Process generates noncontact
and contact water discharges. Noncontact water sources in-
clude the heat exchanger, evaporative cooler, boiler water,
and blow-down. Contact water comes from the TDU quench,
scrubber liquor, tank cleaning, and equipment wash down; it
will likely require further treatment prior to discharge to a
POTW. In any case, wastewater discharge to a sanitary sewer
requires a discharge permit or, at least, concurrence from state
and local regulatory authorities that the wastewater is in
compliance with regulatory limits.
Comprehensive Environmental Response,
Compensation, and Liability Act
The Comprehensive Environmental Response, Compensa-
tion, and Liability Act (CERCLA) of 1980. amended 'by
SARA of 1986, provides federal funding to respond to re-
leases of hazardous substances to air, water, and land. Section
121 of SARA, entitled, "Cleanup Standards" states a strong
statutory preference for remedies that are highly reliable and
provide long-term protection. It recommends that remedial
action utilize on-site treatment that ". .. permanently and
significantly reduces the volume, toxicity, or mobility of
hazardous substances." In addition, remedial actions must
consider the technology's long-term and short-term effective-
ness, implementability, and cost.
EPA employed the TDU Demonstration only as a proof-of-
concept; the TDU requires further development to success-
fully handle contaminated solids. Based on this conclusion,
further development must take place before the TDU's capa-
bility to meet CERCLA requirements can be evaluated.
Occupational Safety and Health Act
Sections 1900 to 1926 of the Occupational Safety and Health
Act (OSHA) govern ECO LOGIC remedial operations:
1910.120 for hazardous waste, operations, 1926 for construc-
tion site activities, and 1910.1200 for worker and community
right-to-know.
Resource Conservation and Recovery Act
RCRA is the primary federal legislation governing hazardous
waste activities. RCRA Subtitle C contains requirements for
generation, transport, treatment, storage, and disposal of haz-
ardous waste, most of which are applicable to CERCLA
activities.
Depending on the waste feed and the effectiveness of the
treatment, the ECO LOGIC TDU/Reactor Process generates
three potentially hazardous waste streams: the scrubber li-
quor, the TDU quench water, and the treated soil. To generate
these wastes, the remedial manager must obtain an EPA
generator identification number and either comply with gen-
erator accumulation and storage requirements under 40 CFR
262, or receive a Part B Treatment, Storage, and Disposal
(TSD) interim status permit. CERCLA mandates compliance
with RCRA TSD requirements. A hazardous waste manifest
must accompany off-site waste shipment; transport must com-
ply with Federal Department of Transportation (DOT) haz-
ardous waste transportation regulations. The receiving TSD
facility must hold a permit and comply with RCRA standards.
Technology or treatment standards apply to many hazardous
wastes; those appropriate for the ECO LOGIC Process depend
on the waste generated. RCRA land disposal restrictions, 40
CFR 268, mandate hazardous waste treatment after removal
from a contaminated site and prior to land disposal, unless a
variance has been granted. The scrubber liquor, quench water,
and/or treated soil will require additional treatment prior to
land disposal, if they do not meet their pertinent treatment
standards.
Toxic Substances Control Act
The ECO LOGIC Process treats wastes containing PCBs.
Therefore, the remedial manager must address TSCA stan-
dards for PCB spill cleanups and disposal. The EPA docu-
ment, CERCLA Compliance with Other Laws Manual,3 dis-
cusses TSCA as it pertains to Superfund actions.
If ECO LOGIC plans to treat PCB-contaminated material
containing no RCRA wastes, they must obtain a TSCA autho-
rization. The conditions of this authorization may contain
operational, throughput, or disposal constraints that could
affect treatment efficiency and costs. If ECO LOGIC chooses
to treat PCB-contaminated material containing RCRA wastes,
a RCRA permit for a TSD facility will also be required.
State and Local Regulations
Compliance with applicable or relevant and appropriate re-
quirements (ARARs) may require meeting state standards that
are mote stringent than federal standards; state standards may
control non-CERCLA treatment activities. Several types of
state and local regulations affect operation of the ECO LOGIC
Process, such as, permitting requirements for construction/
operation, prohibitions on emission levels, and nuisance rules.
References
1. U.S. Office of Technology Assessment, "Dioxin Treat-
ment Technologies" (background paper), OTA-BP-0-93,
U.S. Government Printing Office, Washington, D.C., No-
vember 1991.
18
-------
2. Kalyanam, K. M, and Hay, D. R, Safety Guide for 3. U.S. EPA. CERCLA Compliance with Other Laws Manual
Hydrogen, National Research Council of Canada, Ot- Part II: Clean Air Act and Other Environmental Statutes
tawa, Ontario, 1987. and State Requirements, Interim Final, EPA/540/G-89/
009, OSWER, Washington, D.C., August 1989.
19
-------
Section 4
Economic Analysis
Introduction
Estimating the cost of employing an innovative technology is
a major objective in each SITE demonstration project. This
economic analysis presents data on the costs (excluding profit)
for a commercial-scale remediation using the ECO LOGIC
Gas-Phase Chemical Reduction Process. With a realistic un-
derstanding of the test costs, it should be possible to forecast
the economics of operating similarly sized systems or to
extrapolate these figures for larger systems at other sites.
The SITE Demonstration of the ECO LOGIC TDU/ Reactor
System conducted at the Middleground Landfill treated PCB-
contaminated soil. The demonstration achieved an average
system feed rate of 1.26 kg/min. Target rates for the test runs
were significantly more than the average rate actually achieved.
