c/EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/540/AR-93/522
September 1994
Eco Logic International
Gas-Phase Chemical
Reduction Process—
The Reactor System
Applications Analysis Report
SUPERFUND INNOVATIVE
TECHNOLOGY EVALUATION
-------
EPA/540/AR-93/522
September 1994
Eco Logic International Gas-Phase Chemical Reduction
Process—The Reactor System
Applications Analysis Report
Risk Reduction Engineering Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
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 liquid feedstocks; and a
second, independent AAR, which interprets the data and discusses the applicability of the
Thermal Desorption Unit (TDU) to soil 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 liquid matrix. The report presents data from the
recent EPA SITE Demonstration of the reactor system, provides case studies, and evaluates
the costs of operating the system.
The ECO LOGIC Reactor System thermally separates organics, then chemically
reduces them in a hydrogen atmosphere, converting them to a reformed gas that consists of
light hydrocarbons and water. A scrubber treats the reformed gas to remove hydrogen
chloride and particulates. Of this gas, a portion recycles back into the reactor; the
remainder is either compressed for storage or feeds a propane-fired boiler prior to release
to the atmosphere. The reactor system produced two principal residual streams: reformed
gas and scrubber effluent.
The SITE Program evaluated the ECO LOGIC Process at the Middleground Landfill
in Bay City, Michigan. The reactor system processed 2.9 tons of wastewater and 0.2 tons of
waste oil, both contaminated with polychlorinated biphenyls (PCBs). The reactor system
demonstration revealed that the process can successfully treat both Toxic Substance
Control Act (TSCA) and Resource Conservation and Recovery Act (RCRA) hazardous
compounds. Although the reactor is not classified as an incinerator, stack emissions met
the TSCA destruction and removal efficiency for PCBs and the RCRA destruction and
removal efficiency for tracer compounds, which are specified in the respective incinerator
regulations. The system produced liquid effluent streams that may require further treatment
prior to publicly owned treatment works (POTW) processing.
The ECO LOGIC Process is best suited to sites that contain oily liquid wastes. Costs
fell in the range of $7.68/gal (60% utilization) to $6.41/gal (80% utilization) for liquid
feed. The ECO LOGIC Thermal Desorption Unit (TDU)/Reactor System Demonstration, a
proof-of-concept test that processed contaminated soil, is the topic of a second, indepen-
dent AAR.
IV
-------
Contents
Foreword iii
Abstract iv
Figures vii
Tables viii
Abbreviations ix
SI Conversion Factors xii
Acknowledgements 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 7
Conclusions 8
Technology Evaluation 9
Organics Destruction 10
Air Emissions 10
Intermediate and Residual Stream Characterization 12
Equipment and Operating Considerations 15
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 17
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 19
Introduction 19
Conclusions 19
Issues and Assumptions 19
Site-Specific Factors 19
Costs Excluded from the Estimate 20
Utilities 20
Supplies 20
Operating Conditions 20
Labor 20
Basis for Economic Analysis 20
Site Preparation Costs 20
Permitting and Regulatory Costs 21
Capital Equipment 21
Mobilization and Start-up 22
Operations Labor 22
Supplies 22
Utilities 22
Effluents 22
Residuals 22
Analytical 22
Repairs and Maintenance 23
Demobilization 23
Results of Economic Analysis 23
References 23
Appendices
A. Demonstration Sampling and Analysis 25
B. Vendor's Claims 29
C. Case Studies 36
VI
-------
Figures
1. Gas-phase chemical reduction reactions
2. Reactor system and TDU schematic diagram
3. The ECO LOGIC reactor
A-l. Sampling and monitoring stations 25
B-l. ECO LOGIC process reactions 30
B-2. Commercial-scale process reactor 31
B-3. Commercial-scale process unit schematic 32
C-l. Bench-scale reactor system schematic diagram 37
VII
-------
Tables
1. Summary Results of Reactor System Tests 3
2. MDNR Air Permit Conditions 11
3. Mass Distribution of Selected Streams 12
4. Component Partitioning 13
5. Reformed Gas Comparison to Other Fuels 14
6. Summary of Reactor Operating Conditions 16
7. Consumables Required by the ECO LOGIC Reactor Process 20
8. Operating Labor 21
9. Labor Costs Based on Utilization 21
10. Demonstration Site Preparation Costs 21
11. Capital Equipment for Commercial Operation 22
12. Mobilization/Start-up Costs 23
13. Cost Allocations 23
14. Economic Analysis for the ECO LOGIC Reactor System 24
15. Cost Extrapolations 24
A-l. EPA Sample Locations 26
A-2. ECO LOGIC Process Control Monitoring Stations 26
A-3. Flue Gas Sampling and Analytical Methods 27
A-4. Solids Sampling and Analytical Methods 27
A-5. Liquids Sampling and Analytical Methods 28
B-l. Hamilton Harbor Performance TestResults 33
B-2. U.S. EPA SITE Program Results 34
B-3. Summary of Test Results from the Lab-Scale Thermal Desorption Mill 35
C-l. Sediment Samples 38
C-2. Residue Streams 38
C-3. Performance Results 38
C-4. Waste Input and Effluent Analysis Components 40
C-5. Characterization Test Air Sampling Components 41
C-6. Air Emission Sampling Components 42
C-7. Characterization Test Results 43
C-8. Performance Test Results 44
VIM
-------
Abbreviations
AAQ ambient air quality
AAR Applications Analysis Report
AAS atomic absorption spectroscopy
ALR analytical linear range
ASTM American Society for Testing and Materials
BOC beginning of condition
BOR beginning of run
Btu/lbm British Thermal Unit per pound mass
ARARs applicable or relevant and appropriate requirements
C carbon
CAA Clean Air Act
CB chlorobenzene
CEM continuous emission monitor
CEMS continuous emissions monitoring system
CERCLA Comprehensive Environmental Response, Compensation, and Liability Act
cfm cubic feet per minute
CFR Code of Federal Regulations
CIMS Chemical lonization Mass Spectrometer
Cl chloride
C12 chlorine
CO carbon monoxide
CO2 carbon dioxide
CP chlorophenols
CVAAS cold vapor atomic absorption spectroscopy
CWA Clean Water Act
dP pressure differential
DE destruction efficiency
DOT U.S. Department of Transportation
DRE destruction and removal efficiency
dscm dry standard cubic meter
dscf dry standard cubic foot
IX
-------
Abbreviations (Continued)
ECO LOGIC ELI Eco Logic International, Inc.
EPA U.S. Environmental Protection Agency
EER Energy and Environmental Research Corporation
FID flame ionization detection
FPD flame photometric detector
ft feet
FWEI Foster Wheeler Enviresponse, Inc.
g gram
gal gallon
GC gas chromatography
GF graphite furnace
gr grains
gpm gallons per minute
HR high resolution
hr hour
H2 hydrogen
HCB hexachlorobenzene
HC1 hydrogen chloride
ICAP inductively coupled argon plasma spectroscopy
in. inches
kg kilogram
Kw kilowatt
L liter
Ib pound
m meter
MASA Method of Air Sampling and Analysis
MDNR Michigan Department of Natural Resources
mg milligram
min minute
mo month
MS mass spectroscopy
NAAQS National Ambient Air Quality Standards
NDIR non-dispersive infrared
NDUV non-dispersive ultraviolet
ng nanogram
NOx nitrogen oxides
NPDES National Pollutant Discharge Elimination System
O2 oxygen
ORD Office of Research and Development
-------
Abbreviations (Continued)
OSHA Occupational Safety and Health Act
OSWER Office of Solid Waste and Emergency Response
PAHs polycyclic aromatic hydrocarbons
PCBs poly chlorinated biphenyls
PCDD polychlorinated dibenzo(p)dioxin
PCDF polychlorinated dibenzofuran
PCE perchloroethylene (tetrachloroethene)
pH a measure of acidity/alkalinity
PICs products of incomplete combustion
PIR product of incomplete reduction
POHC principal organic hazardous constituent
POTW publicly owned treatment works
PPE personal protective equipment
ppb parts per billion
ppm parts per million
ppmv parts per million by volume
psi pounds per square inch
psig pounds per square inch gauge
QA quality assurance
QI quality indicator
RCRA Resource Conservation and Recovery Act
RREL Risk Reduction Engineering Laboratory
SARA Superfund Amendments and Reauthorization Act
S sulfur
sec second
scf standard cubic feet
scfm standard cubic feet per minute
SITE Superfund Innovative Technology Evaluation Program
SO2 sulfur dioxide
SVOC semivolatile organic compounds
TCLP Toxicity Characteristic Leaching Procedure
TDU thermal desorption unit
TER Technology Evaluation Report
THC total hydrocarbons
TSCA Toxic Substances Control Act
TSD treatment, storage, and disposal
[ig microgram
VOC volatile organic compounds
VOST volatile organic sampling train
WTC Wastewater Technology Center
XI
-------
SI Conversion Factors
Multiply
Area:
Flow rate:
Length:
Mass:
Volume:
Temperature:
Concentration:
Pressure:
Heating value:
English (US)
Units by
1ft2
lin.2
1 gal/min
1 gal/min
1MGD
1ft
lin.
lib
lib
1ft3
1ft3
Igal
Igal
°F-32
1 gr/ft3
1 gr/gal
1 lb/ft3
1 lb/in.2
1 lb/in.2
Btu/lb
Btu/scf
Factor
0.0929
6.452
6.31xlO-5
0.0631
43.81
0.3048
2.54
453.59
0.45359
28.316
0.028317
3.785
0.003785
0.55556
2.2884
0.0171
16.03
0.07031
6894.8
2326
37260
Metric (SI)
to get Units
m2
cm2
m3/s
L/s
L/s
m
cm
g
kg
L
m3
L
m3
°C
g/m3
g/L
g/L
kg/cm2
Newton/m2
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
technical support; 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 sound 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 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 the Ontario Ministry of the Environment and Energy;
EPA conducted the demonstration of the ECO LOGIC Pro-
cess at Bay City's Middleground Landfill. The landfill ac-
cepted municipal and industrial wastes for approximately 40
years. A 1991 remedial investigation indicated elevated levels
in groundwater of trichloroethene, PCBs, 1,2-dichloroethene,
methylene chloride, toluene, and ethylbenzene. The ground-
water contained lesser concentrations of benzidine, benzene,
vinyl chloride, chlorobenzenes, polycyclic aromatic hydrocar-
bons (PAHs), lindane, dieldrin, chlordane, and DDT metabo-
lites.
The patented ECO LOGIC Gas-Phase Chemical Reduction
Process treats organic hazardous waste in a hydrogen-rich
atmosphere at approximately 900°C (1,650°F) and ambient
pressure to produce a 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 reduction 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 particulates. Of this gas, 95% recycles back into the
reactor; 5% 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 the RCRA
requirements, making the reformed gas environmentally ac-
ceptable 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 a complementary TDU, the topic of a second, inde-
pendent AAR. Each reactor test condition consisted of three
runs. The reactor program treated approximately 2.9 tons of
wastewater and 0.2 tons of waste oil. This report presents only
the results of the reactor tests. The program also conducted a
72-hour test to evaluate the reactor system reliability.
EPA collected extensive samples at points around the major
system components and stored or logged important data on
system operation and utility usage. Laboratory analyses pro-
vided information on the principal process streams: reactor
grit, scrubber residuals, reformed gas, and boiler stack emis-
sions. EPA evaluated these data against established program
objectives 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 contaminated liquids and
providing environmentally acceptable air emissions.
In general, ECO LOGIC'S Reactor System effectively de-
stroyed PCBs, reducing them to lighter hydrocarbons. Theo-
retically, the destruction process could depend on both the
reactor system's gas phase reduction reactions, which pro-
duced the reformed gas, and on the propane/reformed gas-
fired boiler, a combustion device.
Although the result was not listed as a primary or secondary
objective for the demonstration, destruction and removal effi-
ciencies (DREs) for PCBs in the scrubbed reformed gas were
essentially equal to the DREs achieved at the boiler stack.
This shows that combustion of the reformed gas in the boiler
is not required to complete PCB destruction.
Stack emissions generally met stringent regulatory levels.
