EPA/540/A5-89/010
June 1989
Shirco Infrared Incineration 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
under Contract No. 68-03-3255 and the Superfund Innovative Technology Evaluation (SITE)
Program. It has been subjected to the Agency's peer review and administrative review and it has
been approved for publication as a USEPA document. Mention of trade names or commercial
products does not constitute an endorsement or recommendation for use.
11
-------
Foreword
The Superfund Innovative Technology Evaluation (SITE) program was authorized in the 1986
Superfund amendments. The program is a joint effort between EPA's Office of Research and
Development and Office of Solid Waste and Emergency Response. The purpose of the program is to
assist the development of hazardous-waste treatment technologies necessary to implement new
cleanup standards which require greater reliance on permanent remedies. This is accomplished
through technology demonstrations designed to provide engineering and cost data on selected
technologies.
This project consists of an analysis of the Shirco Infrared Thermal Destruction System and
represents the first and third field demonstrations in the SITE program. The first technology
demonstration took place during the operation of a transportable Shirco system during an
emergency cleanup at a former waste-oil rerefining-facility designated as the Peak Oil Superfund
site in Brandon, Florida. The other technology demonstration occurred during the operation of a
pilot-scale Shirco system at an abandoned waste site, the Demode Road Superfund site in Rose
Township, Michigan. The demonstration efforts were directed at obtaining information on the
performance and cost of the transportable and pilot-scale systems for use in assessments at other
sites. Documentation will consist of three reports. Two Technology Evaluation Reports describing
the field activities and laboratory results of each demonstration have been previously issued. This
Applications Analysis Report provides an interpretation of the data and presents overall conclusions
on the results and potential applicability of the technology.
Additional copies of this report may be obtained at no charge from EPA's Center for Environmental
Research Information, 26 West Martin Luther King Drive, Cincinnati, Ohio, 45268, using the EPA
document number found on the report's front cover. Once this supply is exhausted, 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. You can also call the SITE Clearinghouse hotline at 1-800-424-9346 or 202-382-3000 in
Washington, D.C. to inquire about the availability of other reports.
Margaret M. Kelly, Director
Technology Staff,Office
ofProgram Management and
Technology
Alfred W. Lindsey, Acting
Director, Office of
Engineering and Technology
Demonstration
111
-------
Abstract
This document is an evaluation of the Shirco infrared thermal destruction technology and its appli-
cability as an onsite treatment method for waste site cleanup.
A demonstration was conducted at the Peak Oil Superfund site in Brandon, FL in Aug. 1987 during
a removal action employing a commercial-scale transportable unit. A second demonstration was con-
ducted at the Demode Road Superfund site in Rose Township, MI in Nov. 1987 using the pilot-scale
unit. Operational data and sampling and analysis information were carefully monitored and con-
trolled to establish a database against which other available data and the vendor's claims for the
technology could be compared and evaluated. Conclusions were reached concerning the technology's
suitability for use in cleanup of the types of materials found at the test site, and extrapolations were
made to cleanups of other materials.
Other operations using the Shirco technology range from pilot-scale tests to obtain TSCA permits
and evaluate technology effectiveness to commercial incineration of thousands of tons of PCB-
contaminated soil.
The conclusions drawn from the test results and other available data are that: (1) the commercial
unit is a viable transportable thermal-system consisting of 5 main component trailers and other aux-
iliary facilities; (2) the unit operation is sensitive to the physical and chemical characteristics of the
waste feed and requires a relatively dry soil-like material with a particle sized between 5 micron and
2 in.; (3) the process can thermally decontaminate feed and destroy organic contaminants and, in
general, meet applicable
incineration performance standards; (4) the process cannot reduce the mobility of heavy metals, thus
requiring further furnace ash processing, if applicable; and (5) the unit is an attractive economical
alternative to other established transportable thermal- treatment systems and technologies.
IV
-------
Contents
Page
Foreword . . . iii
Abstract iv
Figures vi
Tables ;..' vii
Abbreviations and Symbols x
Conversions xiii
Acknowledgment xiv
l.Executive Summary 1
2.Introduction 7
The SITE program 7
SITE program reports 7
Key contacts 8
S.Technology Applications Analysis 9
Introduction 9
Conclusions 9
Evaluation of technology performance 11
Environmental regulations and comparison with Shirco performance 17
Waste characteristics and their impact on the performance of the technology 21
Ranges of site characteristics suitable for the technology 24
Material handling required by the demonstrated technology 25
Personnel issues 28
Tests to evaluate application and performance of technology 28
4.Economic Analysis 31
Introduction 31
Results of economic analysis 31
Basis of economic analysis 33
References 36
Appendices
A. Process description 39
B. Vendor's claims for the technology 43
C. SITE demonstration results 51
D. Case studies , 65
References for Appendices 119
-------
Figures
Number Page
A-l Systems process flow and overall test site layout — Peak Oil 40
A-2 System process flow for the pilot-scale unit 41
B-l Shirco thermal treatment costs - 9x61-ft transportable (commercial) unit 49
VI
-------
Tables
Number
1 SITE demonstration: Comparison of results to environmental standards
2 Applicable range of waste characteristics
3 Estimated costs in $/ton: Site cleanup cost-element breakdown
4 Summary of estimated costs
A-l Transportable Shirco unit design parameters
B-l Waste characteristics - General
B-2 Economic model for Shirco transportable (commercial) unit
C-l.l SITE demonstration: Summary of test results
C-2.1 Operations summary
C-2.2 SITE demonstration: Summary of test results
D-l.l Operating parameters
D-1.2 Demonstration Test summary
D-1.3 Waste characterization of drummed soils
D-1.4 Concentration of PCBs in waste feed
D-1.5 Particulate emissions
D-1.6 HC1 emissions
D-1.7 Total organic halogens
D-1.8 Continuous monitoring emission results
D-1.9 Concentration of PCBs in flue gas samples