This economic analysis extrapolated the demonstration data
to commercial feed quantities, assumed at 300 tons of soil,
based on the ratios of wastewater, oil, and soil actually treated
during the demonstration.
Since the process would experience some downtime, three
different utilization factors have been presented: 60%, 70%,
and 80%. Certain cost elements were fixed; others were time-
sensitive.
Because the testing of the stand-alone reactor system and the
combined TDU/reactor system were concurrent, actual cost
data could not be completely isolated. However, financial
analysis allowed the extrapolation of data to provide for the
addition of the TDU.
Decreased process efficiency (lower utilization factor) would
require an extended time to process the same amount of
material, reflecting higher costs. Final figures have been
expressed as cost (1994 U.S. dollars) of material processed.
Conclusions
Previous demonstration data, reported in an independent AAR,
showed the commercial-scale ECO LOGIC Reactor Process
to be an acceptable remedial alternative for liquids contami-
nated with PCBs. Since the process was effective in treating
the PCB-contaminated Middleground Landfill liquids and
soils, it should be applicable to the remediation of other
similar sites.
The TDU/reactor process demonstration was, as explained
earlier, a proof-of-concept test. The results of the two runs in
Condition 2 showed a need for further development of the
TDU. The incremental TDU treatment costs (1994 U.S. dol-
lars) ranged from a low of $500/ton to a high of $630/ton,
depending on the utilization factor. Because of limited data,
the cost estimates presented in this analysis may range in
accuracy from +50% to -30%. an order of magnitude guide-
line suggested by the American Association of Cost Engi-
neers.
The cost effectiveness of employing a TDU in conjunction
with the ECO LOGIC Reactor Process cannot be assured
without further development and demonstration.
Issues and Assumptions
The costs associated with this technology were calculated on
the basis of demonstration parameters such as the following:
. A small to medium hazardous waste site.
. One ton of soil feed.
. A short treatment period during the SITE Demonstration.
While the equipment used for the demonstration was a small
commercial size, it may not be applicable where time con-
straints require increased capacity. The targeted test through-
put rates were considerably higher then those actually realized
during the demonstration. Variations in throughput could
significantly affect costs.
The reactor system, described in an independent AAR, treated
the desorbed gases from the TDU. This economic analysis
addresses only those costs associated with the soil feed pro-
cessed through the TDU.
Important assumptions regarding specific operating condi-
tions and task responsibilities, described below, will impact
cost estimates.
Site-Specific Factors
The demonstration site presented certain site-specific charac-
teristics that affected the cost estimate. Variations to these
characteristics may improve or worsen the project economics:
20
-------
* Proximity to utilities, with capacity sufficient to service
project
* Favorable ambient conditions
* Clear, level work area
* Small, specialized project with minimal requirements for
storage, administration, services, etc.
Fixed costs are not related to time or volume, nor are they
affected by project magnitude. Such costs include the trans-
portation/setup/removal of trailers, sanitary facilities, decon-
tamination facilities, process equipment, foundations, roads,
and utilities. In employing the results of this SITE economic
analysis to forecast a unit cost ($/ton), the potential user
should recognize that these same fixed costs spread over
larger volumes of contaminated material would lower the unit
cost. The reverse would be true of a smaller project.
Costs Excluded from the Estimate
Although the SITE Program provides 12-iatm list of costs
on which the economic analysis of a demonstration should be
calculated, not all 12 apply to every project. Certain cost items
were excluded from this analysis because they were either
site-specific, project-specific, or the obligation of the site
owner/responsible party.
Basis for Economic Analysis
To provide a basis of cost effectiveness comparison among
technologies, the SITE Program links costs to 12 standard
categories listed below:
Site preparation
Permitting and regulatory
Capital equipment
Mobilization and start-up
Operations labor
Supplies
Utilities
Effluents
Residuals
Analytical
Repair and maintenance
Demobilization
The cost estimates are representative of charges assessed to
the client by the vendor but do not include profit.
Some of the cost categories do not apply to this analysis
because they are site-specific, project-specific, or the obliga-
tion of the site owner/responsible party:
. Project engineering and design, specifications, requisi-
tions
. Permits, regulatory requirements, plans
. Wells, pipelines, excavation/stockpiling/handling of waste
(except for feed to process equipment),
. Backfilling, landscaping, any major site restoration
. Sampling and chemical analysis except as required for
disposal of miscellaneous effluents and wastes
. Initiation of monitoring programs
. Post-treatment reports, regulatory compliance
Wherever possible, applicable information has been provided
on these excluded costs so that potential users may calculate
site-specific economic data for their projects.
The TDU costs associated with the standard categories listed
below reflect only the incremental costs for the TDU compo-
nent of the combined TDU/reactor system. These costs are
shown in Table 7.
Site foundations for TDU and
preparation material handling equipment.
Capital TDU (and its internal components)
equipment conveyors
feed hopper
quench tank
[equipment costs have been annualized based on the formula
below:!
A - annualized cost, $
C - capitalized cost, $
i - interest rate, 6%
n - useful life, 10 years
Mobilization/ delivery
start-up taxes
insurance
working capital
equipment setup
trial bum
Operations
labor
Supplies
Utilities
Repairs and
maintenance
Demobilization
one technician per shift
propane
electric usage (mechanical
equipment)
$150/mo allowance
dismantling equipment
preparation for shipment
demolishing foundations
Results of Economic Analysis
The largest single cost component of this treatment technol-
ogy was the cost of TDU fuel-accounting for 67% of the
total treatment cost at 80% utilization. Labor comprise
capital equipment, 11%. The remaining categories comprised
only 13% of the total treatment cost. Assuming that the
throughput had achieved test target rates, the incremental
costs per ton would be less than $100.