However, average benzene concentrations in the stack gas—
corrected to 7% oxygen—(Condition 1-73 jj^g/dscm; Condi-
tion 3-113 |j,g/dscm) and scrubber liquor (Condition 1-18.5
|J.g/L; Condition 3 - 347 |J,g/L) required close monitoring. The
reformed gas composition resembled coal-gas fuel. The scrub-
ber liquor required either disposal as a RCRA waste or
-------
recycling through the system for additional treatment. Table 1 independent AAR presents the results of the TDU (Condition
correlates the program conclusions with program objectives. 2) tests.
Waste Applicability Costs
The SITE Program concluded that the ECO LOGIC Process The 12 categories established for the SITE Program formed
efficiently treated liquid wastes containing oily PCBs, other the basis for the cost analysis. Costs relate to the reactor
organics, and water containing PCBs, other organics, and system, processing an average of 2.2 kg/min, as operated at
metals. Stack emissions met stringent regulatory levels. The the Middleground Landfill. Based on the economic analysis,
principal residual stream—the scrubber effluent—concentrated the estimated cost (1994 U.S. dollars) for treating liquid
metals and some organics (benzene, PCBs, and PAHs), indi- wastes similar to those at the Bay City site range from $2,0007
eating that additional treatment might be required prior to ton (60% utilization factor) to $l,670/ton (80% utilization
disposal. factor). The most important element affecting cost is labor
(52% of cost), followed by site preparation (15%), supplies
The reactor did not directly process soil. Instead, ECO LOGIC (12%), and start-up/mobilization (12%).
provided a complementary front-end TDU to treat soils. An
-------
Table 1.
Summary Results of Reactor System Tests
Objective
Not
Met Met
Results
Range
Conclusions
Demonstrate ORE for PCBs: 99.9999% X
Demonstrate DE for PCE: 99.99% X
Ensure no formation PCDD/PCDF X
Characterize PIC emissions X
Characterize HCI emissions
Document MDNR air permit compliance
Characterize criteria air pollutants X
Document TSCA permit compliance X
Validate key cost assumptions X
Characterize effluents and residuals X
Determine suitability of reformed gases
for reuse/resale
Demonstrate system reliability
99.9999% to 99.99999%
99.99%
PCDD DE 63.05% to 98.36%
PCDF DE 99.91% to 99.98%
Benzene: 73 to 113 |ig/dscm
0.659 to 0.807 mg/dscm;
109.1 to 197.8 mg/hr;
99.98% removal
Benzene: 61 to 109 |ig/dscm
Throughput reliability: 20 to 55%
of design. System availability: 24%
Good destruction.
Good destruction.
No net PCDD/PCDF
formation.
PICs characterized;
benzene emissions
exceeded regulatory
limit.
Acceptable emissions.
Air permit compliance
documented; benzene
emissions exceeded
MDNR conditions.
Easily met permit
conditions.
Met permit conditions.
Cost elements identified.
Organics destroyed;
metals partitioned to
scrubber effluents;
after further treatment,
scrubber liquor may
be suitable for POTW.
Closely matched
composition of other
commercial fuel gases.
Process reliability
requires improvement.
Develop mass balances
Characterize scale-up parameters
Validate CIMS
Document system operation
Generally good closures,
except for certain
metals.
Characterized.
May reflect data trends
useful for process
control.
Data available for
commercial scale-up.
-------
Section 2
Introduction
The SITE Program
In 1986 EPA's Office of Solid Waste and Emergency Re-
sponse (OSWER) and 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 corrections. The SITE Program includes four major
elements: the Demonstration Program, the Emerging Tech-
nologies Program, the Measurement and Monitoring Tech-
nologies Program, 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 technolo-
gies 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 ef-
fective dissemination 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
applications, and discusses the advantages, disadvantages,
and limitations of the technology. The AAR attempts to
synthesize available information and draw reasonable conclu-
sions 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
applications.
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 devel-
opment at the commercial scale. The AAR can assist remedial
managers in planning Superfund cleanups; it represents an
-------
important tool in the development and commercialization of James Nash
the technology. ELI Eco Logic International, Inc.
143 Dennis St.
Key Contacts Rockwood, Ontario NOB 2KO
Canada
The sources listed below can provide additional information Phone: 519-856-9591
concerning the SITE Demonstration, the site, or the ECO Fax: 519-856-9235
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
-------
Section 3
Technology Applications Analysis
This AAR assesses the capability of the ECO LOGIC Process
to treat liquids contaminated with PCBs and other hazardous
substances. 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 that are 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; and Appendix C, case studies of the
technology's application.
Process Description
The patented ECO LOGIC Gas-Phase Chemical Reduction
Process 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
HC1, from the reduction of chlorinated organics, such as
PCBs, and lighter hydrocarbons, such as methane and ethyl-
ene from the reduction of straight-chain and aromatic hydro-
carbons. A scrubber treats the reformed gas to remove hydro-
gen chloride and particulates. Of this gas, a portion recycles
back into the reactor; the remainder 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) was 2 m (6 ft) in diameter and 3 m (9 ft) tall, mounted
on a 15 m (45 ft) drop-deck trailer. The trailer carried a
scrubber system, a recirculation gas system, and an electrical
control center. A second trailer held a propane boiler, a waste
preheating vessel, and a waste storage tank.
ECO LOGIC designed the process to treat 4 tons/day of waste
oil, 10 tons/day of wastewater, and 25 tons/day of soil, de-
pending on the nature of the contaminants, their degree of
chlorination, and their water content. The ECO LOGIC TDU—
designed to remove most volatile, most semivolatile, and
some metallic contaminants—treats the soil. The TDU is the
subject of an independent AAR.
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
of the reactor tangentially. The tangential entry swirled the
fluids to provide effective mixing. As indicated in Figure 3,
the swirling mixture traveled downward in the annulus formed
by the reactor wall and the central ceramic-coated steel tube,
past electrically heated bars. These bars heated the mixture to
900°C (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 fine 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 HC1 removal efficiency. The scrubber 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 remainder
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 concentrated organ-
ics, 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.
-------
5H2
PCB
Benzene
+ 4HC!
Hydrogen
Chloride
Polycyclic Aromatic
Hydrocarbons
3H2
(Phenanthrene)
Benzene
+ C2H4
Ethylene
+ 9H2
Benzene
9CH4
Methane
C2H4
2H2
2CH4
cnH(2n+2)
Straight-Chain
Hydrocarbons
(n-1)H2
Methane
CH4
H2O
CO
+ 3H2
Figure 1. Gas-phase chemical reduction reactions.
Test Conditions
In preparation for the SITE Demonstration, 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. Condition 1 treated 2.9 tons of wastewater
contaminated with 3,757 ppm PCBs and 3,209 ppm perchlo-
roethylene (PCE) (tracer). Condition 3 treated 0.2 tons of
waste oil contaminated with 25.4% (254,000 ppm) PCBs and
6,203 ppm PCE.
The ECO LOGIC SITE Demonstration objectives were as
follows:
• Demonstrate at least 99.9999% ORE for PCBs.
• Demonstrate at least 99.99% destruction efficiency (DE)
for PCE in the liquid feedstock.
• Ensure that no dioxins or furans were formed.
• Characterize emissions from products of incomplete com-
bustion (PICs).
• Characterize HC1 emissions.
-------
N2-
H2-
Recirculating Gas
Propane Exhaust 900°^
Scrubt
er
35°C
Gas
Booster
Hydrocarbon
Gas
Propane
Thermal
Desorption
Figure 2. Reactor system and TDU schematic diagram.
• Document compliance with Michigan Department of Natu-
ral Resources (MDNR) air permit conditions.
• Characterize criteria air pollutant emissions.
• Document compliance with TSCA permit requirements.
• 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 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 from con-
taminated liquids and providing environmentally acceptable
air emissions.
In general, ECO LOGIC'S Reactor System effectively de-
stroyed PCBs, reducing them to lighter hydrocarbons. Theo-
retically, the destruction process could depend on both the
reactor system's gas phase reactions, which produced the
reformed gas, and on the propane/reformed gas-fired boiler, a
combustion device.
Although the result was not listed as a primary or secondary
objective for the demonstration, DREs for PCBs in the scrubbed
reformed gas were essentially equal to the DREs achieved at
the boiler stack. This shows that combustion of the reformed
gas in the boiler is not required to complete PCB destruction.
Stack emissions generally met stringent regulatory levels.
However, average benzene concentrations in the stack gas—
corrected to 7% oxygen—(Condition 1-73 u.g/dscm; Condi-
tion 3-113 u.g/dscm) and scrubber liquor (Condition 1-18.5
u.g/L; Condition 3 - 347 u.g/L) required close monitoring. The
reformed gas composition resembled coal-gas fuel. The scrub-
ber liquor required either disposal as a RCRA waste, or
recycling through the system for additional treatment. Table 1
(Executive Summary) correlates the program conclusions with
program objectives.
-------
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.
Technology Evaluation
The demonstrated ECO LOGIC Gas-Phase Chemical Reduc-
tion Process is a pilot or small commercial-scale, trailer-
mounted system, capable of treating wastewater and waste oil.
The SITE Demonstration of the reactor system consisted of
initial shakedown runs, a blank run to determine train capaci-
ties, and six liquid runs (Conditions 1 and 3). An independent
AAR discusses the TDU (Condition 2) demonstration results.
A liquid pool of waste within the Middleground Landfill
provided feedstock for the tests. Test Condition 1 treated 1.73
kg/min (totalling 2.9 tons) of wastewater containing 3,757
ppm PCBs. Condition 3 treated 0.385 kg/min (totalling 0.2
tons) of waste oil containing 25.4% (254,000 ppm) PCBs.
PCB concentrations were sufficient to calculate the DREs.
PCE added to the feedstock at levels of 3,209 and 6,203 ppm
respectively, served as a tracer to determine DEs. Additional
feedstock contaminants included fluoranthene, naphthalene,
phenanthrene, other PAHs, chlorobenzene, chlorophenol, me-
thyl chloride, tetrachlorethene, toluene, and various metals.
Based on the SITE test data—covered in detail in the TER—
analysts assessed the applicability of the ECO LOGIC Process
to the test wastes, as summarized in Table 1 and as discussed
in the following paragraphs.
-------
Organics Destruction
Dioxins and Furans
To determine the efficiency of organics destruction, EPA
evaluated DREs and DEs, benzene ring destruction, and for-
mation of dioxins, furans, and other PICs.
ORE
DRE compares the mass flow rate of selected feedstock
compounds—in this case PCBs—to their mass flow rate in the
boiler stack gas.
DRE (%) = ( 1 - Massstack / Massmput) x 100
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 LOGIC Reactor Process achieved PCB destruction
at the boiler stack ranging from 99.9999% to 99.99999%. This
met established TSCA DRE requirements, potentially quali-
fying the process for use as a PCB treatment device. (Other
TSCA requirements affecting residuals, stack emissions, and
paniculate emissions must also be considered.) The SITE
Program results, supported by ECO LOGIC'S laboratory-
scale tests and results from their Hamilton Harbor Test (Ap-
pendix C) provide evidence of acceptable PCB destruction at
the stack.
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^ / Massmput) x 100
PCE added to the feedstock acted as a tracer compound to
calculate DEs; the system achieved the target objective—
99.99%—for this tracer.
Benzene concentrations in the output streams were higher
than expected. In scale-up, ECO LOGIC must address ben-
zene concentrations in residual and effluent streams, since
high benzene concentrations can affect the costs of waste
disposal.
The DRE and DE results indicate that the ECO LOGIC
Reactor System—with boiler—can achieve RCRA hazardous
waste incinerator DEs (99.99% measured at the boiler stack)
for most organic compounds. ECO LOGIC'S lab-scale tests,
their Hamilton Harbor test on sediment contaminated with
PAHs (Appendix C), and the SITE Demonstration test of the
ECO LOGIC TDU, discussed in an independent AAR, further
support this conclusion. Assuming that scale-up factors main-
tain the same DE, a commercial-scale system would meet
RCRA emissions criteria.
The ECO LOGIC Process reduces organics in a high-tempera-
ture hydrogen environment, as opposed to combustion by
incineration in an oxygen environment. The absence of oxy-
gen inhibits formation of polychlorinated dibenzo(p)dioxin/
polychlorinated dibenzofuran (PCDD/PCDF). Although veri-
fying the reduction mechanisms inside the reactor was not an
objective of the demonstration, the test confirmed a net de-
struction at the stack of trace PCDD/PCDF in the feedstock.