D-1.10 Destruction and removal efficiency of PCBs
D-2.1 Operating parameters for the 1987 test runs
D-2.2 Operating parameters for the 1988 diagnostic runs ,
D-2.3 Operating parameters for the 1988 test runs
D-2.4 Feed characteristics and DRE for the 1987 tests
D-2.5 Feed characteristics and DRE for the 1988 tests
D-2.6 Furnace ash analysis for the 1987 tests
D-2.7 Furnace ash analysis for the 1988 tests
D-2.8 Stack emissions data for the 1987 tests
D-2.9 Stack emissions data for the 1988 tests
D-2.10 Scrubber water data for the 1987 tests
Page
. 19
.. 21
.. 32
.. 33
.. 42
.. 44
.. 46
.. 55
.. 60
.. 61
.. 66
.. 67
.. 68
.. 69
.. 69
.. 69
.. 69
.. 70
.. 71
.. 71
.. 74
,. 75
.. 75
.. 75
.. 76
.. 77
.. 77
.. 79
.. 79
.. 80
vn
-------
Tables (continued)
Number Page
D-2.11 Scrubber water data for the 1988 tests 80
D-5.1 Operational data for the Jan. 1987 tests 88
D-5.2 Operational datafor the May 1987 tests 88
D-5.3 Demonstration Test results summary 90
D-5.4 Dioxins and furans in furnace ash -Jan. 1987 tests 91
D-5.5 Dioxins and furans in furnace ash and scrubber water - May 91
D-5.6 PCS DRE for the Jan. 1987 test 92
D-5.7 PCB DRE for the May 1987 tests 92
D-5.8 Dioxin and furans in stack emissions - Jan. 1987 tests 92
D-5.9 Dioxin and furans in stack emissions - May 1987 tests 92
D-5.10 Stack emissions - Jan. tests 92
D-5.11 Stack emissions - May tests 92
D-5.12 Scrubber water PCB levels 92
D-5.13 Dioxins and furans in scrubber water - Jan. 1987 tests 93
D-6.1 Brio site process conditions 96
D-6,2 Typical contaminants in Brio site feed material 97
D-6.3 Weight and volume reduction of waste feed materials 97
D-6.4 Brio site stack-gas analyses 98
D-6.5 Destruction and removal efficiency of CCLt 99
D-6.6 Onsite mobile incineration service - Estimated economic model 98
D-7.1 Operation summary 102
D-7.2 PCB concentrations in waste feed 102
D-7.3 PCDD/PCDF concentrations in waste feed 103
D-7.4 Concentrations of PCBs in furnace ash 103
D-7.5 Concentrations of PCDD/PCDF infurnace ash 103
D-7.6 Concentrations of PCBs in stack gas 103
D-7.7 Concentrations of PCDD/PCDF in stack gas 104
D-7.8 Destruction and removal efficiency of PCBs and PCDDs/PCDFs 104
D-7.9 Stack gas composition 105
D-7.10 PCDF/PCDD concentrations in scrubber effluent 105
Vlll
-------
Tables (continued)
Number
D-8.1
D-8.2
D-8.3
D-8.4
D-8.5
D-9.1
Page
Thermal process test data summary 109
Test results summary 110
Pretest waste analysis data : 110
Waste feed analysis Ill
Flue gas destruction and removal efficiencies 112
Summary of results 114
D-10.1 Simulated creosote feed analysis 115
D-10.2 Simulated creosote waste incineration operating conditions 115
D-10.3 Furnace ash analysis 116
D-10.4 Destruction and removal efficiencies 117
IX
-------
Abbreviations and Symbols
ACL
acm
afp
ATSDR
APC
ARARs
BACT
Br
Btu
C
CAA
cc
CERCLA
CE
OEM
CPR
Cl
CO
C02
CWA
DE
DOHS
DRE
alternate concentration
limit
actual cubic meters
mactual feet per minute
Agency for Toxic
Substances and Disease
Registry
air pollution control
applicable or relevant and
appropriate requirements
Best available control
technology
bromine
british thermal unit
carbon
Clean Air Amendments
cubic centimeter
Comprehensive
Environmental Response,
Compensation, and
Liability Act
carbon tetrachloride
combustion efficiency
continuous emission
monitor
Code of Federal
Regulations
chlorine
carbon monoxide
carbon dioxide
Clean Water Act
decontamination efficiency
Department of Health
Services
destruction and removal
efficiency
dscf dry standard cubic feet
dscfm dry standard cubic feet per
minute
dscm dry standard cubic meters
dscmm dry standard cubic meters
per minute
EC environment control
EDR equivalent daily rate
EPA Environmental Protection
Agency
EP Tox EP Toxicity Test
Procedure
F
FRP
ft
FWQC
g
gal
GC/MS
gr
GRAY
gpm
H
H2O
HC1
HHV
h
I
ID
IEPA
fluorine
fiberglass reinforced
plastic
feet
federal water quality
criteria
grams
gallons
gas chromatography/ mass
spectrometry
grains
gravimetric
gallons per minute
hydrogen
water
hydrogen chloride
high heating value
hour
iodine
induced draft
Illinois Environmental
Protection Agency
-------
Abbreviations and Symbols
(continued)
in. inches
K potassium
kg kilograms
kVA kilovolt ampere
kW kilowatts
kWh kilowatt hour
L liters
LAER lowest achievable emission
rate
Ib pound
LEU LaSalle Electric Utilities
M million
m meters
MCLG maximum contaminant
level goal
MDNR Michigan Department of
Natural Resources
mg milligrams
min minutes
mL milliliters
MM5 Modified Method 5
mo month
ug micrograms
N2 nitrogen
Na sodium
NAAQS National Ambient Air
Quality Standards
NBS National Bureau of
Standards
NCP National Contingency
Plan
ND not detected
ng . nanograms
NOx nitrogen oxide
NPDES National Pollutant
Discharge Elimination
System
NPL National Priorities List
NSR New Source Review
OHM OH Materials Corporation
ORD Office of Research and
Development
OSHA Occupational Safety and
Health Act
OSWER Office of Solid Waste and
Emergency Response
Oa oxygen
P phosphorous
PAH polynuclear aromatic
hydrocarbon
Pb lead
PCBs polychlorinated biphenyls
PCC primary combustion
chamber
PCDD polychlorinated dibenzo-
p-dioxin
PCDF polychlorinated
dibenzofurans
PCP pentachlorophenol
pH a measure of acidity or
alkalinity
PIC product of incomplete
combustion
PL public law
POHC principal organic
hazardous constituent
POTW publicly owned treatment
works
PP priority pollutant
XI
-------
Abbreviations and Symbols
ppb
ppm (v)
ppt
PSD
psi
%
QA7QC
RACT
RCRA
RI/FS
ROD
RPM
RREL
S
S&A
SARA
SASS
SCAQMD
sec
SCFH
SCFM
s
SITE
parts per billion
parts per million
(volume)
parts per trillion
prevention of significant
deterioration
pounds per square inch
percent
quality assurance/quality
control
reasonably available
control technology
Resource Conservation
and Recovery Act
Remedial Investigation/
Feasibility Study
Record of Decision
revolutions per minute
Risk Reduction
Engineering Laboratory
sulfur
sampling and analysis
Superfund Amendments
and Reauthorization Act
Source Assessment
Sampling System
South Coast Air Quality
Management District
secondary combustion
chamber
standard cubic feet per
hour
standard cubic feet per
minute
seconds
Superfund Innovative
(continued)
SC-2
SPCCSpill
ft*
sv
TAT
TCDD
TCDF
TCLP
TCO
TDS
TGA
THC
TOC
tpd
TSCA
TSS
UHC
V
V
VGA
VOST
we
WHI
wt
<
>
sulfur dioxide
Prevention, Control, and
Countermeasure Plan
square feet
semivolatile
Technical Assistance
Team
tetrachlorodibenzo-p-
dioxm
tetrachlorodibenzofuran
Toxicity Characteristic
Leaching Procedure
total chromotographable
organics
total dissolved solids
thermogravimetric
analyses
total hydrocarbons
total organic carbon
tons per day
Toxic Substances Control
Act
total suspended solids
unburned hydrocarbon
volume
volt
volatile organic analysis
volatile organic sampling
train
water column
Westinghouse/Haztech,
Inc.