21
-------
Table 7. Economic
System
Activity
Site preparation
Capital equipment
Start-up/mobilization
Labor
Supplies
Utilities
Maintenance costs
Demobilization
Totals
costs
Analysis for the
60%
(250 days)
2,600
20,000
5,000
50,000
100,000
8,500
1,200
2,000
189,300
$630/ton
ECO LOGIC
Utilization
70"%,
(21 4 days)
2,600
18,000
5,000
22,000
100,000
8,500
1,000
2,000
159,100
$530/ton
TDU
80%
(188 days)
2,600
16,000
5,000
13,800
100,000
8,500
900
2,000
149,800
$500/ton
References
1. Richardson Engineering Services Cost Estimating Guide,
Vol 1, 1993 edition.
R. S. Means. "General Building Construction," Cost 'Esti.
mating Services.
3. Evans, G. M. "Estimating Innovative Technology Costs
for the SITE Program." EPA/RREL for Journal of Air
Waste Management Association. July, 1990. Volume 40,
No. 7.
22
-------
Appendix A
Demonstration Sampling and Analysis
Introduction
The ECO LOGIC Reactor System SITE Demonstration con-
sisted of two test conditions with three runs each. Condition 1
treated PCB-contaminated wastewater; Condition 3,
PCB-contaminated waste oil. The TDU demonstration com-
prising Condition 2 processed contaminated soil-the subject
of an independent AAR.
Sampling and analysis of the feedstock, intermediate streams,
and residuals followed the procedures outlined in the demon-
stration plan. EPA subjected the entire sampling and analysis
program to a rigorous Category II Quality Assurance (QA)
procedure designed to generate reliable test data. The demon-
stration plan also contains the QA procedure. The TER pre-
sents a detailed account of the demonstration results.
Figure A-l shows the sampling locations. An SS designation
represents EPA contractor sampling locations shown in Table
A-l; MS indicates an ECO LOGIC Process monitoring sta-
tion, listed in Table A-2.
Methodologies
The EPA program sampled three matrices: gases, liquids, and
solids. EPA sampled and analyzed all key input and output
streams; they selected intermediate streams for physical prop-
erties (flow rate, density, moisture), PCBs, PCDD/PCDF,
PAHs, PCE, chlorobenzenes, chlorophenols, VOCs, 13 trace
metals, hydrogen chloride, oxygen, carbon dioxide, carbon
monoxide, sulfur dioxide, nitrogen oxides, THC, and other
selected compounds. Tables A-3, A-4, and A-5 list the sam-
pling and analysis methods used by EPA. The demonstration
plan and TER contain further details about the Sampling and
Analysis Program.
Rtt
Tank Condensais
Figure A-l. Sampling and monitoring stations.
23
-------
Table A-l. EPA Sample Locations
Stream Description
Location
SSI Wastewater
SS2 Waste oil
SS3 Contaminated soil
SS4 Caustic soda
SS5 Scrubber makeup water
SS6 Propane
SS7 Hydrogen
SS9 Combustion air
SSIO Treated soil
SS11 Reactor grit
ss1 2 Scrubber sludge
ss1 3 Scrubber decant
ss1 4 Reformed gas
SS15 Tank condensate
SS16 Stack gas
SS18 Heat exchanger
SS19 TDU gas
SS20 TDU molten bath
SS22 Scrubber liquor
SS24 Quench water
Feed line before pump
Oil drum
Feed drum
Caustic soda reservoir tank
Feed line
Feed line
Feed line
Boiler inlet
Treated soil collection drum
Reactor grit catchpot
Scrubber effluent tank
Scrubber effluent tank
Duct after gas booster fan
Bottom of condenser
Boiler stack
Heat exchanger residue
waste drum
TDU-to-reactor feed line
Bath vessel
Scrubber tank
Quench water tank
Table A-2. ECO LOGIC Process Control Monitoring Stations
Parameter
Temperature
Stations
2,3,4,5,6,7,9,
11, 12, 13, 15, 16,
17.18
Frequency
Continuous
Method
Thermocouple
Pressure
12. 13, 16,
1.4,7
7, 10
Continuous
Continuous
I/2 hour
Pressure transmitter
Differential pressure
transmitter
Gauge
Flow rate
7, 10
13
8
Continuous
Continuous
Hourly
Differential pressure
transmitter
Vortex flow meter
Orifice meter
Feed rate
13
14
Hourly
I/2 hour
Vortex flow meter
Tracer injection
PH
Gas constituents
Continuous
Continuous
pH meter
02 analyzer; CIMS
24
-------
Table A-3. Flue Gas Sampling and Analytical Methods
Analyte
PCBs
Dioxins/furans
PAHs
CB/CP
Volatile organics
Metals
HCI
Particulates
NOx
S02
02
C02
CO
THC
Fixed gases
Sulfur compounds
Heating value
Sampling
Principle
XAD-2
XAD-2
XAD-2
XAD-2
Tenax
Impinger
Impinger
Filter
OEMS
OEMS
OEMS
OEMS
OEMS
OEMS
Tedlar bag
Tedlar bag
Tedlar bag
Reference
Method 0010*
Method 0010
Method 0010
Method 0010*
Method 0030*
EPA Method 29 (draft)
EPA Method 26**
EPA Method 5"
EPA Method 7E"
EPA Method 6C"
EPA Method 3A"
EPA Method 3A"
EPA Method 10"
EPA Method 25A"
EPA Method 18"
EPA Method 18"
EPA Method 18"
Analytical
Principle
HR GC/HR MS
HRGC/HR MS
GC/MS
GC/MS
GC/MS
CVAAS, ICAP, GFAAS
1C
Gravimetric
Chemiluminescence
NDUV
Paramagnetic
NDIR
NDIR
FID
GC
GC/FPD
GC
Reference
EPA 680*
EPA 23**
EPA
EPA 8270*
EPA 5041*
EPA 29 (draft)
EPA 26**
EPA 5"
EPA 7E"
EPA 6C"
EPA 3A"
EPA 3A"
EPA 10"
EPA 25A**
MASA133*"
EPA 15**
ASTM 2620M
Test Methods for Evaluating Solid Wasfes, SW-646, U.S. EP^November 1966. reissued July 1992
and November 1992).