Stack emissions (corrected to 7% oxygen, dry basis) ranged
from 0.156 to 0.368 ng/dscm dioxin and 0.007 to 0.011 ng/
dscm furan—results well within incineration regulatory guide-
lines. The low PCDD/PCDF stack concentrations support the
conclusion that the system can effect a net PCDD/PCDF
destruction, resulting in PCDD/PCDF stack emission concen-
trations 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. In 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 (CO), total PAHs, and benzene as indicators
of PIC/PIR formation. For the ECO LOGIC Reactor System
tests, the three indicators—THC, CO, and total PAHs—were
much lower than regulatory guidelines and well within the
MDNR permit conditions. THC averages ranged from 1.53 to
15.5 ppmv, CO from 2.3 to 23.3 ppmv, and total PAHs from
24.0 to 28.6 j^g/dscm (all corrected to 7% O2, dry basis). The
remedial manager can expect that the ECO LOGIC system
will meet anticipated permit limits for THC, CO, and PAH
emissions at other sites.
Benzene, ranging from 73 to 113 |^g/dscm, exceeded both the
MDNR permit guidelines and allowable air emission concen-
trations. A benzene ring balance, calculated at the stack,
ranged from 80 to 96% DE. This DE did not reduce benzene
concentrations to acceptable levels in the stack gas and scrub-
ber effluent. Because benzene is a major intermediate product
in the reduction of PAHs and PCBs, high benzene concentra-
tions probably formed as these high molecular weight com-
pounds degraded. Benzene is a by-product of the normal
combustion process; this may have further increased stack
emission concentrations. The remedial manager should closely
monitor benzene levels.
Air Emissions
EPA evaluated emissions of criteria air pollutants and HC1, as
well as compliance with the MDNR air permit.
10
-------
Criteria Air Pollutants
During the tests, continuous emission monitors (CEMs) mea-
sured the concentrations of the criteria air pollutants at the
stack: nitrogen oxides (NOx), sulfur dioxide (SO2), particu-
lates, THC, and CO. Each of these pollutant emission concen-
trations was low, well under the level established in the
MDNR air permit. NOx averages ranged from 60.8 to 63.5
ppmv; SO2, from 1.4 to 2.2 ppmv; particulates, from 0.17 to
0.99 mg/dscm; THC, from 1.53 to 15.5 ppmv; and CO, from
2.3 to 23.3 ppmv (all corrected to 7% O2, dry basis). 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 CO (indicators of combustion efficiency) during
low-fire operation, most notably in Condition 1, Run 1 when
the boiler was cycling between high and low fire. Future users
must be alert to the potential for decreased combustion effi-
ciency and increased emissions of criteria air pollutants dur-
ing 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.
HCI
The ECO LOGIC Reactor System reduced stack HCI emis-
sions to below the MDNR-permitted levels. RCRA emission
limits set incinerator HCI emissions at 4 Ib/hr (or less), or 99%
removal. The reactor system easily achieved this—average
stack concentrations ranged from 0.66 mg/dscm at 109 mg/hr
to 0.81 mg/dscm at 198 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. Of the 15 permit criteria, only benzene
stack concentrations exceeded the permit criteria. However,
the total quantity of emitted materials is low, probably lower
than levels that normally present health risks to exposed
populations. For future commercial units, a taller stack might
resolve this problem; greater dispersion could allow less
Table 2.
MDNR Air Permit Conditions
Parameter
Unit
Permit limit
Program average
HCI (7% 02, 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
|ig/dscm
Ib/hr
|ig/dscm
Ib/hr
%
°C
PH
Yes/No
°C
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
<10
O.04
<16
100
<0.67
<0.00034
5.5
0.0016
11.0
0.0059
0.00060
3.3E-07
65
>0.000034
(ND) O.88
4.3E-07
0
520
8.9
Yes
907
1.97
0.045
9.57
110
ND Not detected.
BQL Detected below the quantitation limit.
< Emission rate is less than the mass indicated. The mass indicated assumes that the substance is present at the detection
limit.
11
-------
restrictive stack concentrations. However, benzene emissions
could potentially exceed permit levels at other sites. Scale-up
designs should address these problems.
Intermediate and Residual Stream
Characterization
Intermediate and residual stream evaluations provided pro-
cess mass balance data; major effluent, intermediate process,
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. The waste oil was the major
waste input stream, containing the greatest mass of PCBs. The
major effluent streams were the stack gas and scrubber decant.
Most of the material in these streams entered the process
through combustion air and process water. Boiler combustion
air contributed most of the mass to the stack gas stream;
scrubber water, to the scrubber decant stream.
Table 4 shows the concentration of the major contaminants in
the intermediate and effluent streams. These data indicate the
tendency of contaminants to concentrate in the intermediate
and residual streams.
Process Mass Balance
The test objectives included a system mass balance for metals,
carbon, hydrogen, oxygen, sulfur, chlorine, and total mass.
These balances were needed to evaluate system performance
and to determine the fate of metals and other compounds in
the feedstock.
The program established a value of 0 + 50% (deviation from
perfect closure) as the quality indicator (QI) of mass balance.
Total mass balance closures ranged from -4.3 to +20.1%,
indicating that data based on process mass balances (such as
DRE, DE, and stack emission rates) can be considered very
reliable. Carbon, chlorine, hydrogen, oxygen, and sulfur mass
Table 3.
Mass Distribution of Selected Streams
Material quantity*
Stream
Input
Wastewater
Waste oil
Residual/output
Reactor grit
Scrubber sludge
Scrubber decant
Scrubber liquor
Compressed tank
condensate
Stack gas
SS1
SS2
SS11
SS12
SS13
SS22
SS15
SS16
Condition
1
0.984
0.016
0.001
0.032
1.097
0.122
N/A
0.738
Condition
3
0.781
0.219
0.001
0.172
3.659
0.005
0.002
0.980
*kg material per kg total feed.
balance closures ranged from -41 to +29.1%. Only the hydro-
gen in Condition 1 (-53.8%) and sulfur (+129%) and carbon
(+98%) in Condition 3 exceeded the QI criterion. Therefore,
the elemental mass balances further support DRE, DE, and
partitioning data reliability. Closure of the metals balances,
typically difficult to achieve in any system, ranged widely
(from -153 to +175%). However, metal balance closures are
of less concern than metals partitioning and their concentra-
tions in residual streams.
Reactor Streams
The reactor system demonstration evaluated an intermediate
stream—the reformed gas exiting the scrubber, and three
major residual streams: reactor grit; scrubber residuals con-
sisting of sludge, decant, and liquor; and stack gas emissions.
It also analyzed miscellaneous streams, and compared the
TSCA permit conditions to residual stream analyses. Stack
gas emissions have already been discussed in the section on
air emissions.
Intermediate Process Streams
Table 5 compares the reformed gas composition to several
commercially available fuels. The scrubbed reformed gas was
similar to blue water gas; its quality could be adequate to burn
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 liquify. There-
fore, storage as a gas would require excessively large tanks.
Residual principal organic hazardous constituents (POHCs)
and PIRs may affect the end use of the gas. As previously
discussed, PCB DEs measured for the scrubbed reformed gas
were essentially equal to the DEs measured at the boiler stack.
This was also true for the PCE DEs, with the exception of two
Condition 3 runs. The measured PCE DE in the reformed gas
was an order of magnitude lower than that measured at the
boiler stack. Run 1 and Run 3 achieved 99.988% and 99.97%
DEs of PCE in the reformed gas, slightly below the 99.99%
target level for this tracer compound. The DE levels demon-
strated for the chlorinated organic compounds indicate that a
commercial-scale system can achieve consistent DEs of
99.99%.
Benzene was the most prevalent PIR in the reformed gas.
Benzene concentrations ranged between 522 and 1,780 mg/
dscm. PAHs were not measured. However, as shown in Table
4, the combustion step in the boiler destroys most of the
residual benzene. PAH emissions from the boiler stack also
were low. The reformed gas is generated from a hazardous
waste, presenting a further difficulty in its utilization as a fuel
outside of the process. However, the results of the demonstra-
tion show that burning the reformed gas in combustion equip-
ment would adequately destroy any residual hazardous organ-
ics.
12
-------
Table 4.
Component Partitioning
Stream (ppb)*
Component
Total PCBs (mono-deca)
Total PAHs
Total PCDD/PCDF
Total chlorobenzenes
Total chlorophenols
Benzene
PCE
Condition
1
3
1
3
1
3
1
3
1
3
1
3
1
3
SS1
Waste-
water
22.8
25.2
ND
24.8
0.00054
1.57
ND
ND
ND
ND
3.71
8.7
21
4.3
SS2
Waste
Oil
2.38E+08
2.54E+08
320,400
366,000
327
393
253,500
235,000
ND
ND
31
ND
ND
ND
SS11
Reactor
Grit
2,160,000
3,310
655,000
846,000
0.179
162
ND
ND
ND
ND
430
17.0002
ND
ND
SS12
Scrubber
Sludge
15,490
17,665
12,800,000
40,700,000
4.03
1.92
ND
ND
ND
ND
ND
ND
ND
ND
SS13
Scrubber
Decant
203
40.6
6,640
15,200
0.00063
0.00013
ND
ND
ND
ND
7,160
7.7002
ND
41
Component
Total PCBs (mono-deca)
Total PAHs
Total PCDD/PCDF
Total chlorobenzenes
Total chlorophenols
Benzene
PCE
Condition
1
3
1
3
1
3
1
3
1
3
1
3
1
3
SS22
Scrubber
Liquor
31.5
48.9
2,697
20,300
0.0004
0.001
ND
ND
ND
ND
18.5
347
ND
7
SS14
Reformed
Gas3
2.84
32.6
N/A
N/A
0.00021
0.000162
ND
ND
ND
ND
521 ,6002
1,781,000
8.89
2,481
SS15
Tank
Condensate
N/A
16,800
N/A
6,420,000
N/A
ND
N/A
ND
N/A
ND
N/A
81 92
N/A
7.751
SS18
Heat
Exchanger
8.67
10.79
5.25
43.84
0.00021
0.00053
ND
ND
ND
ND
9.3
120
ND
ND
SS16
Stack3
0.21
1.23
28.74
24.044
0.0004
0.0002
ND
ND
ND
ND
73.1
113
3.85
4.51
1 Compound(s) detected at concentrations below the quantitative limit.
2 Compound detected at concentrations above the linear range for analysis.
3 Concentration given as |ig/dscm.
4 Essentially naphthalene.
ND Not detected.
* Averages of three runs, including NDs.
Major Residual Streams
Reactor Grit—The first test ran 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), PCB concentrations
detected in the grit ranged from 1.67 to 2,100 ppm. A conge-
ner consists of all PCB compounds having the same number
of chlorine atoms but arranged in different positions for any
individual congener compound. The grit from Condition 1
exceeded the 2 ppm (per congener) TSCA criterion. If mono-
chlorobiphenyls, dichlorobiphenyls, and nondetected conge-
ners (assumed to be present at the detection level) are in-
cluded, the grit could contain maximum PCB concentrations
between 3.3 and 2,160 ppm. These concentrations could af-
fect the 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.
13
-------
Table 5. Reformed Gas Comparison to Other Fuels
Gaseous
Fuels H2
ECO LOGIC Condition 1 41.1
reformed gas Condition 3 55.7
Blast furnace gas 1 .0
Blue water gas 47.3
Carburated water gas 40.5
Coal gas 54.6
Coke-oven gas 46.5
Natural gas (15.8% C2H6) —
Producer gas 14.0
Gaseous
Fuels
ECO LOGIC Condition 1
reformed Gas Condition 3
Blast furnace gas
Blue water gas
Carburated water gas
Coal gas
Coke-oven gas
Natural gas (15.8% C2H6)
Producer gas
Composition, percent by volume
N2
33.7
13.7
60.0
8.3
2.9
4.4
8.1
0.8
50.9
MW
16.7
11.6
29.6
16.4
18.3
12.1
13.7
18.3
24.7
02 CH4
0.05 12.2
0.06 16.9
— —
0.7 1.3
0.5 10.2
0.2 24.2
0.8 32.1
— 83.4
0.6 3.0
HHV
Btu/lbm
6,250
12,610
1,170
6,550
1 1 ,350
16,500
17,100
24,100
2,470
CO C02
5.1 6.7
8.0 3.3
27.5 11.5
37.0 5.4
34.0 3.0
10.9 3.0
6.3 2.2
28.0 4.5
HHV
Btu/scf
269
376
89
277
535
514
603
1,136
157
C2H4 C6H6
0.4 —
0.7 —
— —
— —
6.1 2.8
1.5 1.3
3.5 0.5
— —
Sp. gr.
air = 1.0
0.58
0.54
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 60°F
The grit also contained PAH levels exceeding 846 ppm,
benzene levels up to 17 ppm, and PCDD/PCDF reaching
0.162 ppm. Chlorobenzenes, chlorophenols, and PCE were
not detected.