weight
less than
greater than
Technology Evaluation
xu
-------
To convert from
Btu/lb
ft3
yds
ft
op
gal
hp
Ib
Btu/h
Conversions
to
J/g
m3
m3
m
kW
kg
J/h
Multiply by
2.326 E +00
2.832 E-02
7.646 E-01
3.048 E-01
tc = (tF-32)/1.8
3.785 E-03
7.46 E-01
4.535 E-01
1.055 E +03
Xlll
-------
Acknowledgment
This report was prepared under the direction and coordination of Howard Wall, EPA SITE Program
Manager in the Risk Reduction Engineering Laboratory, Cincinnati, Ohio. Contributors and
reviewers for this report were Dick Valentinetti, John Kingscott, and Linda Galer of USEPA,
Washington, DC; Stephen James, Robert Olexsey, and Ronald D. Hill of USEPA, RREL, Cincinnati,
Ohio; and Mike Hill of ECOVA Corp., Dallas, Texas.
This report was prepared for EPA's Superfund Innovative Technology Evaluation (SITE) Program
by Seymour Rosenthal of Foster Wheeler Enviresponse, Inc. for the U.S. Environmental Protection
Agency under Contract no. 68-03-3255.
XIV
-------
SECTION 1
EXECUTIVE SUMMARY
Introduction
The Shirco infrared thermal destruction system was
tested and evaluated under the Superfund Inno-
vative Technology Evaluation (SITE) Program. The
Shirco thermal destruction technology is similar to
other conventional incineration processes in many
respects. Like other systems such as a rotary kiln,
the Shirco unit utilizes staged combustion where the
organics are driven out of the soil in a primary com-
bustion chamber and then combusted in a secondary
chamber. The major difference is that the Shirco unit
uses an electric-powered infrared heat-source in the
primary chamber instead of fossil fuels. This feature
results in a lower gas flow from the primary combus-
tion chamber, and a subsequently smaller secondary
combustion chamber (SCO with lower fuel use and a
smaller emissions-control system. Particulate emis-
sions are also reduced because the soil is processed
virtually undisturbed on the primary combustion
chamber-conveyor (PCC) belt.
Shirco technology demonstrations under the SITE
program were conducted at the Peak Oil Superfund
site in Brandon, Fla. in Aug. 1987 [1] during a re-
moval action employing a commercial unit and at the
Demode Road Superfund site in Rose Township,
Mich, in Nov. 1987 [2] using a pilot-scale unit. The
major objectives of these demonstrations were to de-
termine the following:
• Characteristics of the site and the waste feed.
• DRE levels for PCBs, and the presence of PICs in
the stack gas. The applicable regulatory standard
for PCBs is 99.9999% DRE under the Toxic Sub-
stances and Control Act (TSCA). In contrast the
regulatory standard for DRE under the Resource
Conservation and Recovery Act (RCRA) is 99.99%
for other POHCs and 99.9999% for dioxins and
furans.
• Level of hydrogen chloride (HC1) and particulates
(including heavy metals concentrations) in the
stack gas. The RCRA standard for HC1 in the
stack gas is the larger of 1.8 kg/h (4 Ib/h) or 99
wt% HC1 removal efficiency.
The RCRA standard for particulate emissions in the
stack gas is 180 mg/dscm (0.08 gr/dscf).
• Level of residual heavy metals, organics, and
PCBs in the furnace ash. The TSCA guidance lev-
el is 2 ppm of residual PCBs in the furnace ash.
• Mobility of heavy metals in the furnace ash as
measured by the Extraction Procedure Toxicity
(EP Tox) and the proposed Toxicity Characteristic
Leaching Procedure (TCLP) tests.
• Mobility of heavy metals, particularly lead, in the
furnace ash as compared to the feed.
• Level of residual heavy metals, organic com-
pounds, and other physical and chemical charac-
teristics in the scrubber water discharged from the
unit.
• Overall reliability of the unit during operation.
• Effect of varying operating conditions on unit per-
formance and energy consumption.
• Costs for commercial applications.
In addition to the SITE demonstrations at the Peak
Oil and Demode Road sites, information is available
on the Shirco technology performance from the pilot-
scale and commercial soil incinerations performed by
different organizations (Appendix D). These range
from the conduct of pilot-scale tests to obtain TSCA
permits, to incineration of thousands of tons of PCB-
contaminated soil. This information was reviewed
and used to supplement the SITE demonstration
data in evaluating the Shirco technology against the
objectives listed above. After the initiation of the
SITE Program, Shirco Infrared Systems, Inc. filed for
bankruptcy, and ECOVA Corp. of Redmond, Wash-
ington, purchased a license from Shirco Infrared Sys-
tems, Inc. to construct 2 commercial and 2 pilot-scale
units. ECOVA intends to construct, own, and operate
-------
the infrared thermal destruction systems as part of
their overall remediation capabilities. Other licenses
are available.
Conclusions
The conclusions drawn from reviewing the data on
the Shirco process, both from the SITE demon-
strations, where the most extensive results were
obtained, and the literature, in relation to SITE
Program objectives, are:
• The commercial and pilot-scale units are designed
for transport to remote sites and to be self-
contained and stand-alone units. In all cases there
was sufficient plot area to accommodate the units;
required site improvements, facilities upgrading
and utility connections were completed without
significant problems.
• The Shirco commercial unit is designed to process
wastes over a specific range of physical and
chemical characteristics, including morphology,
particle size, rheology, moisture content, density,
heating value, pH, and organic and inorganic
compounds including metals. Based on the results,
preoperations testing and analysis is critical to
identify the characteristics and conditions that
will contribute to operating difficulties. In
general, a dry soil-like material that is 5 microns
to 2 in. in size is an ideal waste feed to the Shirco
unit. Unit conditions can be adjusted to accom-
modate other physical and chemical charac-
teristics.
• The Shirco process can thermally destroy organic
contaminants in the waste feed and meet the ap-
plicable RCRA DRE performance standard of
99.99% and TSCA DRE performance standard of
99.9999%. Volatile and semivolatile organic com-
pounds measured in the stack gas are possible
PICs and include halomethanes and chlorinated
organics, aromatic volatiles and semivolatiles,
and oxygenated hydrocarbons. These compounds
were detected at levels significantly lower than es-
tablished standards for their direct inhalation
[23].
• The RCRA performance standard for acid gas re-
moval, which is the larger of 1.8 kg/h HC1 in the
stack gas or 99 wt% HC1 removal efficiency, was
met during the operation of the commercial and
pilot-scale units. In general, the RCRA perform-
ance standard for particulate emissions of 180
mg/dscm (0.0 8 gr/dscf) was met during operations.