Code of Federal Regulations, 40 CFR 60.
Lodge, J.P., Methods of Air Sampling and Analysis. 3rd Edition, Lewis Publishers, Inc., Chelsea. Ml, 1969.
25
-------
Table A-4. Solids Sampling and Analytical Methods'
Analyte
PCBs
Dioxins/furans
CB/CP
PAHs
Volatile organics
Metals
Organic halogens
Inorganic halogens
Hexavalent chromium
Total sulfur
TCLP volatiles
TCLP metals
Ash
Heating value
Ultimate analysis
Total organic carbon
Density
Analytical
Principle
Reference
EPA 680*
EPA 8290*
EPA 8270*
EPA 8270*
EPA 8260*
EPA 6010,7471*
EPA 9020*
ASTM E776
EPA 7196*
ASTM D3177
EPA 8240*
EPA 6010,7470*
Combustion/gravimetric ASTM D482
Bomb calorimeter ASTM D240
Combustion ASTM D3176
GC EPA 9060*
Hydrometer ASTM D1298
GC/MS
HRGC/HR MS
GC/MS
GC/MS
GC/MS
CVAAS, AAS, ICAP
1C
1C
Calorimetric
Gravimetric
GC/MS
CVAAS, ICAP
Usina arab samples, oerformed in accordance with U.S. EPA Office of
Solid \Afeste document Test Methods for Evaluating
So/id Wastes, SW-846, 3rd Edition, Volume 11, Chapter 9, November
1966.
Table A-5. Liquids Sampling and Analytical Methods
Analyte
PCBs
Dioxins/furans
CB/CP
PAHs
Volatile organics
Metals
Organic halogens
Inorganic halogens
Hexavalent chromium
Total'sulfur
TCLP volatiles
TCLP metals
Ash
Heating value
Ultimate analysis
Total organic carbon
Density
PH
Analytical
Principle
GC/MS
HR GC/HR MS
GC/MS
GC/MS
GC/MS
CVAAS, ICAP
1C
1C
Calorimetric
ICAP
GC/MS
CVAAS, ICAP
Combustion/gravimetric
Bomb calorimeter
Combustion
GC
Hydrometer
pH meter
Reference
EPA 680*
EPA 8290*
EPA 8270*
EPA 8270'*
EPA 8260*
EPA 6010,7470*
EPA 9020*
EPA 325.2
EPA 7196*
EPA 6010*
EPA 8240
EPA 6010.7470*
EPA 160.4
ASTM 0240
ASTM D3176
EPA 9060*
ASTM D1298
EPA 9040*
Using grab samples, performed in accordance with U.S. EPA Office of
Solid Waste document Test Methods for Evaluating
Solid Wastes, SW-846, 3rd Edition, Volume II, Chapter 9. November
1966.
26
-------
Appendix B
Vendor's Claims
Introduction
Following the 1992 SITE Demonstration of the ECO LOGIC
Gas-Phase Chemical Reduction Process in Bay City, Michi-
gan, several advancements have been made. Further research
and development has focused on optimizing the process for PrOCBSS Chemistry
commercial operations, and improving the design of the soil/
sediment processing unit. These advancements along with
relevant background information are described herein.
organic, and soil matrices. This paper describes the process,
the commercial-scale system under construction, and the re-
sults of demonstration testing in Canada and the United
States.
Since 1986, ECO LOGIC has been conducting research with
the aim of developing a new technology for destroying aque-
ous organic wastes, such as contaminated harbor sediments,
landfill soil and leachates, and lagoon sludges. The goal was a
commercially viable chemical process that could deal with
these watery wastes and also process stored wastes '(e'.g.
contaminated soils, solvents, oils, industrial wastes, pesticides
and chemical warfare agents). Other companies and agencies
at that time were focusing their efforts primarily on incinera-
tion and were investigating a variety of predestruction clean-
ing or dewatering processes to deal with the problem of
aqueous wastes. The process described in this paper was
developed with a view to avoiding the expense and technical
drawbacks of incinerators, while still providing high destruc-
tion efficiencies and waste volume capabilities.
Following bench-scale testing supported by the National Re-
search Council, a lab-scale process unit was constructed in
1988 and tested extensively. Based on the results of these
tests, a mobile pilot-scale unit was constructed with funding
support from the Canadian Department of National Defence.