Scrubber Residuals—The scrubber is a critical component
in the gas-phase chemical reduction process. The scrubber
effectively removes a variety of organic and metallic com-
pounds, 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 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 (40.6 to 203 ppb total) and scrubber liquor
streams (31.5 to 48.9 ppb total) met 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 of 203 ppb; the scrubber liquor could contain
PCB concentrations up to 48.9. If these streams are not
recycled through the process, they will require further treat-
ment as a RCRA waste.
The scrubber residuals did not contain detectable levels of
chlorobenzene and chlorophenols. In Condition 3, PCE was
detected at very low levels in the scrubber decant and scrubber
liquor, but not in the sludge.
The absence of chlorobenzene, chlorophenols, and the rela-
tive absence of PCE in the residual streams downstream of the
reactor provide further evidence that the ECO LOGIC Process
effectively removes chlorine from chlorinated organic com-
pounds. PAHs were the principal organic compounds detected
in the residue; benzene occurred in elevated concentrations in
the residual streams. The benzene and PAHs are most likely
PIRs resulting from the dechlorination of the PCBs. Also,
residual PCB concentrations and PCDD/PCDF concentra-
tions, although low, were present in all residue streams. The
remedial manager should evaluate concentrations of these
compounds at other sites, as they will likely be found at
detectable levels.
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.
14
-------
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 POTW imposed
stricter PCB effluent concentrations 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, accep-
tance/rejection of the scrubber 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 schedule that includes obtaining a
TSCA permit and addressing any process and operating con-
straints that the permit may impose.
Equipment and Operating Considerations
The remedial manager considering the use of the ECO LOGIC
Reactor Process should understand the function of major
process equipment components and potential operating prob-
lems associated with them.
System Components
The principal components of the ECO LOGIC Reactor Sys-
tem are the reactor, the scrubber system, the recirculating fan,
the propane-fired boiler, the liquid feed systems, and the
process instrumentation. Each of these components presented
operating problems that future users should consider.
Reactor—The reactor is the principal component of the sys-
tem. Here the combination of temperature, residence time,
feed rate, and hydrogen concentration determines the DE. The
reliability and performance of the subsystems controlling
these critical parameters affect the reformed gas quality and
the appropriate disposition of process residues and emissions.
During the demonstration, the steam flow control valve, used
to control reactor pressure, did not operate stably. Control
improved as the operators gained experience, made system
modifications, and formulated program logic adjustments.
During one run, reactor overpressurization resulted in a sys-
tem shutdown, underlining the importance of reactor pressure
control.
Scrubber system—The scrubber system is a key component
in achieving acceptable emissions. Gases exiting the reactor
first enter the spray tower leg for quenching, then pass to the
packed tower. The scrubber removes residual organics, met-
als, particulates, and chlorides—cleaning the reformed gas.
Initially, as a result of incorrect installation of internal piping,
the scrubber produced foam, affecting its efficiency. After
ECO LOGIC modified the piping, the foaming stopped and
the scrubber operated efficiently.
During processing, the pH meter did not perform satisfacto-
rily because of radio frequency interference emanating from
the recirculating heater spark plug wires. As a result, the
addition of excess caustic contributed to the scrubber foam-
ing. However, ECO LOGIC was able to manually measure
and adjust the scrubber pH, preventing any program delays.
This demonstrates the importance of relatively close scrubber
pH control.
Recirculating fan—The 5-hp recirculating fan moved the
scrubbed, recirculated, hydrogen-rich gas to the reactor inlet,
the reformed gas to compression and storage, and a reformed
gas slip stream to the boiler as supplementary fuel.
Scrubber foaming and water carryover caused excessive mois-
ture in the fan casing. Eventually this condition required fan
shutdown, cleaning, and motor winding replacement.
Boiler—The boiler provided clean steam to heat aqueous
wastes in the heat exchanger and burned a portion of the
reactor product—the reformed gas. Under normal operation,
the boiler cycles between high fire and low fire, depending on
process steam requirements. However, during Condition 1,
Run 1, the boiler, operating at low fire, emitted high spike
concentrations of THC and CO. Operation improved after
ECO LOGIC adjusted the linkage controlling the air/fuel ratio
to the boiler. However, during the remainder of the demon-
stration, ECO LOGIC vented steam to maintain boiler high-
fire. Future design considerations should address the appro-
priate sizing of the boiler and control of fuel/air ratio to
prevent excessive criteria pollutant emissions.
In Condition 3, the PCB-rich feedstock generated surplus
reformed gas, more than the boiler could process. The boiler
capacity, therefore, limited the system's throughput. To over-
come this, ECO LOGIC added a compressed gas storage tank
to the system. If, at the commercial scale, the process were
operated as a fuel producer, the boiler would not restrict
system throughput. In future operations, ECO LOGIC intends
to compress and store the surplus reformed gas for sale and
reuse. However, the remedial manager should address this
report's earlier cautions concerning storage capacity and sal-
ability.
Liquid feed systems—The ECO LOGIC Reactor System had
separate feed systems for organic liquid feed and aqueous
liquid feed. The organic liquid feed system consisted of a feed
tank and a feed pump. The aqueous liquid feed system con-
sisted of a feed tank, a feed pump, and a heat exchanger. The
aqueous liquid feed pump operated unstably, requiring fre-
quent adjustment. ECO LOGIC should undertake further de-
sign work to improve the pump's reliability.
Process monitoring—The oxygen analyzer did not operate
reliably. This is an important consideration since elevated
levels of O2 in the system can create an explosive atmosphere.
Apparently, blockages in the analyzer sampling line caused
the problem. Future configurations of this critical system
should address this deficiency.
A differential pressure transmitter and a magnehelic gauge
control the hydrogen content in the system, ensuring suffi-
15
-------
cient hydrogen to reduce (destroy) the organics. Insufficient
hydrogen content can slow the reaction kinetics, causing
incomplete reduction. During the demonstration, the lines to
the pressure sensors plugged, resulting in an insufficient hy-
drogen content and generating oily residue that coated equip-
ment and instrumentation, further affecting system opera-
tions. ECO LOGIC should consider a design change to im-
prove instrument reliability.
System Reliability
The program evaluated system reliability during processing
and during a 72-hour uninterrupted test. The reliability has
been expressed in terms of planned availability compared to
actual on-line availability. The number of days planned for the
entire demonstration was 10 (reactor and TDU/reactor tests);
the program actually took 42. This translates to a 24% equip-
ment availability.
In addition, the program evaluated actual waste throughput as
a percent of the planned input—a throughput reliability. The
wastewater test was designed to treat 8 tons of material, but
processed 2.9 tons. The waste oil test nearly achieved the
planned throughput of 0.8 tons. The resulting throughput
reliability percentages varied between 20% and 55% over the
six runs. However, during the 72-hour continuous operation
using liquid feedstock, the system operated without interrup-
tion.
Scale-up Parameters
One program objective sought to identify the critical scale-up
parameters. Knowing these parameters assists future users in
evaluating a proposed commercial-size operation. This report
has addressed scale-up considerations as they pertain to the
immediate discussion.
CIMS Validation
The CIMS is the primary process control unit of the ECO
LOGIC Process. It records and stores data. It measures se-
lected compounds and their decomposition products to maxi-
mize organic destruction.
Demonstration results show that the CIMS may reflect data
trends useful for process control, but it is not, at this stage of
its development, a reliable source of quantitative data. Further
testing will determine whether the CIMS can provide ad-
equate 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.
Table 6.
Summary of Reactor Operating Conditions
Test condition
averages
Equipment
Reactor
Scrubber
Parameter
Temperature (°C)
Pressure (in. H2O)
Residence time (sec)
Inlet temperature (°C)
Outlet temperature (°C)
Water pH
1
892
1.8
8
546
33
8.78
3
933
1.8
6.1
527
32
9.32
Recirculating fan
Vaporizer
Differential pressure
(in. H20)
Flow rate (cfm)
Gas pressure (in. H2O)
Temperature (°C)
Pressure (psi)
11.6
110
6.5
148.3
51.8
7.8
110
0.66
149
51.4
Technology Applicability
This section describes the applicability of the technology to
the site, waste media, safety, and staffing.
Site 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 sits on two mobile
trailers.
Cold-weather operations may inhibit efficient destruction be-
cause of the incremental amount of energy required to heat the
reactor. In addition, feedstock liquids would require melting
prior to treatment, and liquid residuals could freeze in the
unheated storage tanks. Winterization, including heat tracing,
is necessary to provide adequate feedstock and to ensure
uninterrupted processing.
Applicable Media
Initially, ECO LOGIC designed the reactor system to process
liquids, with soil processing limited to about 30% solids1.
ECO LOGIC added the TDU to gain greater feedstock pro-
cessing capabilities. As explained in an independent AAR, the
demonstration indicated that the TDU requires further devel-
opment.
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.
16
-------
Safety Considerations
Process Safety System
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
1926] addressing such topics as slips, trips, and falls; confined
space entry; contingency planning; etc. The regulations in 29
CFR 1910.120 address PPE. High voltage electrical equip-
ment standards are also a concern.
Chemical Use
The chemical hazards of the ECO LOGIC Process accompany
the use of propane, liquified 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
usage 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.
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 infiltration 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.
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 fan, scrubber pump failure, 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. Reformed gas flow to the boiler stops. Either
an operator or an automatic computerized process controller
initiates these events.
Staffing Issues
The CIMS facilitates monitoring and remote adjustment of
process parameters. This reduces labor requirements for moni-
toring and maintenance personnel. The monitoring personnel
must be capable of evaluating system problems and directing
maintenance personnel in problem resolution. Since opera-
tions can be controlled remotely, only those personnel need-
ing to manually adjust or maintain the system components
require personal protective equipment. Since the system will
be processing hazardous substances, the medical monitoring,
training, and personal protection requirements of 29 CFR
1910.120 will remain in effect.
Regulatory Considerations
Several pieces of federal legislation and any state or local laws
present compliance considerations in operating the ECO
LOGIC Reactor 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
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-
17
-------
elude 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.
The ECO LOGIC Reactor Process satisfies the SARA man-
date to reduce the toxicity, mobility, and volume of hazardous
substances by reducing organic contaminants in the feed-
stock—such as PCBs—to lighter, nontoxic hydrocarbons,
such as methane and ethylene. The demonstration showed that
the reactor system destroyed more than 99.99% (DE) of the
contaminants, illustrating both long-term and short-term ef-
fectiveness with respect to organic compounds. It indicated
that metals were mainly concentrated in the scrubber effluent,
which required additional treatment prior to disposal. EPA
cost estimates are found in Section 4.
The system appears implementable as currently designed.
Relatively mobile, it requires water and electric utilities;
hydrogen, oxygen, nitrogen, and propane storage; and front-
end material handling equipment to deliver feedstock to the
feed storage tanks.
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 specific waste feed and the effectiveness of
the treatment, the ECO LOGIC Reactor Process generates two
potentially hazardous waste streams: the scrubber liquor and
the treated soil. To generate these wastes, the remedial man-
ager must obtain an EPA generator identification number and
either comply with generator accumulation and storage re-
quirements 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 haz-
ardous waste manifest must accompany off-site waste ship-
ment; transport must comply with Federal Department of
Transportation (DOT) hazardous waste transportation regula-
tions. 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 and 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 may require meeting state standards that are more
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.
2. Kalyanam, K. M., and Hay, D. R., Safety Guide for
Hydrogen, National Research Council of Canada, Ot-
tawa, Ontario, 1987.
3. U.S. EPA. CERCLA Compliance with Other Laws Manual
Part II: Clean Air Act and Other Environmental Statutes
and State Requirements, Interim Final, EPA/540/G-89/
009, OSWER, Washington, D.C., August 1989.