In some cases this standard was not met. Current
design modifications to operating commercial
units are addressing this problem with the addi-
tion or planned-addition of a high efficiency scrub-
ber system [7,22].
• The Shirco unit has demonstrated the ability to ef-
fectively decontaminate the waste feed and pro-
duce a furnace ash that contains minimal levels of
organic contaminants consistent with applicable
regulatory standards and guidelines. In particu-
lar, unit operations on waste feed contaminated
with PCBs has consistently resulted in a furnace
ash that meets the TSCA guidance level of 2 ppm
of residual PCBs. The majority of the heavy met-
als in the feed concentrate in the furnace ash and
may require further treatment to meet the toxic-
ity characteristic standards.
• There is no definitive trend or evidence from the
data that the Shirco thermal-treatment process
reduces the mobility of heavy metals in the fur-
nace ash as compared to the feed.
• Scrubber water quality was satisfactory in most
operations and with appropriate onsite treatment
can be discharged to local POTWs.
• In general, recent operations of the commercial
units exhibit overall project utilization or operat-
ing factors ranging from 24%-61%. Intermittent
operations of the commercial units over 1-3 mo.
periods have realized utilization factors up to 90%,
which are in agreement with the 85% factor
claims by the vendor. Based on the data and the
complexities of incineration systems in general, it
is expected that an overall utilization factor of
50%-75% is a realistic and achievable range.
• During the operation of the pilot-scale unit at the
Demode Road site, the unit was able to successful-
ly decontaminate the feed and destroy PCBs using
less electrical power when fuel oil was added to
the waste and when PCC temperature was re-
duced. The addition of fuel oil also permitted a
higher feed rate. Additional energy savings were
obtained when the SCC temperature was also re-
duced. Cost savings in specific applications will
depend on local fuel and electrical costs. Minimum
PCC and SCC temperatures must be maintained
to achieve adequate desorption and the necessary
destruction of the organics.
• The Shirco unit is an attractive economical alter-
native to other established transportable thermal
treatment systems and technologies. Costs based
on the economic analysis range from approximate-
ly $182/ton- $24 I/ton of waste feed - excluding
waste excavation, feed preparation, vendor profit,
and ash disposal costs.
-------
Results of Applications Analysis
Site Characteristics
The commercial and pilot-scale units are designed
for transport to remote sites and to be self-contained
and stand-alone units. In all cases there was suffi-
cient plot area to accommodate the units. Required
site improvements, facilities upgrading, and utility
connections were completed without significant
problems that either delayed, aborted, or affected the
operations of any specific cleanup action.
Waste Characteristicslnformation on both the phys-
ical and chemical characteristics of the waste matrix
is necessary to determine: the suitability of that
waste for thermal processing using the Shirco tech-
nology, and the possible need for waste preparation
and pretreatment.
Preoperations waste-feed characterization and labo-
ratory analysis, and thermogravimetric analyses
(TGA) (including pilot or bench-scale testing) are
mandatory in order to define the waste feed matrix
and its impact on the Shirco unit's pretreatment and
waste-feed preparation requirements, metallurgical
requirements and/or limitations, potential design
limitations (particularly in the offgas treatment sec-
tion), and operating conditions.
The following summary presents a range of waste
characteristics suitable for processing in the Shirco
unit based on an analysis of available data.
Characteristics
Applicable Range
morphology
particle size (diameter)
moisture content
density
heating value
organics (including POHCs)
chlorine
sulfur
phosphorous
pH
alkali metals
heavy metals
soil/solid
semi-solid
oily-sludge/solid
5 microns - 2 in.
0-50 wt% (no free liquids or free-
flowing sludges)
30-130 Ib/ft3
0-10,000 Btu/lb
0-100 wt% (determined by
preoperation testing)
0-5 wt%
0-5 wt%
0-300 ppm
5-9
0-1 wt%
0-1 wt% (determined by
preoperation testing)
In order to match specific-site waste-matrices to
these waste characteristics, preprocessing to meet
the physical and chemical characteristics require-
ments of the Shirco unit may be required. This in-
cludes sizing, classifying, screening, dewatering,
soils blending, and/or lime addition prior to process-
ing to ensure a solid/semi-solid matrix with charac-
teristics within the applicable range of the unit. Pure
liquids can also be processed if blended with a suit-
able carrier such as soil or vermiculite to form a
semi-solid waste matrix.
Specific examples from the SITE demonstrations and
case studies where departure from the recommended
range of waste characteristics caused unit operating
problems include:
Operation
Departure from
applicable range
Unit problem
Peak Oil [ 1 ] Improperly
prepared
oily/clumpy sludge
High concentration
of lead
Florida Steel [3,4,5] Particle size
High chlorine
content
Brio [13]
International
Paper [15]
Low Btu
Lumpy/clay feed
Tar-like feed,
adhesive and
cohesive, high
moisture
Materials handling
Feed handling
Ash handling
Emissions control
Particles passing
through belt
Emissions control
Maintaining PCC
temperature
Materials handling
Feed handling
Materials handling
Feed handling
Destruction and Removal Efficiency
(ORE) and Stack Gas Emissions
In general, the Shirco unit has demonstrated the
ability to achieve DREs of:
• organics greater than the RCRA performance
standard of 99.99% (Appendix D-6)
• dioxins and furans greater than the RCRA perfor-
mance standard of 99.9999% (Appendix D-9)
• PCBs greater than the TSCA performance stan-
dard of 99.9999% (Appendix D-2)
Volatile and semivolatile organic compounds mea-
sured in the stack gas were typical incinerator PICs
— including halomethanes and chlorinated organics,
-------
aromatic volatiles and semivolatiles and oxygenated
hydrocarbons. These compounds were detected at
levels significantly lower than established standards
for their direct inhalation [23].
Acid Gas Removal and Particulate
Em/ss/ons
The Shirco unit has demonstrated the ability to meet
the RCRA performance standard of the larger of 1.8
kg/h of HC1 in the stack gas or a 99% acid gas remov-
al efficiency.
The commercial and pilot-scale Shirco operations
have not consistently met the RCRA performance
standard of 0.08 gr/dscf for particulate emissions. As
with any transportable incineration system design,
the need to meet a trailer size should not preclude ad-
herence to an emissions-control system-design that
will provide efficient control over a wide range of gas
flows and particulate loadings. Recent commercial
designs [7,22] have incorporated or plan to incorpo-
rate a high efficiency scrubber system.
Preoperations testing and analysis of the waste-feed
for particle size and elemental (halides, S and P),
metals (heavy and alkali), and organic species and
concentrations are necessary to identify contami-
nants that may cause potential emissions problems.
These tests will also define the limits of operation
consistent with the waste-matrix and possible waste-
pretreatment options.
Residual Contaminants in Furnace
Ash and Scrubber Water
The Shirco unit has demonstrated the ability to effec-
tively decontaminate the waste feed and produce a
furnace ash that contains minimal levels of organic
contaminants consistent with applicable regulatory
standards and guidelines. The majority of the heavy
metals in the feed will concentrate in the furnace ash
and may require further treatment to meet the toxic-
ity characteristic standards.