The pilot-scale plant was completed and commissioned in
1991. It was taken through a preliminary round of tests at
Hamilton Harbor, Ontario, where the waste processed was
coal-tar-contaminated harbor sediment. That demonstration
received funding from both Environment Canada's Contami-
nated Sediment Treatment Technology Evaluation Program
and the Ontario Ministry of Environment's Environmental
Technologies Program. In 1992, the same unit was taken
through a second round of tests as part of the EPA SITE
program in Bay City, Michigan. This demonstration was
partially funded by the Environment Canada Development
and Demonstration of Site Remediation Technology Program,
the Ontario Ministry of Energy and Environment Environ-
mental Technologies Program and the Canadian Department
of National Defense Industrial Research Program. In this test
program, the pilot-scale unit processed PCBs in aqueous,
The process involves the gas-phase reduction of organic com-
pounds by hydrogen at temperatures of 850°C or higher.
Chlorinated hydrocarbons, such as PCBs, and polychlorinated
dibenzo-p-dioxins (dioxins), are chemically reduced to meth-
ane and hydrogen chloride, while non-chlorinated organic
contaminants, such as PAHs, are reduced substantially to
methane and minor amounts of other light hydrocarbons. The
hydrogen chloride produced can be recovered as acid or
scrubbed out in a caustic scrubber downstream of the process
reactor.
Figure B-l shows some of the reduction reactions, including
intermediate steps, for the destruction of a variety of contami-
nants using the ECO LOGIC Process. Unlike oxidation reac-
tions, the efficiency of these reduction reactions is enhanced
by the presence of water, which acts as a reducing agent and a
source of hydrogen. The water shift reactions shown produce
hydrogen, carbon monoxide, and carbon dioxide from meth-
ane and water. These reactions can be used at higher efficien-
cies by subjecting scrubbed methane-rich product gas to cata-
lytic steam reforming, reducing the requirements for pur-
chased hydrogen.
A benefit of using an actively reducing hydrogen atmosphere
for the destruction of chlorinated organic compounds, such as
PCBs, is that no formation of dioxins or furans occurs. Any
dioxins or furans in the waste are also destroyed effectively.
The reducing hydrogen atmosphere is maintained at more
than 50% hydrogen (dry basis) to prevent formation of PAHs.
This makes the scrubbed recirculation gas suitable for con-
tinuous monitoring using an on-line CIMS. By measuring the
concentrations of intermediate reduction products, the CIMS
produces a continuous indication of destruction efficiency.
SE25 Commercial-Scale Process Unit
Figure B-2 is a schematic of the reactor where the destruction
of the waste takes place. The various input streams are in-
jected through several ports mounted tangentially near the top
of the reactor. Special nozzles are used to atomize liquid
27
-------
Cl Cl
CK>
O
+ 4HCI
Cl
Cl
PCB molecule and hydrogen react to produce benzene
and hydrogen chloride
+ 4 HCI + 2 H2O
14CH4
9 H2
6 CH4
Dioxin molecule and'hydrogen react to produce ben-
zene, hydrogen chloride, and water
PAH molecule and hydrogen react to produce methane
Benzene and hydrogen react to produce methane
C n H(2n + 2) + (n-1 )
CH4 + H20
n CH4
Water Shift Reactions
CO + 3H2
Hydrocarbons and hydrogen react to produce methane
Methane and water react to produce carbon monoxide
and hydrogen
CO + H20
C02 + H2
Carbon monoxide and water react to produce carbon
dioxide and hydrogen
Figure B-1.
ECO LOGIC Process reactions.
wastes to accelerate liquid vaporization. The gas mixture
swirls around a central stainless steel tube and is heated by 18
vertical radiant tube heaters with internal electric heating
elements. By the time it reaches the bottom of the reactor, the
gas mixture has reached a temperature of at least 850°C. The
process reactions take place from the bottom of the central
tube onward and take less than one second to complete.
Figure B-3 is a process schematic of the entire system, includ-
ing the reactor. Most of the system components are mounted
on highway trailers for ease of mobility. The reactor trailer
houses the reactor, the electric heating control system, the
scrubber system, the recirculation gas blower, the recircula-
tion gas heater, and the watery waste preheater vessel. A
second trailer contains the main power distribution room, the
dual-fuel steam boiler, the catalytic steam reformer, and an
auxiliary burner for excess product gas. Cooling water for the
scrubbing system is generated by skid-mounted evaporative
coolers, and scrubber stripping operations are carried out on a
small skid situated near the boiler. The product gas compres-
sion and storage system is also skid-mounted to allow flexibil-
ity in site layout. For processing soils and other solids, the
thermal desorption mill (TDM) is housed on a separate trailer,
and the sequencing batch vaporizer (SBV) is a skid-mounted
unit. The process control system, gas analyzer systems, and
command center are housed in a standard office trailer. Sev-
eral feed systems are available for various types of wastes,
depending on whether watery waste, oil waste, or solid waste
is being processed. Watery waste is preheated in a preheater
vessel using steam from the boiler. The contaminated steam
from the preheater vessel is metered into the reactor at a rate
determined by the process control system. Hot contaminated
liquid exits the bottom of the preheater vessel at a controlled
flow rate and enters the reactor through an atomizing nozzle.
Oil waste can be metered directly from drums into atomizing
nozzles using a diaphragm pump.