18
-------
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 Reactor System
conducted at the Middleground Landfill treated both PCB-
contaminated wastewater and waste oil. This economic analy-
sis is an extrapolation of that experience based on the com-
mercial use of a system similar to that employed during the
demonstration program. For the purposes of this analysis it
was assumed that 100,000 gallons of wastewater and 30,000
gallons of waste oil were stockpiled for treatment. The waste
streams are assumed to be identical in composition to those
treated during the demonstration program. The following
feedrates were utilized forthis analysis: 1.73 kg/min of waste-
water and 0.485 kg/min of waste oil, simultaneously injected
into the reactor. Since the process could experience some
downtime, a sensitivity analysis presents three different on-
line utilization factors: 60%, 70%, and 80%. Certain cost
elements were fixed; others were time-sensitive.
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 (U.S. dollars per ton) of material processed.
Conclusions
The data showed the ECO LOGIC Reactor Process to be an
acceptable remedial alternative for liquids contaminated with
PCBs. Since the process was effective in treating the PCB-
contaminated Middleground Landfill liquids, it should be
applicable to the remediation of other similar sites.
The treatment costs (1994 U.S. dollars) ranged from a low of
$l,670/ton to a high of $2,000/ton, depending on the utiliza-
tion factor. Because of limited data, the cost estimates pre-
sented in this analysis may range in accuracy from +50% to -
30%, an order of magnitude guideline suggested by the Ameri-
can Association of Cost Engineers.
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
• Three tons of liquid 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.
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
site-specific characteristics may improve or worsen the project
economics:
• 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 (dollars/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.
19
-------
Coste Excluded from the Estimate
Labor
Although the SITE Program provides a 12-item 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.
Utilities
Electrical power was required for the operation of the various
pumps, blowers, feeders/conveyors, and electric heating ele-
ments, in addition to instrumentation, lighting, and miscella-
neous power outlets. The total system demand at full effi-
ciency averaged 30 kW.
Scrubber make-up water requirements were minimal; actual
volume used was not available. For calculations, the addition
of 178 kg/hr of water (about 50 gal/hr) was assumed.
The recirculating gas heat exchanger and boiler needed a
natural gas source along with required piping and appurte-
nances. During the demonstration, propane fuel was used. For
Condition 1, the propane consumption rate was 7.62 kg/hr; for
Condition 3, 12.9 kg/hr.
With the exception of propane, the analysis assumed that all
utilities, in appropriate capacities, were available at the site.
Supplies
Table 7 shows the types and quantities of consumable sup-
plies required by the ECO LOGIC Reactor Process.
Operating Conditions
This analysis assumed that the facility would operate 24 hours
a day, seven days a week. At the throughput rates discussed
earlier, the required operating times were calculated with the
three different utilization factors, as follows:
250 days (60% utilization)
214 days (70% utilization)
188 days (80% utilization)
These periods excluded mobilization, shakedown, start-up,
and demobilization times.
Table 7
Consumables Required by the ECO LOGIC
Reactor Process
Item
Caustic
Hydrogen
Propane
Nitrogen
Measure
kg/hr
kg/hr
kg/hr
m3
Condition 1
(Water)
24.7
0.138
7.62
15
Condition 3
(Oil)
116.7
0.072
12.9
15
Four crews, consisting of a shift supervisor and two techni-
cians, would provide coverage for a 24-hour, seven-day,
three-shift reactor operation. The project engineer/manager
would work Monday to Friday during the day shift; a part-
time clerk, the same schedule. The first and second shifts
would require two technicians. The third shift (midnight to
8:00 am) would require only one.
The Project Engineer/Manager would hire all non-union local
workers. Table 8 lists the labor classification, number of
workers, and unit labor rates used in the forecasts. Table 9
shows the data totals based on utilization percentage.
The estimates excluded costs for OSHA training time, medi-
cal screening for all personnel on-site, and operations training.
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
Some of the cost categories above do not apply to this analysis
because they are site-specific, project-specific, or the obliga-
tion of 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 particular projects.
Site Preparation Costs
The extent of preparation depends on the specific site charac-
teristics. Such activities include site design, layout, surveys,
acquisition of access rights, establishment of support and
decontamination facilities, and utility connections.
20
-------
Table 8. Operating Labor*
Category Hire Number
Engineering Direct 1
Shift supervisor Local 1
Technicians Local 2**
Clerk (part-time) Local 1
Subtotal
Time/labor for engineer
Total per week
* 24 hours per day, 7 days per week operation.
** 4 shifts (shift 3 - only 1 technician).
Table 9. Labor Costs Based on Utilization
Utilization (%) Cost/week Weeks Labor cost
60 $16,600 34 $564,400
70 16,600 30 498,000
80 1 6,600 26 431 ,600
Despite the fact that most of these activities are site-specific,
they represent a typical percentage of the overall cost that can
be expected on any project. Therefore, they have been in-
cluded in the cost analysis.
The analysis excluded site engineering, work plan prepara-
tion, and pretreatment of hazardous waste feed. Table 10 lists
the cost elements associated with site preparation for the ECO
LOGIC SITE Demonstration.
Permitting and Regulatory Costs
Permitting and regulatory costs are generally the obligation of
the responsible party or site owner. These costs may cover
actual permit application, monitoring, and the development of
Shifts
1
4
4
1
Table
Item
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Total
Hr/wk
40
160
280
20
10. Demonstration
Description
Site fencing (chain link)
Snow fencing (wood)
Access roads
Gravel and stone
Concrete pads
HOPE liner
Temporary piping
Temporary electric
Telephone
Sump pumps
Security
Signs, etc.
All-in
cost/hr
$40.00
34.00
30.00
20.00
Site Preparation Costs
Cost
$2,500
100
12,000
2,000
4,500
3,500
1,300
2,000
500
1,000
8,000
1,000
Total
per week
$1,600
5,440
8,400
400
1 5,840
760
$16,600
Personal protective equipment 2,000
Engineering support
Administrative support
Site supervision
Travel and living
Miscellaneous
5,000
5,000
60,000
12,000
5,000
$127,400
tory costs vary greatly because they are specific to the site,
waste, and technology. Therefore, no permitting and regula-
tory costs have been included in this analysis. Depending on
the treatment site, however, they could be a significant factor,
since such activities can be both expensive and time-consum-
ing.
Capital Equipment
This cost category includes all equipment provided by the
technology developer; it generally encompasses equipment
integral to the process. For this analysis, holding tanks and
incidental equipment have been relegated to other categories.
Table 11 provides a breakdown of the reactor capital equip-
ment costs. Comparable costs associated with the TDU/reac-
tor combination are addressed in the complementary TDU
AAR.
Prices for the various pieces of equipment were obtained from
vendor catalogs, Richardson's cost estimating handbooks, and
historical data. General specifications were provided by the
developer. The figures excluded all research and development
costs. No license fees or royalties have been included.
In terms of size and throughput capacity, the actual equipment
used for the demonstration was also used for the analysis.
The operational duration of the project in the forecast is less
then 1 year. Therefore, the equipment costs have been annual-
ized based on the following formula:
A = C
i (1 + i)n
A = annualized cost, $
C = capitalized cost, $
i = interest rate, 6%
n = useful life, 10 years
21
-------
Table
Item
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Total
11. Capital Equipment for Commercial
Description
Reductive heater
Heat exchangers (2)
Reactor
Scrubber
Recirculating fan
Gas storage vessel
Boiler
Pumps
Sludge/oil tank
Wastewater tank
Lowboy trailers
Ductwork
Electrical system
Control system
Miscellaneous pipes and valves
Miscellaneous structures and supports
Instrumentation
Electrical bulks
Miscellaneous process items
Operation
Total
$20,000
60,000
74,000
36,000
10,000
10,000
113,000
19,000
10,000
10,000
55,000
5,000
21,000
40,000
32,000
5,000
32,000
1 1 ,000
22,000
$585,000
Mobilization and Start-up
Mobilization includes the setup of the work site including
trailers, utilities, and miscellaneous materials, as well as the
transportation and assembly of the process equipment. Table
12 breaks down the significant cost items associated with this
category.
Transportation costs were calculated on the basis of 500 one-
way road miles at an average of $2.00 per mile. One-way
miles were used because it was assumed that the equipment
travels from active site to active site.
The process equipment consisted of two pre-assembled flat-
bed trailers. A separate allowance covered final hook-up at the
site, while shakedown costs comprised part of the allowance
for start-up (Table 13).
Transportation costs for personnel have been included in the
Labor category.
Taxes and insurance were calculated as 5% of capital equip-
ment. An allowance for working capital, equivalent to ap-
proximately one month's inventory of supplies, has been
included ($9,000).
Start-up costs included labor for a five-man crew during one
60-hour week, plus an allowance for consumables and miscel-
laneous. The cost analysis did not provide for a separate
contingency; however, start-up costs included a 10% contin-
gency.
Operations Labor
Personnel requirements for operating the facility under vari-
ous conditions have already been discussed. They included
wages and travel expenses for on-site operations personnel.
Fully burdened wage rates were shown by labor category in
Table 8. It was assumed that all personnel would be local hires
except the project engineer/manager. Per diem for the project
manager—lodging, food, and rental car for a seven-day-
week—has been calculated at $110/day.
Supplies
This cost category, consisting of chemicals and fuels, was
based on consumption rates for the various operating condi-
tions. Northeastern area vendors provided verbal quotes with
no consideration for large bulk quantity or unusual transporta-
tion. (Chemicals and fuels could be purchased locally for
approximately the same price.)
Supplies also encompassed disposable personal protective
equipment (PPE) for Level D. The category also included a
$5,000 allowance for unspecified consumables and spares.
Utilities
Electrical usage, make-up water consumption, and telephone
charges comprised the utilities category. Electrical demand in
kilowatt hours was extrapolated from actual demonstration
experience at $0.08/kWh. Make-up water was calculated at
approximately 50 gal/day and $0.05/gal. Telephone charges
were set at $300/month.
All utilities were assumed to be available at the site. However,
costs excluded installation, hook-up, etc., which were covered
under Mobilization.
Effluents
There were no costs associated with effluents in this analysis
since no material would be introduced into normal effluent
streams.
Residuals
Residuals generated by this process would include grit and
fines that would be stored in drums and transported to an
approved disposal site. This category also included the trans-
portation and disposal of PPE stored in drums. The process
generated approximately 0.071 kg/hr of grit, slightly more
then 500 Ibs for this application.
Analytical
No analytical costs have been included in this cost estimate.
The client could elect (or might be required by local authori-
ties) to initiate a sampling and analytical program to meet
local regulatory criteria. These analytical requirements could
significantly affect costs.
22
-------
Table 12. Mobilization/Start-Up Costs
Description Cost/Month ($)
Fixed costs
Delivery /blocking trailers
Trailer furnishings
Hooking up process equipment
Storage tanks and vessels
Drums and pails
Crane rentals, etc.
Monthly costs
Trailers (5) 500
Portable toilets (2) 150
Dumpsters 150
Job vehicles 1 ,000
60%
$1,550
2,000
15,000
1 1 ,000
2,600
1,400
$4,000
1,200
1,200
8,000
Utilization
70%
$1,550
2,000
15,000
1 1 ,000
2,600
1,400
$3,500
1,050
1,050
7,000
80%
$1,550
2,000
15,000
1 1 ,000
2,600
1,400
$3,000
900
900
6,000
Table 13.
Cost Allocations
Description
Transportation
Working capital
Insurance
Start-up (4,000)
60%
$1 ,000
9,000
32,000
20,000
Utilization
70%
$1 ,000
9,000
28,000
20,000
80%
$1,000
9,000
24,000
20,000
Repairs and Maintenance
Maintenance labor and material costs vary with the nature of
the waste, the performance of the equipment, and the site
conditions. For estimating purposes, roughly $500/mo has
been allowed. This represents approximately 10% of capital
equipment.
The key maintenance items associated with the ECO LOGIC
Process are the electrically heated bars in the reactor. The
anticipated life span, under the operating conditions described,
has not yet been defined.
Demobilization
Demobilization costs were limited to disassembly, site cleanup,
and limited restoration. Disassembly covered the following:
disconnection of equipment and utilities, surface decontami-
nation (for transportation off-site) of all process equipment,
and loading. Transportation to the next destination was not
included.