Scrubber water quality was satisfactory in most op-
erations and with appropriate onsite treatment can
be discharged to local POTWs.
Mobility of Heavy Metals
Despite high levels of metals in the waste-feed and
furnace ash, the concentrations of metals in the EP
Tox and TCLP leachates were low and in most cases
met their respective toxicity characteristic stan-
dards. However, there is no trend or evidence from
the data that the Shirco thermal treatment process
reduces the mobility of heavy metals in the furnace
ash as compared to the feed.
Overall Reliability of the Shirco Unit
In general, recent operations of the commercial units
exhibit utilization or operating factors ranging from
50%-90%. The initial operation of a Shirco unit on a
commercial basis at Peak Oil was at an overall utili-
zation factor of only 24%. This was a first-of-a-kind
operation on a difficult waste-feed where materials
handling, feed system, and emissions control prob-
lems plagued the unit operation.
The commercial operation at Florida Steel initially
ran at a utilization factor of 50%, which then in-
creased to more than 90% during the final month of
operation for an overall project factor of 61%. The op-
eration at LaSalle Electric [11] — which is the same
unit used at Peak Oil but with modifications address-
ing the operating problems encountered at Peak Oil
— has been reported to be currently operating at a
utilization factor of 80% to 90%.
Optimum Operating Conditions
During the SITE demonstration of the pilot-scale
unit at the Demode Road Superfund site, a series of
runs was conducted to examine the effect on energy
-consumption and changes in the residual levels of
heavy metals and organics in the furnace ash versus
the levels in the feed by varying operating condi-
tions.
Based on the data obtained, an analysis was conduct-
ed to compare energy consumption in the unit at op-
erating conditions that did not affect the perfor-
mance of the unit. A reduction in the PCC operating
temperature from 1,600° to 1,200°F reduced the aver-
age PCC power usage 48%. A reduction in the SCC
operating temperature from 2,200° to 1,800°F re-
duced the average propane fuel consumption by 51%.
The use of 3 wt.% fuel oil to supplement the fuel val-
ue of the feed further decreased PCC power usage by
26% to 67% at PCC operating temperatures of 1,600°
and 1,200°F, respectively, with accompanying in-
creases in overall feed rate of 32% and 26%. The costs
for fuel oil and its attendant facilities still must be
examined for specific applications to determine the
cost effectiveness of a fuel oil additive to the waste
feed.
The results did not provide any trend or change in
the residual levels of the heavy metals and organics
in the furnace ash versus the levels in the feed as the
operating conditions were varied and PCC operating
temperatures were maintained between 1,200° and
1,600°F. At an abnormally low PCC operating tern-
-------
perature of 900°F, without the input of combustion
air to simulate non- oxidizing or pyrolytic combus-
tion conditions, total PCB and TCDF concentrations
in the furnace ash increased. The increases may indi-
cate that these PCC conditions led to incomplete de-
sorption or incineration of PCBs and to the produc-
tion" of low levels of TCDF in the furnace ash from the
incomplete combustion of PCBs in the feed.
Costs for Commercial Operations
The economic analysis is based on the processing of
36,500 tons of waste feed in a commercial unit. This
quantity is based on the amount of waste that would
be processed if the commercial unit operated at the
design capacity of 100 ton/d, and a 100% operating
(or utilization) factor over a 365-day annual period.
However, the costs were adjusted to reflect real-time
operations of the unit since periodic shutdowns are
required in order to respond to maintenance or oper-
ational problems. Costs were based on operating fac-
tors ranging from 85% to 50%, equivalent to a range
of 429 to 730 days at the site to process the 36,500
tons of waste feed. Additional cost data is provided in
Appendices B, C, and D. A summary of the estimated
costs obtained from the economic analysis and other
data is presented below.
Data source
Unit
capacity,
tpd
Operating
factor, %
Brio Site 150
Friendswood, Tex. 220
(Shircocostest.)[13]
LaSalle Electric 100
LaSalle, 111.,
(Haztech proposal)
[11]
Florida Steel 100
Indiantown, Fla.
(OH Materials est.)
[4,6]
Peak Oil Brandon, 100
Fla. (SITE Tech.
Eval. Report) [1]
ECO VA Dallas, Tex. 100
(Vendor's claims)
[19]
Economic Analyses 100
(Section 4)
82
82
60
61
80
37
85
85
80
70
60
50
Unit cost
$/ton
143(a)
300(a)
<300(b)
197(c)
416(c)
161-257(a)
182(c)
187(c)
200(c)
217(0
241(c)
(a) Cost includes vendor profit but excludes waste excavation, feed
preparation and ash disposal.
(b) Cost includes vendor profit, waste excavation and feed
preparation but excludes ash disposal.
(c) Cost excludes vendor profit, waste excavation, feed
preparation and ash disposal.
-------
-------
SECTION 2
INTRODUCTION
The SITE Program
In 1986, the EPA's Office of Solid Waste and Emer-
gency Response (OSWER) and Office of Research and
Development (ORD) established the Superfund Inno-
vative Technology Evaluation (SITE) Program to
promote the development and use of innovative tech-
nologies to clean up Superfund sites across the coun-
try. Now in its third year, SITE is helping to provide
the treatment technologies necessary to implement
new federal and state cleanup standards aimed at
permanent remedies, rather than quick fixes. The
SITE Program is composed of three major elements:
the Demonstration Program, the Emerging Technol-
ogies Program, and the Measurement and Monitor-
ing Technologies Program.
The major focus has been on the Demonstration Pro-
gram, which is designed to provide engineering and
cost data on selected technologies. To date, the dem-
onstration projects have not involved EPA funding
for technology developers. EPA and developers par-
ticipating in the program share the cost of the dem-
onstration. Developers are responsible for demon-
strating their innovative systems at chosen sites,
usually Superfund sites. EPA is responsible for sam-
pling, analyzing, and evaluating all test results. The
result is an assessment of the technology's perfor-
mance, reliability, and cost. This information will be
used in conjunction with other data to select the most
appropriate technologies for the cleanup of Super-
fund sites.
Developers of innovative technologies apply to the
Demonstration Program by responding to EPA's an-
nual solicitation. EPA also will accept proposals at
any time when a developer has a treatment project
scheduled with Superfund waste. To qualify for the
program, a new technology must be at the pilot or
full-scale and offer some advantage over existing
technologies. Mobile technologies are of particular
interest to EPA.
Once EPA has accepted a proposal, EPA and the de-
veloper work with the EPA regional offices and state
agencies to identify a site containing wastes suitable
for testing the capabilities of the technology. EPA
prepares a detailed sampling-and-analysis plan de-
signed to thoroughly evaluate the technology and to
ensure that the resulting data are reliable. The dura-
tion of a demonstration varies from a few days to sev-
eral months, depending on the length of time and
quantity of waste needed to assess the technology.