Solid wastes such as Soil or decanted sediment are decontami-
nated in the TDM with the desorbed contaminants being sent
to the reactor through a separate port. The internal workings
of the TDM are designed to vaporize ail water and organic
contaminants in the waste soil/sediment while mechanically
working the solids into a fine granular mixture for optimum
desorption. The water vapor and organic contaminants are
swept into the reactor by a sidestream of scrubbed recircula-
tion gas.
Solids such as contaminated electrical equipment can be
thoroughly desorbed using the SBVs. These chambers take
advantage of the reheated recirculation gas stream to heat the
equipment and carry contaminants into the reactor. The hy-
drogen atmosphere is nonreactive with most metals, and there
are none of the problems with metal oxide formation associ-
ated with rotary kilns. The SBV can also be used for vaporiza-
tion of drummed solid chemical wastes, such as HCB. Signifi-
cant stockpiles of "hex wastes" exist and are still being
generated as by-products of chlorinated solvent production.
Advantages of vaporizing hex wastes directly from the drum
28
-------
«"*">"
*r,V"^
i|r™s"™T smr r"T «° f
$ tx L s™
Waste Injection Ports
- Reactor Wail
8" Thick 'Pyro-Bloc1 "Y"
Ceramic Fibre Insulation
Stainless Steel Inner Lining
Radiant Tubes
Stainless Steel Central Tube
,j Grit box
Figure B-2. Commercial-scale process reactor.
29
-------
Recirculation Gas
Steam
Fuel
Exhaust
t
TT
H2N2.
I Steam I Recirculation
Reformer I Gas I
Fuel Heater
uu
i ny MV •
f j ICIMSI
Contaminated _ Fuel
Soil Exhaust
Fuel
1
Thermal
Desorption
Mill
Electrical
Energy
Desorbed
Gases
Cleaned Soil
Desorbed Gases
t
\ I T
Scrubber
I System
I
I
—€3
Gas
Blower
Reactor
HCL
Fuel for Boiler
Steam Reformer
TDM
Contaminated
Equipment
=L Gas
J* Storag<
Steam Waste
Fuel
Exhaust
Hot Watery Waste
J^Vapourizer
intaminate
: Oil
J
Contaminated
Water
Fue
Figure B-3. Commercial-scale process unit schematic.
include decreases in worker exposures and fugitive emissions
from drum transfer operations, cleaning of the drums in place,
and segregation of inorganic contaminants into the existing
drums. The SBV has been tested at lab-scale with hex waste
samples and PCB-contaminated electrical equipment.
The product gas leaving the reactor is scrubbed to remove
hydrogen chloride, water, heat, fine particulates, aromatic
compounds, and carbon dioxide. The first stage of the scrub-
ber can be operated to recover medium-strength hydrochloric
acid, which avoids the cost of neutralization with caustic. The
cost saving can be considerable if the waste stream is heavily
chlorinated, the acid can usually be recycled, and generation
of large volumes of salty wastewater is avoided. The second
stage of scrubbing drops the temperature of the gas to remove
water and completes the removal of hydrogen chloride by
caustic packed tower scrubbing. Particulate matter, which
may have entered the reactor as dissolved or suspended solids
in the watery waste, is removed in both the first and second
stages of the scrubber and is filtered out of the scrubber tanks
continuously. Heat is removed using plate heat exchangers on
the first two stages and cooling water from the evaporative
cooling system.
The third stage of scrubbing removes low levels of benzene
and naphthalene from the gas stream by neutral oil washing.
The oil is stripped end regenerated with the benzene and
naphthalene going to the inlet of the catalytic steam reformer.
The fourth scrubbing stage is removal of carbon dioxide using
monoethanolamine (MEA) absorption. The MEA is stripped
and regenerated with the carbon dioxide going to the boiler
stack.
The scrubber water from the stage-two scrubber leg returns to
the covered section of the scrubber tank through a drop-tube
that extends well below the water surface. This acts as a seal
against air infiltration and as an emergency pressure relief
mechanism. There will be no gas release if a short-term
pressure surge forces gas out of the bottom of this tube since a
check valve allows the gas to re-enter the system once the
pressure returns to normal. The system normally operates
within 10 inches water gauge (0.36 psi) of atmospheric pres-
sure.
As waste is processed through the system, acid and water are
produced as effluents. Filtered acid is pumped to a storage
tank for further activated carbon treatment prior to recycling.
30
-------
Excess water is also filtered and carbon-treated to remove any
trace of organic contamination and is then stored for analysis
prior to discharge. Carbon can be regenerated on-site in the
SBV, and the minor amount of scrubber sludge produced can
also be processed through the TDM or SBV.
The cooled and scrubbed product gas is a clean dry mixture of
hydrogen, methane, carbon monoxide, and other light hydro-
carbons. Some of the gas is reheated and recirculated back
into the reactor to increase the methane concentration in the
reactor when processing low-strength wastes. Recirculation
gas is also directed to the TDM as sweep gas, to the SBV as
sweep gas, to the catalytic steam reformer for hydrogen
generation, or to the compressor for storage.
Throughout waste processing operations, the product gas is
sampled for analysis by the QMS and other gas analyzers.
The CIMS is capable of accurately monitoring up to 10
organic compounds every few seconds at concentrations rang-
ing from percent levels down to ppb levels. It is used as part of
the ECO LOGIC Process to monitor the concentrations of
certain compounds indicative of the process destruction effi-
ciency. The compounds selected for monitoring depend on the
waste being processed. For example, during PCB processing,
monochlorobenzene is typically monitored as an indicator of
destruction efficiency. Low levels of this volatile compound
indicate that destruction of the PCBs is proceeding to comple-
tion. The CIMS readings are monitored by the process control
system, and the exceedance of alarm limits sends a message to
the operator (low-level alarm) or automatically curtails waste
input (high-level alarm). The CIMS also provides a continu-
ous record of the quality of the product gas being compressed
and stored.