Site restoration included the removal of all utilities, trailers,
and rental equipment. Requirements regarding permanent fenc-
ing, grading, landscaping, etc., vary by site. Depending on the
future use of the site, they were assumed to be the obligation
of the site owner or the responsible party. They were not
included in this analysis.
Results of Economic Analysis
Table 14 presents the total treatment cost for the reactor
system. The table was organized in accordance with the 12
EPA cost categories. In addition to total treatment costs, a unit
cost (dollars/ton) has been provided. In an effort to address
unforeseen job conditions, a range of costs for 60%, 70%, and
80% utilization factors has been calculated.
The largest single cost component of this treatment technol-
ogy was operational labor—accounting for 52% of the total
treatment cost at 80% utilization. Supplies accounted for 12%
of the total, while site preparation made up 15%, and mobili-
zation/start-up, 12%. The remaining eight categories com-
prised only 9% of the total treatment cost, with three having
no cost associated with them for this SITE project analysis.
Considering the effect of the labor component on price and
the relative constancy in scale-up of the other components, it
is likely that unit costs would benefit significantly from
commercial scale-up. Increasing equipment capacity would
decrease process time and labor cost.
Table 15 compares the costs per ton for the actual test through-
puts with the costs estimated for targeted throughputs. If
targeted throughputs had been achieved, costs per ton would
have been substantially lower. A commercial-scale unit would
further decrease these figures.
References
1. Richardson Engineering Services. Cost Estimating Guide,
Vol 1, 1993 edition.
2. 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.
23
-------
Table 14. Economic Analysis for the ECO LOGIC Reactor System
Activity
Site preparation
Capital equipment
Start-up/mobilization
Labor
Supplies
Utilities
Residuals
Maintenance costs
Demobilization
Totals
Costs
Table 15. Cost Extrapolations
At actual
throughput
60% 2,000
70% 1 ,850
80% 1 ,670
60%
(250 days)
$127,400
50,400
109,950
564,000
110,000
10,500
2,500
4,000
20,000
$998,750
$2,000/ton
Cost, $/ton
At targeted
throughput
670
620
550
Utilization
70%
(21 4 days)
$127,400
44,700
104,150
498,000
106,000
10,500
2,500
3,500
20,000
$916,750
$1,850/ton
80%
(188 days)
$127,400
37,800
98,350
431 ,000
103,000
10,500
2,500
3,000
20,000
$833,550
$1,670/ton
24
-------
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 comprising
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, volatile organic
compounds (VOCs), 13 trace metals, HC1, O2, CO2, CO, SO2,
NOx, THC, and other selected compounds. Tables A-3, A-4,
and A-5 list the sampling and analysis methods used by EPA.
The demonstration plan and TER contain further details about
the Sampling and Analysis Program.
Clean
Soil
Quench (SS24
Water
SS15J
Tank Condensate
Figure A-1. Sampling and monitoring stations.
25
-------
Table A-1. EPA Sample Locations
Stream
SS1
SS2
SS3
SS4
SS5
SS6
SS7
SS9
SS10
SS11
SS12
SS13
SS14
SS15
SS16
SS18
SS19
SS20
SS22
SS24
Description
Waste water
Waste oil
Contaminated soil
Caustic soda
Scrubber make-up water
Propane
Hydrogen
Combustion air
Treated soil
Reactor grit
Scrubber sludge
Scrubber decant
Reformed gas
Tank condensate
Stack gas
Heat exchanger
TDU gas
TDU molten bath
Scrubber liquor
Quench water
Location
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
Pressure
Flow rate
Feed rate
PH
Gas constituents
Stations
2, 3, 4, 5, 6, 7, 9,
11, 12, 13, 15, 16,
17, 18
12, 13, 16,
1,4,7
7, 10
7, 10
13
8
13
14
5
7
Frequency
Continuous
Continuous
Continuous
1/2 hour
Continuous
Continuous
Hourly
Hourly
1/2 hour
Continuous
Continuous
Method
Thermocouple
Pressure transmitter
Differential pressure transmitter
Gauge
Differential pressure transmitter
Vortex flow meter
Orifice meter
Vortex flow meter
Tracer injection
pH meter
O2 analyzer; CIMS
26
-------
Table A-3. Flue Gas Sampling and Analytical Methods
Sampling
Analyte Principle Reference
PCBs
Dioxins/furans
PAHs
CB/CP
Volatile organics
Metals
HCI
Participates
N0x
S02
°2
C02
CO
THC
Fixed gases
Sulfur compounds
Heating value
XAD-2
XAD-2
XAD-2
XAD-2
Tenax
Impinger
Impinger
Filter
OEMS
OEMS
OEMS
OEMS
OEMS
OEMS
Tedlar bag
Tedlar bag
Tedlar bag
Method 001 0*
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
HR GC/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 8270*
EPA 8270*
EPA 5041*
EPA 29 (draft)
EPA 26**
EPA 5**
EPA 7E**
EPA 6C**
EPA 3A**
EPA 3A**
EPA 10**
EPA 25A**
MASA 1 33***
EPA 15**
ASTM 2620M
Test Methods for Evaluating Solid Wastes, SW-846, U.S. EPA (November 1986, 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, 1989.
Table A-4. Solids Sampling and Analytical Methods*
Analytical
Analyte Principle Reference
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
GC/MS
HR GC/HR MS
GC/MS
GC/MS
GC/MS
CVAAS, AAS, ICAP
1C
1C
Colorimetric
Gravimetric
GC/MS
CVAAS, ICAP
Combustion/gravimetric
Bomb calorimeter
Combustion
GC
Hydrometer
EPA 680*
EPA 8290*
EPA 8270*
EPA 8270*
EPA 8260*
EPA 6010, 7471*
EPA 9020*
ASTM E776
EPA 71 96*
ASTM D31 77
EPA 8240*
EPA 60 10, 7470*
ASTM D482
ASTM D240
ASTM D31 76
EPA 9060*
ASTM D1298
* 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 1986.
27
-------
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
Colorimetric
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 71 96*
EPA 6010*
EPA 8240*
EPA 60 10, 7470*
EPA 160.4
ASTM D240
ASTM D31 76
EPA 9060*
ASTM D1298
EP A9040*
* 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 1986.
28
-------
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
commercial operations, and improving the design of the soil/
sediment processing unit. These advances along with relevant
background information are described herein.
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 Defense.
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 EPA's 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,
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.
Process Chemistry
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 HC1, while nonchlorinated organic contaminants,
such as PAHs, are reduced substantially to methane and minor
amounts of other light hydrocarbons. The HC1 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 DE.
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
wastes to accelerate liquid vaporization. The gas mixture
29
-------
Cl
Cl
5H2
ci
ci
O
+ 4HCI
PCB molecule and hydrogen react to produce benzene
and hydrogen chloride
23 H2
+ 4HCI + 2H2O
14CH4
9H2
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
H2
n CH4
Hydrocarbons and hydrogen react to produce methane
CH4 + H2O
Water Shift Reactions
CO + 3H2
Methane and water react to produce carbon monoxide
and hydrogen
CO + H2O
CO2 + H2
Carbon monoxide and water react to produce carbon
dioxide and hydrogen
Figure B-1.
ECO LOGIC process reactions.
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 all 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 hexachloro-
benzene (HCB). Significant 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 include decreases in worker exposures
30
-------
To Scrubber System
Waste Injection Ports
Reactor Steel Wai!
6" Thick 'Pyro-Bloc' "Y"
Ceramic Fibre Insulation
Stainless Steel Inner Lining
Radiant Tubes
Stainless Steel Central Tube
To Grit Box
Figure B-2. Commercial-scale process reactor.
31
-------
Recirculation Gas
Reformer I Gas
Fuel Heater
Contaminated Fuel
Soil Exhaust
Electrical
Energy
Thermal
Desorption
Mill
Desorbed
Gases
Fuel for Boiler
Steam Reformer
TDM
Cleaned Soil
Desorbed Gases
Contaminated
I Equipment,
L
Hot Watery Waste
SBV
Gas
Heater
Cleaned Equipment
-S. *~*t40
r^ Storage
Fuel
Figure B-3. Commercial-scale process unit schematic.
and fugitive emissions from dram transfer operations, clean-
ing of the drams in place, and segregation of inorganic con-
taminants into the existing drams. 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
HC1, water, heat, fine particulates, aromatic compounds and
carbon dioxide. The first stage of the scrubber can be operated
to recover medium-strength HC1, which avoids the cost of
neutralization with caustic. The cost saving can be consider-
able 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 HC1 by caustic packed tower scrubbing. Particu-
late 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 and 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 in. water gauge (0.36 psi) of atmospheric pressure.
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.
Excess water is also filtered and carbon-treated to remove any
32
-------
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 CIMS 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 DE. The com-
pounds selected for monitoring depend on the waste being
processed. For example, during PCB processing,
monochlorobenzene is typically monitored as an indicator of
DE. Low levels of this volatile compound indicate that de-
struction of the PCBs is proceeding to completion. 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 continuous 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. 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 analysed in the boiler stack
emissions.1 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-contami-
nated 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 forerun-
ner of the current TDM, and demonstrated 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 AAR.
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
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
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.
(M/kg)
483
680
423
ND
Grit
Cone.
(mg/kg)
1.67
7.76
0.37
ND
Sludge
Cone.
(mg/kg)
32.8
56.1
4.3
ND
Stack
Gas Cone.
(|ig/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
33
-------
Table B-2. U.S. EPA SITE Program Results
Water/Oil and High-Strength Oil Tests
Run
1
2
3
4
5
6
Soil
Run
1
2
Waste Type
Water/Oil
Tracer
Water/Oil
Tracer
Water/Oil
Tracer
Oil
Tracer
Oil
Tracer
Oil
Tracer
Tests
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
1 2,400
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
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 HCB was also
contaminated with significant levels of octachlorodibenzo-p-
dioxin (OCDD). The desorption efficiencies achieved for the
HCB and OCDD for Test 2 were 99.99% and 99.8%, respec-
tively. 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 independent of the destruction
process. The reactor destruction efficiencies 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 efficiencies of
at least 99.9999%. The ability to compress and store the
product gases generated during processing means that no
uncontrolled air emissions occur.
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-
34
-------
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 SB V 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.
Table B-3. Summary of Test Results from the
Thermal
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)
Desorption Mill
Waste PCB
Concentration
(ppm)
39
440
520
710
790
750
7.3
8.3
8.3
420
420
2000
1200
8.3
Lab-Scale
Grit PCB
Concentration
(ppm)
0.011
0.0039
0.0016
0.028
0.0097
0.065
0.0029
0.0066
0.0013
0.0017
0.012
0.044
ND (0.011)
ND (0.005)
References
1 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.
35
-------
Appendix C
Case Studies
Introduction
Two case studies illustrate the use and performance of the
ECO LOGIC Gas-Phase Chemical Reduction Process.
Case Study C-1:
Bench-Scale Demonstration on
Contaminated Harbor Sediment
Introduction
Environment Canada asked for a series of laboratory tests on
harbor sediment wastes prior to funding a pilot-scale unit. The
Canadian Contaminated Sediment Treatment Technology Pro-
gram provided funding for the tests.
Description
ECO LOGIC designed the 3 kg/hr reactor system to mimic the
operation of the pilot-scale field demonstration unit, to pro-
vide DE data, and to develop the process control and continu-
ous monitoring systems for the pilot-scale work.
As shown in Figure C-1, the reactor (LS) was a single cylin-
drical chamber with a 12-in. diameter and 72-in. length elec-
trically heated by glo-bars passing through the central axis.
The insulated reactor contained a relatively cool area (G)
where solids collected after passing through the reaction zone.
Thermocouples—at three locations inside and outside the
inner stainless steel liner (T1-T6)—measured temperatures.
Liquid waste (L) and hydrogen (H2) flowed into the reactor at
known, metered rates.
As the gases and fine particulates left the reactor, the CIMS
drew a small sidestream; the remainder of the gas flowed to
the first condensation flask. This flask (S) simulated the
scrubber in the pilot system. After the first knockout flask,
most of the gas flow passed through a heat exchanger tube
(HX), condensing the rest of the water in the second knockout
flask (KO). A valved pump (P) and rotameter (R) drew some
of the gas through an XAD2 resin trap cartridge (X). The
remaining gas was vented (V). Analyses of the scrubber flask
water, the knockout flask water, and the XAD2 resin deter-
mined the reactor's DE.