After the completion of a technology demonstration,
EPA prepares two reports, which are explained in
more detail below. Ultimately, the Demonstration
Program leads to an analysis of the technology's
-overall applicability to Superfund problems.
The second principal element of the SITE Program is
the Emerging Technologies Program, which fosters
the further investigation and development of treat-
ment technologies that are still at the laboratory
scale. Successful validation of these technologies
could lead to the development of a system ready for
field demonstration. The third component of the
SITE Program, the Measurement and Monitoring
Technologies program, provides assistance in the de-
velopment and demonstration of innovative technol-
ogies to better characterize Superfund sites.
SITE Program Reports
The analysis of technologies participating in the
Demonstration Program is contained in two docu-
ments, the Technology Evaluation Report and the
Applications Analysis Report. The Technology Eval-
uation Report contains a comprehensive description
of the demonstration sponsored by the SITE program
and its results. This reported costs obtained from the
economic analysis and other data is presented below.
The purpose of the Applications Analysis Report is to
estimate the Superfund applications and costs of a
-------
technology based on all available data. This report
compiles and summarizes the results of the SITE
demonstration, the vendor's design and test data,
and other laboratory and field applications of the
technology. It discusses the advantages, disadvan-
tages, and limitations of the technology. Costs of the
technology for different applications are estimated
based on available data on pilot-scale and commer-
cial applications. The report discusses the factors,
such as site and waste characteristics, that have a
major impact on costs and performance.
The amount of available data for the evaluation of an
innovative technology varies widely. Data may be
limited to laboratory tests on synthetic wastes, or
may include performance data on actual wastes
treated at the pilot- or full- scale. In addition, there
are limits to conclusions regarding Superfund appli-
cations that can be drawn from a single field demon-
stration. A successful field demonstration does not
necessarily assure that a technology will be widely
applicable or fully developed to the commercial size.
The Applications Analysis attempts to synthesize
whatever information is available and draw reason-
able conclusions. This document will be very useful
to those considering the technology for Superfund
cleanup and represents a critical step in the develop-
ment and commercialization of the treatment tech-
nology.
Key Contacts
For more information on the demonstration of the
Shirco technology, please contact:
1. Regional contact concerning the Peak Oil, Bran-
don, Fla. site:
Mr. Fred Stroud
USEPA, Region IV
345 Courtland St. NE
Atlanta, GA 30365
404-347-3931
2. Regional contact concerning the Rose Township,
Mich, site:
Mr. Kevin Adler
USEPA, Region V
230 South Dearborn St.
Chicago, IL 60604
312-886-7078
3. EPA Project Manager concerning the SITE
demonstrations:
Mr. Howard Wall
USEPA
Risk Reduction Engineering Laboratory
26 W. Dr. Martin Luther King Jr. Drive
Cincinnati, OH 45268
513-569-7691
4. Vendor concerning the process:
ECOVA Corp.
Mr. Mike Hill
12790 Merit Drive #220
Dallas, TX 75251
214-404-7540
-------
SECTION 3
TECHNOLOGY APPLICATIONS ANALYSIS
Introduction
This section of the report addresses the applicability
of the Shirco infrared thermal destruction technol-
ogy to various potential feedstocks, based on the re-
sults obtained from the SITE demonstration tests
and other Shirco applications test data. Since the re-
sults of the Demonstration Tests provide the most ex-
tensive database, conclusions on the technology's ef-
fectiveness and its applicability to other potential
cleanups will be strongly influenced by them; these
are presented in detail in the Technology Evaluation
Reports. Additional information on the Shirco tech-
nology, including a process description, vendor
claims, a summary of the demonstration test results,
and summaries of reports on outside sources of data
using the Shirco technology are provided in Appendi-
ces A-D.
The subsequent discussions are presented in the fol-
lowing subsections:
• Conclusions — that have been drawn on the per-
formance and applicability of the technology.
• Evaluation of Technology Performance — that
discusses the available data from the demonstra-
, tions, ECOVA Corp., and other commercial and
pilot-scale applications of the technology, and
provides details on the analytical conclusions
and applicability of the Shirco technolpgy.
• Environmental Regulations and Comparison
with Shirco Performance — that summarizes
the regulations and environmental standards
that apply to the operation of the Shirco unit.
• Waste Characteristics and Their Impact on the
Performance of the Technology — that provides
information defining the appropriate range and
limits of physical and chemical characteristics of
the waste-feed suitable for processing in the
Shirco unit.
• Range of Site Characteristics Suitable for the
Technology — that discusses the specific site re-
quirements for a commercial unit operation in-
cluding site physical characteristics, site area
and utilities availability.
• Material Handling Required by the Demonstrat-
ed Technology — that discusses the type of exca-
vation and materials handling procedures and
equipment that are required and have been em-
ployed at commercial operations to complement
the Shirco unit.
• Personnel Issues — that defines the personnel re-
quirements for the operation of the Shirco com-
mercial unit.
• Tests to Evaluate Technology Applicability and
Performance — that discusses the necessary
treatability testing and data required to estab-
lish the suitability of the waste-feed and the
range of recommended operating parameters for
the commercial unit to assure optimum perfor-
mance within regulatory requirements.
Conclusions
The overall conclusions drawn from reviewing the
data on the Shirco technology are discussed below.
These conclusions are based on the information pre-
sented in detail in the remainder of this section.
Site Requirements
The Shirco commercial-scale unit is an easily-
transportable modular design consisting of 5 main
component trailers and auxiliary equipment that al-
lows for cleanup at any location where access and
site layout can accommodate standard tractor-
trailers. Site area requirements include 15,000 ft2 for
processing, and 30,000 ft2 for feed preparation, ash
-------
storage, and other auxiliary service areas; total site
requirements are approximately 45,000 ft2 [19,22].
Characteristics of Waste Matrix and
Waste Feed
The Shirco unit is designed to process solid wastes or
sludges that do not contain free liquids. Waste prep-
aration is necessary to ensure that particle size of the
feed matrix is controlled between 5 microns and 2 in.
Liquid wastes or sludges with free liquids can be
combined with solid carriers such as sand, soil, or
conditioning lime to render them suitable for pro-
cessing. All large bulk objects must be shredded or
reduced to within the above-mentioned size range to
accommodate the unit's design operations con-
straints. In general, the waste feed to the Shirco unit
should be relatively dry soil-like distinct particles
sized as defined above.
The Shirco unit design incorporates a metal conveyor
belt, conveyor belt rollers, and a series of rotary
rakes for conveyor bed agitation. These internals
will be affected by corrosive combustion atmosphere
contaminants, thereby requiring preoperations ana-
lysis of the waste feed to ensure that the materials of
construction are adequate. A pH below 5 and sulfur
and chloride contents of more than 5 wt% in the
waste feed will affect the materials integrity of the
Shirco design. As in other thermal processes, flu-
orine and phosphorous concentrations greater than
300 ppm can affect the integrity of silica-based re-
fractories and must be taken into account before pro-
cessing a specific waste-feed matrix. Heavy metals,
particularly lead, at concentrations greater than 1
wt% may possibly overload the air pollution control
system if they vaporize and carryover to the scrub-
ber. Alkali metals such as sodium and potassium at
concentrations greater than 1 wt% may affect the in-
tegrity of the silica-based refractories and cause slag-
ging and fouling problems in the air pollution control
system. The concentrations of these metals must be
determined before processing a designated waste-
feed.