Storage of the product gas under pressure permits the analysis
of large batches of gas prior to using it as fuel and allows the
operation of the system in a "stackless" mode. Should the
product gas not meet the quality criteria established, there will
have been no emissions to the environment, and the gas can
simply be reprocessed. Potential applications for the stored
product gas include heating the TDM, the catalytic steam
reformer, and the steam boiler. If more gas is generated than
can be used for fuel, an auxiliary burner located at the bottom
of the common boiler/steam reformer stack is used.
Demonstration Testing
The pilot-scale process plant was tested for the first time at
Hamilton Harbor, Ontario, in 1991. The waste processed
during those tests was harbor sediment contaminated with
coal-tar at concentrations of up to 300 g/kg (dry weight basis).
The harbor sediment was injected directly into the reactor as a
5- 10% solids slurry, since, at that time, the TDM had not been
developed. Also, the system had no catalytic steam reforming
or gas compression and storage capabilities, and the product
gas was sent directly to the dual-fuel boiler burner. DREs of
99.9999% were calculated (see Table B-l), based on the total
organic input and the PAHs analyzed in the boiler stack
emissions.' During one test, the liquid waste input was spiked
with PCBs to create a waste with a PCB concentration of 500
mg/kg. The concentration of PCBs in the air emissions, liquid
effluent, and processed solids were below the detection limits
for each, respectively. Based on the detection limits for the
stack sampling trains, a PCB ORE of at least 99.9999% was
achieved.
A second round of tests of the pilot-scale unit was conducted
in 1992 in Bay City, Michigan, as part of the EPA SITE
program. The wastes processed included oily
PCB-contaminated water, high-strength PCB oil, and
PCB-contaminated soil. As part of the demonstration, ECO
LOGIC constructed and commissioned a prototype TDU,
which was the forerunner of the current TDM, and demon-
strated the capability to compress and store the product gas
generated. The results for the test program, confirmed by
EPA,2 are shown in Table B-2. The SITE Program Project
Bulletins and TER have been published and will be followed
by the AARs.
The waste oil was obtained from beneath the Bay City landfill
and was analyzed by EPA to contain 25% PCBs and percent
levels of other chlorinated solvents. The contaminated soil
was obtained from installation of the sump wells used to
collect the oil, and the contaminated water was groundwater
from the landfill. The test matrix called for three water/oil
tests, three oil tests, and three soil tests.
The water/oil tests were to be nominally 4000 mg/kg PCBs,
based on injecting the water and oil in a 100:1 ratio through
Table B-1. Hamilton Harbor Performance Test Results
Run
P1
P2
P3
P3
Target
Analytes
PAHs
PAHs
PAHs
PCBs
Cone, in
Waste
{mg/kg}
21,000
30,000
30,000
500
Decant
Water Cone.
(H9*g)
483
680
423
NO
Grit
Cone.
(mg/kg)
1.67
7.76
0.37
ND
Sludge
Cone.
(mg/kg)
32,8
58.1
4,3
ND
Stack
Gas Cone.
Oig/m3)
0,27
0.23
0.14
ND
ORE
{%)
99.9999
99.9999
99.9999
99.9999
ORE - (Total Input - Stack Emissions) / (Total Input)
ND - Non-Detect
31
-------
Table B-2. U.S. EPA SITE Program Results
Water/Oil and High-Strength Oil Tests
Run
1
2
3
4
5
6
Soil Tests
Run
1
2
Waste Type
Water/Oil
Tracer
Water/Oil
Tracer
Water/Oil
Tracer
Oil
Tracer
Oil
Tracer
Oil
Tracer
Waste Type
Soil
Tracer
Tracer
Soil
Tracer
Tracer
Contaminant
PCBs
Perchloroethene
PCBs
Perchloroethene
PCBs
Perchloroethene
PCBs
Perchloroethene
PCBs
Perchloroethene
PCBs
Perchloroethene
Contaminant
PCBs
HCB
OCDD
PCBs
HCB
OCDD
Concentration
(mg/kg)
4,800
4,670
2,450
2,360
5,950
6,100
254,000
33,000
254,000
26,000
254,000
34,000
Concentration
(mg/kg)
538
12,400
0.744
718
24.800
1.49
Target
DRE/DE
99.9999
99.99
99.9999
99.99
99.9999
99.99
99.9999
99.99
99.9999
99.99
99.9999
99.99
Achieved
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Desorption Efficiency
{%}
94
72
40
99
99.99
99.8
the atomizing nozzle. Perchloroethene was added as a tracer
compound. The oil tests were designed to process the high-
strength oil at higher throughputs while demonstrating the
ability to compress and store the product gas generated. Steam
was added through a separate port, but liquid water was not
co-injected with the PCB oil. Again, perchloroethene was
added as a tracer compound. After oil waste processing, the
stored gas was directed to the boiler for about 24 hours, and
stack testing by the EPA subcontractor was conducted. The
target ORE for the PCBs was 99.9999%. and this was achieved
for all six tests. The target DE for the perchloroethene was
99.99%, and this was also achieved for all six tests. The SITE
program analytical results for the input concentrations of the
water/oil mixture and the high-strength oil are shown in Table
B-2.