A second sidestream, drawn from the main stream immedi-
ately after the scrubber flask (S) passed through a quartz tube
furnace (Q), along with air for combustion. This stream simu-
lated the DE obtained by using the boiler and reactor combi-
nation. After drying in a water knockout flask, the gas stream
passed through an XAD2 resin tube to a valved pump and
rotameter.
Monitored process parameters included the hydrogen flow
rate, reactor pressure, reactor temperatures, boiler tempera-
ture, scrubber flask temperature, knockout flask temperature,
and quartz oven temperature. The CIMS also monitored and
recorded concentrations of 10 organic compounds.
Testing Protocol
Each run processed about 5 liters of sediment over a period of
several hours. Environment Canada provided eight sediment
samples: four from Hamilton Harbor, two from Sheboygan
Harbor, two from Thunder Bay Harbor, and two from Hamilton
Harbor that were subsequently spiked with trichlorobenzene
(Table C-1). ECO LOGIC performed analysis on the samples,
except for the metals analyses, which were done by XRAL
Environmental.
Half of the samples were split for duplicate analysis by the
Wastewater Technology Center Laboratory, whose personnel
also observed most of the test runs. For two test runs, the
laboratory analyzed samples for dioxins, furans, PCBs, PAHs,
organo-chlorines, base neutrals, chlorobenzenes, chlorophe-
nols, and metals; for the other runs, only target compounds.
To begin the test, the operator charged a measured amount of
well mixed sediment to the waste flask (boiler). After the test,
the operator emptied the flask and recirculation pump and
then flushed them with a measured volume of water. The test
operator combined some of the exit sample extracts prior to
analysis. Table C-2 lists the final five output samples.
Data Summary
Table C-3 summarizes the results of the ten test runs requested
by Environment Canada. Runs 1 and 2 processed Hamilton
Harbor sediment diluted to about 4% solids. Both tests ob-
tained a 99.99% DRE. Run 1 samples received full analysis;
there were no dioxins or furans in any of the samples, includ-
ing the waste. Runs 1 and 2 achieved a solids reduction of
about 80%; the remaining grit contained no PAHs. A metals
analysis of the grit from Run 1 revealed sodium, manganese,
36
-------
Air
KO
Figure C-1. Bench-scale reactor system schematic diagram.
37
-------
Table C-1.
Run
1
2
3
4
5
6
7
8
9
10
Table C-3.
Run
1
2
3 PAHs
CB
4 PAHs
CB
5
6
7
8
9
10
Sediment Samples
Input
Source Analysis
Hamilton Full
Hamilton Target
Ham/TCB Target
Ham/TCB Target
Hamilton Target
Sheboygan Target
Thunder Bay Target
Hamilton Full
Sheboygan Target
Thunder Bay Target
Table C-2.
Sample
Reactor grit
Scrubber catch
Scrubber exit
Scrubber exit
Incinerator exit
Performance Results
ORE*
99.9939
99.9960
99.9980
99.9990
99.9944
100.0000
99.991 1
99.9990
100.0000
99.9836
99.9941
99.9960
Output
Analysis
Full
Target
Target
Target
Target
Target
Target
Full
Target
Target
Residue Streams
Type
Solids
Solids
Liquid
Solids
Liquid
Gas
Liquid
Gas
DE
67.9
85.2
61.3
99.9954
81.6
99.9999
-150.2
99.4
100.0
-1.1
99.8
96.8
Target WTC Lab
Compound Duplic.
N/A No
PAHs Yes
PAHs/CBs Yes
PAHs/CBs No
PAHs No
PCBs No
CPs Yes
N/A Yes
PCBs Yes
CPs No
Component Source
Reactor
Scrubber flask, lines
Scrubber flask
Heat exchanger, KO flask, lines
Heat exchanger, KO flask, lines
XAD2 resin
Knockout flask
XAD2 resin
Recirc. Solid
Rate Content
98.4 4.2
98.4 3.6
98.4 3.1
98.4 3.2
98.4 3.0
98.4 3.0
98.4 17.6
98.4 3.0
98.4 3.5
98.4 8.0
Mass
Balance
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Solids
Reduction
87.4
76.7
49.3
70.9
9.7
2.7
32.6
7.0
25.8
43.4
* DREs based on total organics fed and PAHs analyzed in the stack.
38
-------
phosphorous, titanium, copper, and lead. ECO LOGIC sug-
gested that some of the metals could be artifacts from the
reactor's stainless steel inner liner because concentrations of
manganese, titanium, copper, and lead in the waste were low.
The primary metals in the waste included iron, calcium,
sulfur, aluminum, magnesium, sodium, and phosphorous.
Runs 3 and 4 processed Hamilton Harbor sediment that was
diluted to about 3% solids and spiked with approximately
1,000 ppm trichlorobenzene. These runs obtained a 99.99%
ORE for the PAHs; for trichlorobenzene, 99.999% and
100.0000%, respectively. Solids reduction from destruction
of organic materials averaged 60%.
Runs 6 and 9 processed Sheboygan Harbor sediment con-
taminated with PCBs. The waste was diluted to about 3%
solids. The resultant PCB concentration in the feed ranged
from 5 to 7 ppm. The runs achieved DREs of 99.999% and
99.99%; solids reduction averaged 15%.
Runs 7 and 10 processed Thunder Bay sediments contami-
nated with chlorophenols. ECO LOGIC reasoned that the
sample matrix of the waste may have caused analytical
procedure problems; the values obtained did not match ECO
LOGIC'S expectations. There were no problems in analyzing
the other samples; ECO LOGIC reported DREs of 100.0000%
[sic] and 99.999% for the two runs. Solids reduction was
40%.
Runs 5 and 8 processed Hamilton Harbor sediment diluted to
about 3% solids. A large amount of naphthalene formed
during these runs, resulting in net negative total DEs. How-
ever, naphthalene combustion in the quartz tube furnace was
good. The DRE for Run 5 was 99.99%. Glassware breakage
lost the Run 8 sample for the furnace XAD; the WTC lab
audit analysis provided the DRE for Run 8.
A larger-scale test will likely provide better DREs because
ECO LOGIC encountered a number of problems involving
size restrictions. As these will be eliminated in pilot-scale
tests, ECO LOGIC expects even better results than at bench
scale.
Conclusions
The conclusions from this study were as follows:
• The bench-scale system demonstrated that the gas-phase
chemical reduction reaction can decontaminate polluted
harbor sediment.
• PAHs, especially large ones (coal), were more difficult
to process than chlorinated wastes.
• Harbor sediments can contain amounts of organic mate-
rial sufficient to show a substantial volume decrease
after treatment. The treated solids were free of organic
material.
• The test program demonstrated proof-of-concept on ac-
tual wastes.
Based on the interim results of the bench-scale test program,
Environment Canada and Environment Ontario contracted
ECO LOGIC to undertake a demonstration test program at
Hamilton Harbor, funded by the Environment Canada Con-
taminated Sediments Treatment Program and the Ontario
Environmental Technologies Program.
Case Study C-2:
Pilot-Scale Demonstration of
Contaminated Harbor Sediment
Introduction
ECO LOGIC'S research and development on the treatment of
harbor sediment began with laboratory testing of surrogate
compounds, followed by the bench-scale tests described in
Case Study C-l. The Canadian National Research Council
Industrial Research Assistance Program, the Defense Indus-
trial Research Program, the Environment Canada Contami-
nated Sediments Treatment Program, the Environment Ontario
Environmental Technologies Program, and ECO LOGIC
funded the work.
ECO LOGIC began construction of the mobile pilot-scale
field unit during laboratory testing and undertook a demon-
stration program—the topic of this case study—at Hamilton
Harbor, Ontario, Canada. ECO LOGIC installed its equip-
ment on Hamilton Harbor Commission property, adjacent to a
highly contaminated section of the harbor. The test ran from
April to August, 1991.
Description
The pilot-scale research and development proceeded in four
phases. First, laboratory testing proved the gas-phase chemi-
cal reduction reactions and established parameters for resi-
dence time, temperature, and ratios of hydrogen-to-waste.
ECO LOGIC conducted these tests using laboratory glass-
ware and a quartz tube furnace as a reactor. Next, a larger
reactor (Case C-l) processed 5-10 liters of actual waste
samples, primarily harbor and lagoon sediment. This estab-
lished the capability of the process to treat actual wastes in
complex matrices. During the third phase, ECO LOGIC de-
veloped a computer model to simulate operation of the reactor
system. At the fourth phase, they built the pilot-scale reactor
system and undertook proof-of-concept testing. This phase
included materials, component, and system integrity tests to
ensure leak-free system operation at the test temperatures,
flow rates, and pressures.
ECO LOGIC designed the pilot-scale demonstration system
to process contaminated harbor sediment. The system con-
sisted of an electrically heated reactor that heated and mixed
the watery sediment and hydrogen; a scrubber that removed
particulates, heat, water, and hydrogen chloride from the gas
product; a recirculation system that reheated most of the clean
dry product gas for reinjection into the reactor; a boiler fueled
by propane and the gas product; and a heat exchanger that
steam-heated the watery sediment prior to injection into the
reactor. Two standard drop-deck highway trailers held the
entire system.
39
-------
Testing Protocol
A cable arm bucket removed approximately 12 m3 of contami-
nated sediment from the Sherman Inlet of Hamilton Harbor
and placed it in a 30 m3 lugger box. The bucket crane
transferred the lugger box of contaminated sediment to the
demonstration site spill pad and positioned it at the rear of the
boiler trailer.
ECO LOGIC took water quality samples and measurements
before the removal, immediately afterward, 24 hours later,
and 72 hours later. The laboratory analyzed water and sedi-
ment samples for PCBs, PAHs, oil, grease, and heavy metals.
Bioassays were performed on daphnia and fat-head minnows.
Characterization Tests—After commissioning and system
integrity tests, ECO LOGIC processed a surrogate waste of
clean water and diesel fuel under a variety of conditions. ECO
LOGIC designed the characterization tests to evaluate system
performance on actual harbor sediment while operating within
design parameters, using various feed rates and sediment
concentrations. During these short (2- to 4-hour) tests, Air
Testing Services of Toronto measured organic compound
emission rates (PAH, PCB, chlorobenzene, chlorophenol, di-
oxin, and furan) from the boiler stack emissions using the
Canadian regulatory methods. The stack gas organic com-
pound concentrations were within the regulatory limits for
ambient air; the DEs were satisfactory.
Effluents from the process consisted of reactor grit and slag,
scrubber decant water, and scrubber sludge. These streams
were analyzed for PAHs, PCBs, and metals. The grit and slag
were free of organic contamination and contained only the
inorganic and metallic components of the harbor sediment.
The Wastewater Technology Center (WTC) collected the grit
and slag from the program to evaluate disposal options.
ECO LOGIC tested the decant water for organic compounds
and metals; in all cases it was organic-free. Most of the metals
in the sediment exited with the grit. The decant water repre-
sented the largest volume of effluent from the process, equiva-
lent to the amount of water processed with the sediment. In all
cases, the decant water was acceptable for disposal at munici-
pal sewage treatment plants.
The scrubber sludge represented a minor by-product of the
process and contained primarily lime, carbon, fine particu-
lates, and water. The sludge resulted from recirculating the
scrubber water; some organic contamination of the sludge
occurred. As ECO LOGIC gained experience with the scrub-
ber, they modified system operating parameters to minimize
the amount of sludge production. Although the sludge could
be sent to a landfill, ECO LOGIC found it more economical to
recycle this small effluent stream into the water input stream.
Performance Tests—ECO LOGIC then undertook perfor-
mance testing to demonstrate the capability of the system to
operate for longer periods (days), and to measure a wider
range of emissions during longer sampling periods. During
the third performance test, they spiked the sediment waste
with PCB-contaminated oil to a concentration of 110 ppm.
The performance test effluents paralleled those produced dur-
ing characterization tests—the scrubber decant could be sent
to a POTW, and the scrubber sludge was suitable for landfill
disposal. Sludge production totalled about one percent of the
volume of sediment processed. Hence, it could be economi-
cally recycled into the waste input stream.