Preoperations testing and analysis of the waste ma-
trix found at a potential site is mandatory in order to
ensure that the waste and feed preparation systems
design can accommodate the specific feed require-
ments of the Shirco unit.
Based on the limited unit dimensions, an optimal bed
depth of 2 in., and a maximum particle size of 2 in.,
the design throughput for a Shirco unit is limited to
100-200 ton/d, depending on the physical and ther-
mal characteristics of the waste feed.
Destruction and Removal Efficiency
(ORE) and Stack Gas Emissions
In general, the Shirco unit has demonstrated the
ability to achieve DREs at the following levels:
• organics greater than the RCRA performance
standard of 99.99% (Appendix D-6)
• dioxins and furans greater than the RCRA perfor-
mance standard of 99.9999% (Appendix D-9)
• PCBs greater than the TSCA performance stan-
dard of 99.9999% (Appendix D-2)
Volatile and semivolatile organic compounds mea-
sured in the stack gas were typical incinerator PlCs
— including halomethanes and chlorinated organics,
aromatic volatiles and semivolatiles, and oxygenated
hydrocarbons. These compounds were detected at
levels significantly lower than established standards
for their direct inhalation.
Acid Gas Removal and Paniculate
Emissions
The Shirco unit has demonstrated the ability to meet
the RCRA performance standard of the larger of 1.8
kg/h of HC1 in the stack gas or a 99% acid gas remov-
al efficiency.
The commercial and pilot-scale Shirco operations
have not consistently met the RCRA performance
standard of 0.08 gr/dscf for particulate emissions. As
with any transportable incineration system design,
the need to meet a trailer size should not preclude ad-
herence to an emissions-control system design that
will provide efficient control over a wide range of gas
flows and particulate loadings. Recent commercial
designs [7,22] of the Shirco unit have incorporated or
plan to incorporate a high efficiency scrubber sys-
tem.
Preoperations testing and analysis to determine po-
tential emissions problems are necessary to define
the limits of operation consistent with the waste ma-
trix and possible waste pretreatment options.
Residual Contaminants in Furnace
Ash and Scrubber Water
The Shirco unit has demonstrated the ability to effec-
tively decontaminate the waste feed and produce a
furnace ash that contains minimal levels of organic
contaminants consistent with applicable regulatory
10
-------
standards and guidelines. The majority of the heavy
metals in the feed will concentrate in the furnace ash
and may require further treatment based on toxicity
characteristic standards discussed later in this sec-
tion.
Scrubber water quality was satisfactory in most op-
erations, and with appropriate onsite treatment, can
be discharged to local POTWs. Questions concerning
the effectiveness of the emissions control system, as
discussed above, will impact on the scrubber water
characteristics and quality.
Mobility of Heavy Metals
Despite high levels of metals in the waste-feed and
furnace ash, the concentrations of metals in the EP
Tox and TCLP leachates were low and in most cases
met their respective toxicity characteristic stan-
dards. However, there is no definitive trend or evi-
dence from the data that the Shirco thermal treat-
ment process reduces the mobility of heavy metals in
the furnace ash as compared to the feed.
Evaluation of Technology Performance
For the SITE demonstrations the following technical
and performance criteria were addressed to evaluate
the Shirco technology.
• DRE levels for designated POHCs, PCBs, and the
presence of PICs in the stack gas. The regulatory
standards are 99.99% DRE for POHCs and
99.9999% for dioxins and furans under the Re-
source Conservation and Recovery Act (RC RA)
and 99.9999% DRE for PCBs under the Toxic Sub-
stances and Control Act (T SCA).
• Level of hydrogen chloride (HC1) and particulates
(including heavy metals concentrations) in the
stack gas. The RCRA standard for HC1 in the
stack gas is the larger of 1.8 kg/h (4 Ib/h) or 99
wt% HC1 removal efficiency. The RCRA standard
for particulate emissions in the stack gas is 180
mg/dscm (0.08 gr/dscf).
• Level of residual heavy metals, organics, and
PCBs in the furnace ash. The TSCA guidance lev-
el is 2 ppm of residual PCBs in the furnace ash.
• Mobility of heavy metals in the furnace ash as
measured by the Extraction Procedure Toxicity
(EP Tox) and the proposed Toxicity Characteristic
Leaching Procedure (TCLP) tests.
• Mobility of heavy metals, particularly lead, in the
furnace ash as compared to the feed.
• Level of residual heavy metals and organic com-
pounds, and other physical and chemical charac-
teristics in the scrubber water discharged from the
unit.
• Overall performance and reliability of the unit
during operation.
• Effect of varying operating conditions on unit per-
formance and energy consumption.
The following discussion addresses the above criteria
based on the results obtained during the SITE dem-
onstrations on the Peak Oil commercial unit [1] and
the Demode Road pilot-scale unit [2]. Summaries of
these test results are presented in Appendices C-l'
and C-2, respectively. In addition, the discussion that
follows includes the results obtained during other
pilot-scale and commercial operations involving the
Shirco technology. Summaries of these test results
are presented in Appendix D. These case studies in-
clude:
D-l Florida Steel pilot-scale tests [3-5]
D-2 Florida Steel TSCA trial burns [5]
D-3 Florida Steel commercial cleanup [4,6]
D-4 LaSalle Electric commercial cleanup [7-11]
D-5 Twin Cities pilot-scale tests [12]
D-6 Brio pilot-scale tests [13]
D-7 Tibbetts Road pilot-scale tests [14]
D-8 International Paper pilot-scale tests [15]
D-9 Times Beach pilot-scale tests [16,17]
D-10 Simulated creosote pit pilot-scale tests [18]
Destruction and Removal Efficiency
(DRE)
SITE Demonstration Results—
The SITE demonstration results detected less than
25 ppt total PCBs in the Peak Oil stack-gas samples
and less than 180 ppt total PCBs in the Demode Road
stack-gas samples. In both cases, the data was slight-
ly above or at the detection limits. Total PCB concen-
trations in the feed ranged from 3.48 to 5.8 5 ppm at
Peak Oil and from 10.2 to 35.2 ppm at Demode Road.
Based on these results, the Peak Oil commercial
unit achieved DREs for PCBs ranging from 99.99
80% to 99.99972%, and the Demode Road pilot-scale
unit achieved DREs for PCBs ranging from greater
than 99.9922% to 99.9982%. In general, the low PCB
concentrations in the feed resulted in DRE values
11
-------
that were not able to confirm achievement of the
TSCA regulatory standard of 99.9999%.