Soils with various contamination levels were mixed to pro-
duce a relatively homogeneous quantity of soil with a nominal
1000 mg/kg PCB concentration. The soil test runs were con-
ducted after construction and commissioning of the new TDU
was completed. During the first TDU test, contaminated soil
was processed with a desorption efficiency of 94%, resulting
in a processed soil PCB concentration of 30 mg/kg. This result
was encouraging for a first run, but the desorbed soil was still
above the TSCA disposal criteria of 2 mg/kg. The waste soil
residence time inside the TDU was increased for the second
run, and a desorption removal efficiency of 99% was achieved
according to SITE program results. The tracer compound used
for the soil tests was HCB, which was spiked at significantly
higher concentrations than the PCBs. The hexachlorobenzene
was also contaminated with significant levels of
octachlorodibenzo-p-dioxin (OCDD). The desorption efficien-
cies achieved for the HCB and OCDD for Test 2 were 99.99%
and 99.8%o, respectively. Due to TSCA permit restrictions,
only two runs were performed for the third test condition. It
should be noted that the performance of the TDU is indepen-
dent of the destruction process. The reactor destruction effi-
ciencies for the desorbed contaminants were high for both
TDU runs.
An additional component of the test program was a 72-hour
endurance test aimed at demonstrating the continuous opera-
tion capabilities of the ECO LOGIC Process. The equipment
operated perfectly and the 72-hour test was concluded suc-
cessfully.
Current Status
The ECO LOGIC Process has been demonstrated to be a
high-efficiency alternative to incineration for the destruction
of PCB wastes. High water-content wastes and high-strength
oils can both be processed with destruction removal efficien-
cies of at least 99.9999%. The ability to compress and store
the product gases generated during processing means that no
uncontrolled air emissions occur.
32
-------
The existing pilot-scale unit is presently available for further
research and development work including new applications
such as mixed wastes (low-level radioactive PCBs), chemical
warfare agents, and explosives. Further research and develop-
ment over the last 18 months has focused on optimizing the
process for commercial operations and improving the design
of the soil/sediment processing unit. The TDM design cur-
rently under construction has now achieved excellent results
in lab-scale research and development supported by the Na-
tional Research Council Industrial Research Assistance Pro-
gram. Soils and sediments have been desorbed from ppm and
percent levels down to low ppb levels, which are orders of
magnitude below disposal criteria. Table B-3 shows the re-
sults of a number of lab-scale TDM runs processing a variety
of waste types. The SE25 commercial-scale system now un-
der construction has a design capacity of 100-300 tonnes/day
of contaminated soil or sediment and 20 tonnes/day of PCB
askarel fluid. The cost of processing these waste streams is
estimated at $400 and $2,000 per tonne, respectively. The first
SE25 system is being exported to Australia and will begin
operations with a contract from Australian government agen-
cies for 200 tonnes of obsolete pesticide destruction. Con-
struction of a second SE25 system is also commencing to
serve the North American market, and this unit should be
commissioned for commercial use by the end of 1994. ECO
LOGIC has made proposals to several major North American
corporations and a number of government agencies for the
cleanup of contaminated sites.
Treatability studies using ECO LOGIC'S lab-scale destruction
system are continuing. The lab-scale equipment includes a
TDM for processing soil or sediment and an SBV suitable for
processing samples of chemical wastes or contaminated elec-
trical equipment. Clients find that treatability studies are a
cost-effective method for determining the applicability and
effectiveness of the ECO LOGIC Process to their waste
problems.
The ECO LOGIC Process is a proven technology for the
destruction of high-strength PCB oil wastes and is suitable for
the destruction of askarel fluids used in electrical equipment
and PCBs and other organic contaminants in soils and sedi-
ments. ECO LOGIC offers a cost-effective alternative to
incineration and can provide a complete on-site destruction
service for the owners of hazardous organic wastes.
References
I. WTC Newsletter, published by the Wastewater Technol-
ogy Centre, Environment Canada, No. 2, March 1992.
Contact: Mr. Craig Wardlaw, Project Scientific Author-
ity, 905-336-4691.
2. Technology Evaluation Report, SITE Program Demon-
stration, Risk Reduction Engineering Laboratory, U.S.
EPA, Cincinnati, OH 45268, July 15, 1994. Contact: Mr.
Gordon Evans, SITE Project Manager, 513-569-7684.
Table B-3. Summary of Test Results from the Lab-Scale Thermal Desorption Mill
Waste Type
Soil (tarry, oily)
Soil (dry, sandy, PCB-spiked)
Soil (dry, sandy, PCB-spiked)
Sediment (muddy, fine, PCB-spiked)
Sediment (muddy, fine, PCB-spiked)
Sediment (muddy, fine, PCB-spiked)
Sediment (muddy, fine)
Sediment (muddy, fine)
Sediment (muddy, fine)
Sediment (muddy, fine)
Sediment (muddy, fine)
Sediment (muddy, fine)
Sediment (muddy, fine)
Sediment (muddy, fine)
Waste PCB
Concentration
(ppm)
39
440
520
710
790
750
7.3
8.3
8.3
420
420
2000
1200
8.3
Grit PCB
Concentration
(ppm)
0.011
0.0039
0.0016
0.026
0.0097
0.065
0.0029
0.0066
0.0013
0.0017
0.012
0.044
ND (0.011)
ND (0.005)
33
------- |