Data Summary
The test program progressed in three stages: six initial charac-
terization tests (Cl - C6), followed by a short period for
system modification and repair; five additional characteriza-
tion tests (C7 - Cll), a preperformance test (C12); and three
performance tests (PI - P3). Table C-4 lists the compounds
analyzed in the waste input and effluent streams. Table C-5
lists the characterization test stack components; Table C-6,
the performance test compounds.
Table C-4.
Metals
Waste Input and Effluent Analysis Components
Aluminum
Antimony
Barium
Beryllium
Cadmium
Calcium
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Nickel
Phosphorous
Potassium
Selenium
Silver
Sodium
Polychlorinated aromatic compounds
Polychlorinated
biphenyls
PAHs
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
lndeno(1,2,3-cd)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
40
-------
Table C-5. Characterization Test Air Sampling Components
Polychlorinated aromatic compounds Chlorobenzenes Polychlorinated dibenzofurans Polychlorinated dibenzo-p-dioxins
Polychlorinated biphenyls Chlorophenols
Polyaromatic hydrocarbons
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
ldeno(1,2,3-cd)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Combustion gases
Oxygen
Water vapor
Carbon dioxide
Characterization Tests—Table C-7 summarizes the results
of the characterization tests. In all 12 tests, the PAH stack
concentrations were low; they consisted primarily of naphtha-
lene. Benzo-a-pyrene was not detected in any of the tests. The
concentrations of total PAHs in the stack were below the
Ontario Clean Air Program ambient air quality (AAQ) limits
proposed for both naphthalene (30 |J,g/m3) and coal tar pitch
volatiles (1 |^g/m3). The chlorobenzene emission concentra-
tions also were below the Clean Air Program AAQ guideline
(35 |J,g/m3 for 1,2,4-trichlorobenzene). None of the sample
trains contained chlorophenols, PCBs, or dioxins. Furans
were detected during Characterization Test 5; ECO LOGIC
attributes this to reactor pressure instability, resulting from a
malfunctioning steam flow meter. The malfunction allowed
the flow of waste steam to the reactor to increase substan-
tially, causing the residence time, temperature, and DE to
drop, producing furans from incomplete destruction of
trichlorobenzene.
The residuals from the destruction process included grit from
the bottom of the reactor, decant water from the scrubber, and
scrubber sludge. The first six samples of grit contained mainly
water, presented as water concentrations in units of |J,g/L
(ppb). The next six tests were reported as dewatered concen-
trations of the solid material in units of ng/g (ppb). In general,
the grit contained a total PAH concentration of several ppm.
No PCBs appeared in the grit during any tests. The iron, zinc,
and magnesium levels in the grit make it potentially recy-
clable as an ore for the steel industry. WTC collected the grit
to test for various disposal or recycling options.
The laboratory analyzed the decant water for PAHs, PCBs,
and metals. It did not contain detectable levels of PCBs for
any of the tests. The levels of PAHs and metals in the decant
water met the standards for sewer disposal.
Scrubber sludge consisted mainly of lime, carbon, fine par-
ticulates, and water; the sludge water characteristics were
similar to those of the decant water. The sludge was contami-
nated with PAHs to some extent. Sludge metal content was
low but increased as the sludge concentrated. Although the
sludge was suitable for landfilling, it was more economical to
recycle it into the water input stream.
Performance Tests—Table C-8 summarizes the results of
the performance tests. During these tests, the system operated
24 hr/day, with periodic stoppages for maintenance.
Performance test stack emission sampling was more extensive
than sampling for characterization tests. ECO LOGIC used
three stack sampling trains per test to measure semivolatile
trace organic compounds, VOCs, and metals. The semivola-
tile (MM5) train sampled the stack gas for PAHs, PCBs,
chlorobenzenes, chlorophenols, dioxins, and furans; the vola-
tile train (VOST) and metals train, for the compounds and
metals shown in Table C-6. Continuous analyzers sampled the
stack gas for oxygen, carbon dioxide, carbon monoxide, total
hydrocarbons, water vapor, sulfur dioxide, and nitrogen ox-
ides. The gaseous emissions were all within AAQ guidelines.
The PAH and chlorobenzene concentrations in the boiler
stack were below the Clean Air Program AAQ limits. There
were no detectable emissions of chlorophenols, PCBs, diox-
ins, or furans. The results of the VOST testing indicated that
all of the levels were lower than AAQ guidelines. The metals/
paniculate train measured metals and paniculate emissions,
which also met AAQ guidelines.
The feedstock grit, decant water, and scrubber sludge were
analyzed for PAHs, PCBs, and metals; the feedstock analysis
also included solids content and total organic content. The
effluents from the performance tests were similar to the
characterization tests. The grit was almost free of PAHs and
contained no detectable PCBs. It had a total PAH concentra-
tion ranging from less than 1 ppm for Test 3 to 8 ppm for Test
2. Most of the contamination consisted of naphthalene. PCBs
were not detected in the grit from any tests. The grit contained
iron, calcium, magnesium, aluminum, potassium, and zinc.
Analyses of the decant water for PAHs, PCBs, and metals
indicated no detectable levels of PCBs for any tests. PAHs
and metals in the decant water were well below acceptable
limits for sewer disposal.
PAHs contaminated the scrubber sludge. This sludge, when
combined with the sludge produced during the characteriza-
tion tests, was suitable for landfill disposal but could also have
been reprocessed. The volume of sludge produced during the
41
-------
Table C-6. Air Emission Sampling Components
Participate Material
Metals
Combustion gases
Polychlorinated aromatic compounds
PAHs
VOCs
Components
Aluminum
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Oxygen
Carbon dioxide
Carbon monoxide
Water vapor
Chlorobenzenes
Polychlorinated biphenyls
Polychlorinated dibenzofurans
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Acetone
Dichloromethane
Hexane
Benzene
2-Methylhexane
3-Methylhexane
Heptane
Methylcyclohexane
Toluene
Perchloroethylene
Decane
Undecane
Pentylcyclohexane
Dichlorodifluoromethane
Nickel
Phosphorous
Potassium
Selenium
Silver
Sodium
Sulfur
Tellurium
Thallium
Tin
Titanium
Zinc
Sulfur dioxide
Nitrogen oxides
Total hydrocarbons
Chlorophenols
Polychlorinated dibenzo-p-dioxins
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
lndeno(1 ,2,3-cd)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Trichlorofluoromethane
Methylpentane
Methylcyclopentane
Octane
Dodecane
Tridecane
Naphthalene
Ethylbenzene
Meta/Paraxylene
1 ,4-Dichlorobenzene
Methylnaphthalene
C9-C12 Aliphatics
C5-C10 Heterocompounds
C4 Substituted Benzene
42
-------
Table C-7. Characterization Test Results
Test number 1 2
Waste input (kg)
% Solids
% Organics
Stack concentration
(ng/DSCM)
PAHs
Chlorobenzene
Chlorophenol
Emission rates (ng/s)
PAHs
Chlorobenzene
Chlorophenol
Reactor grit
PAHs (ng/L)*
PAHs (ng/g)*
Metals (ng/ml)**
Metals (ng/g)**
Scrubber decant H2O
PAHs (ng/L)*
Metals (|ig/ml)**
Scrubber sludge
PAHs (ng/L)*
Metals (ng/ml)**
Waste input (kg)
% Solids
% Organics
Stack concentration
(ng/DSCM)
PAHs
Chlorobenzene
Chlorophenol
Emission rates (ng/s)
PAH
Chlorobenzene
Chlorophenol
Reactor grit
PAHs (ng/L)*
PAHs (ng/g)*
Metals (ng/ml)**
Metals (ng/g)**
Scrubber decant H2O
PAHs (ng/L)*
Metals (ng/ml)**
Scrubber sludge
PAHs (ng/L)*
Metals (ng/ml)**
200
5
30
1000
70
0
66
4.5
0
2.3
1.5
ND-0.05
66.0
ND-0.39
240
9
28
190
11
0
17
0.98
0
252.1
ND-3.61
1.1
ND-0.17
4.5
ND-0.84
250
5
30
999
100
0
96
9.8
0
3.6
ND-0.26
0.8
ND
48.2
ND-0.03
240
8
28
620
28
0
37
1.6
0
617.1
ND-550
1.1
ND-0.07
34.0
ND-1.64
3
350
5
30
260
19
0
19
1.4
0
2.2
ND-0.06
1.6
ND-0.06
30.6
ND-0.02
240
8
28
160
11
0
11
0.77
0
192.4
ND-220
2.4
ND-0.2
89.3
ND-0.66
4
400
5
28
1000
510
0
82
40
0
43.5
ND-8.52
0.5
ND-0.65
30.4
ND-0.09
240
8
28
250
13
0
18
0.92
0
61.5
ND-260
2.5
ND-0.07
42.2
ND-0.85
5
250
6
32
140
34000
0
8.9
2100
0
4.1
ND-9.72
1.9
ND-0.12
25.3
ND-0.44
240
9
29
280
11
0
20
0.84
0
423.0
ND-250
5.4
ND-0.05
78.3
ND-5.18
6
350
6
32
460
27
0
33
1.9
0
10.2
ND-6.04
2.1
ND-0.12
41.5
ND-0.14
122.5
6
32
370
10
0
20
0.72
0
292.4
ND-240
7.8
ND
1124.2
ND-40.7
* Average of 16 PAH compounds.
** Reporting only 40 CFR 261 Appendix VIII metals.
ND Not detected.
43
-------
Table C-8. Performance Test Results
performance tests was about one percent of the volume of
waste sediment feed.
Test number
Stream
Waste input (kg)
% Solids
% Organics
1
850
7
30
Stack concentrations (ng/DSCM)
PAHs 270
Chlorobenzene 8.1
Chlorophenol
VOCs*
Metals (ng/DSCM)
Particulates (ng/DSCM)
Stack emission rates (ng/s)
PAHs
Chlorobenzenes
Chlorphenol
VOCs*
Metals (ng/s)
Particulates (ng/s)
Reactor grit
PAHs (ng/g)**
Metals (ng/g)***
Scrubber decant H2O
PAHS (ng/L)
Metals (ng/ml)
Scrubber sludge
PAHS (ng/L)
Metals (ng/g)
0
1821.5
1650
620
18
0.71
0
159.8
120
45
104.1
ND-418
30.1
ND-0.22
2046.9
ND-120
2
900
10
30
230
8.0
0
906.2
1275
622
26
0.87
0
98.7
116
57
484.9
ND-360
42.5
ND-0.08
3507.2
ND-203
3
600
10
30
140
68
0
5151.9
2060
1990
12
6.0
0
452.1
142
137
22.8
ND-140
26.4
ND-0.003
265.8
ND-106
During Performance Test 3, ECO LOGIC spiked the harbor
sediment with PCB oil to a level of 110 ppm to demonstrate
that the process could destroy PCB -contaminated material.
The analytical results found no detectable concentrations of
PCBs in the stack gas, the reactor grit, the scrubber decant
water, or the scrubber sludge. Chlorinated compounds such as
dioxins, furans, and chlorophenols were not detected in the
stack emissions. Based on the detection limits, ECO LOGIC
demonstrated a 99.9999% ORE.
Conclusions
The level of organic emissions produced by Performance Test
3, in which PCB -spiked waste was processed, demonstrated
that the process is suitable for destruction of PCB -contami-
nated material, verifying the bench-scale and laboratory-scale
research. PCBs were not detected in the stack gas, the reactor
grit, the scrubber decant water, or the scrubber sludge. The
stack emissions did not contain dioxins, furans, or chlorophe-
nols. ECO LOGIC demonstrated a 99.9999% ORE.
The process operated successfully for extended periods. Al-
though grit blockages and heating element breakage caused
interruptions in processing, ECO LOGIC has since corrected
both problems.
Key Contacts
Craig Wardlaw
Wastewater Technology Center
867 Chemin Lakeshore Road
P.O. Box 5068
Burlington, Ontario L7R 4L7
Panarta
* Average of reported values.
** Average of 16 PAH compounds.
*** Reporting only 40 CFR 261 Appendix VIII metals.
ND Not detected
Phone: 416-336-4691
James Nash
ELI Eco Logic International, Inc.
143 Dennis St.
Rockwood, Ontario NOB 2KO
Canada
Phone: 519-856-9591
Fax: 519-856-9235
44
------- |