Case Study Results—
Except for a few specific runs in several of the case
study programs, the results met the RCRA ORE per-
formance standard of 99.99% for designated POHCs
and 99.9999% for dioxins and furans, and the TSCA
DRE performance standard of 99.9999% for PCBs.
DRE results are presented below, and the specific
runs that did not meet the DRE performance stan-
dard are discussed.
Case Study
POHC
DRE, %
Florida Steel pilot-scale
tests
Florida Steel TSCA trial
burns
Florida Steel commercial
cleanup
LaSalle Electric commercial
cleanup
Twin Cities pilot-scale tests
Brio pilot-scale tests
Tibbctts Road pilot-scale
testa
International Paper pilot-
scale tests
Times Beach pilot-scale
tests
Simulated creosote pit pilot-
scale tests
PCB
PCB
PCB
PCB
PCB
CC14
PCB
PCB
TCDD
PGP
99.9989/99.99992
>99.9999
No available data
No available data
99.997/99.9999989
>99.99
99.99981/99.99992
>99.99
> 99.9999
>99.99
As presented above, runs on the Florida Steel pilot-
scale tests, Twin Cities Pilot-Scale Tests, -and Tib-
betts Road pilot-scale tests failed to meet the TSCA
DRE performance standard of 99.9999% for PCBs.
The explanations that are presented in the specific
case study reports are listed below:
• Florida Steel — Low oxygen level in secondary
combustion chamber due to misoperation.
• Twin Cities - Possible sample contamination.
* Tibbetts Road - Low concentrations of PCBs in the
stack gas sample.
Summary of Results—
With the exception of cases where low PCB concen-
trations in the feed and the stack gas resulted in low
DRE values, the results tend to indicate that the
Shirco technology can meet the designated RCRA
and TSCA DRE performance standards for stack gas
emissions.
Organic Stack Gas Emissions
SITE Demonstration Results—
The SITE demonstration results detected several
volatile and semivolatile organic compounds in the
stack gas at concentrations less than 50 ppb and be-
low established standards for direct inhalation. They
included:
• Chlorinated methanes, methylene, ethanes, ethy-
lenes, and other halomethanes.
• Aromatic volatiles and semivolatiles, such as ben-
zene, toluene, xylene, chlorobenzene, ethylben-
zene, naphthalene, styrene, and pyridine.
• Oxygenated hydrocarbons, including phthalates,
p-chloro-m-cresol, phenol, benzoic acid, acetone,
butanone, and acetophenone.
Dioxins and furans were not detected in the stack gas
samples.
Case Study Results—
All of the case studies detected volatile and semivola-
tile organic compounds similar to the general species
discussed above and at concentrations less than the
established standards for direct inhalation. In one
run conducted during the Tibbetts Road pilot-scale
test [14] a detectable level of TCDF was found and is
attributable to a combination of a poor SCC opera-
tion and the possibility that TCDF may have been a
PIC for the PCBs in the waste feed.
Summary of Results—
All of the data consistently showed minimal con-
centrations of volatile and semivolatile organic com-
pounds in the stack gas at levels less than estab-
lished standards for direct inhalation [23]. The or-
ganics included chlorinated organics and halo-
methanes; aromatic compounds such as benzene,
toluene, xylene, chlorobenzene, ethylbenzene, and
naphthalene and related compounds; and oxy-
genated hydrocarbons such as phthalates, phenol
and related compounds, benzene-related compounds,
and ketones. Dioxins and furans were not detected,
with the one exception (TCDF) discussed above.
Acid Gas Removal
SITE Demonstration Results—
During the SITE demonstrations, the level of chlo-
rine in the waste feed was less than 0.15wt%, and
12
-------
HC1 mass flows in the stack gas were less than 10 g
/h, which is considerably lower than the RCRA per-
formance standard of 1.8 kg/h. Calculated efficien-
cies for the Demode Road pilot-scale demonstration
ranged from 97.23 to 99.35 wt.%; Peak Oil HC1 re-
moval efficiencies could not be calculated because of
the low concentration of chlorine in the feed.
Case Study Results—
In all of the case study results, the data resulted in
HC1 mass flows in the stack gas that were less than
the RCRA performance standard of 1.8 kg/h.
Summary of Results—
All of the available data on the operation of the pilot-
scale and commercial units indicate that the units
have not experienced any problems in meeting the
RCRA performance standard of the larger of 1.8 kg/h
HC1 mass flow or 99 wt% HC1 removal efficiency.
Paniculate Emissions
SITE Demonstration Results—
Although the Demode Road pilot-scale demonstra-
tion successfully achieved particulate emission lev-
els ranging from 7 to 68 mg/dscm (which are below
the RCRA standard of 180 mg/dscm), the Peak Oil
commercial demonstration had continuing problems
meeting the RCRA particulate emissions standard.
During the Peak Oil demonstration, particulate
emissions ranged from 322 to 155 mg/dsc m. Only
after a thorough cleaning of the system and several
mechanical adjustments were the particulate emis-
sions below the RCRA standard. Analyses of the par-
ticulates indicated extremely high concentrations of
lead, sulfur, and sodium, with an average lead con-
centration of 58 wt%. Based on the high initial con-
centrations of these metals in the feed, the use of
poor quality sodium carbonate solutions in the emis-
sions scrubbing system, and the potential for vapor-
ization of these metals and carryover of lead salts as
fines (as illustrated by the high lead concentrations),
it is possible that these species severely impacted on
the overall particulate emissions fines flow and over-
loaded the emissions control system. It is interesting
to note that the emissions control section of the Shir-
co unit that was employed at Peak Oil has been re-
placed with a totally different design in the high- ef-
ficiency Calvert scrubber, which is designed to pro-
vide improved particulate removal efficiencies over a
wider range of gas flows and fines loadings.
Case Study Results—
All of the case study tests met the RCRA particulate
emissions standard of 0.08 gr/dscf, with the exception
of two runs during the Florida Steel TSCA trial
burns on the commercial-scale unit, four runs during
the Twin Cities pilot-scale tests, and one run during
the International Paper pilot-scale tests. For these
specific runs, the data indicate the following reasons
for the failure of the operation to meet the regulatory
standard.
• Florida Steel - During one run the scrubber mal-
functioned. The reports do not provide any addi-
tional details. A second run was conducted on an
Askarel-spiked feed containing more than 1.9 wt%
total chlorine in the feed. The case study data indi-
cate that the high chlorine content contributed to
the overloading of the venturi scrubber system
and the high particulate emissions. The TSCA
permit restricts the chlorine content in the feed to
a maximum of 0.9 wt% with a feedrate of 133 Ib/h
of chlorine to minimize the problem.
• Twin Cities - During three runs, the particulate
emissions exceeded the regulatory standard be-
cause of plugging/corrosion in the scrubber ventu-
ri and toner scrubbing nozzles. During a fourth
run, the 100 Ib/h feedrate to the pilot-scale unit
apparently overloaded the scrubbing system.
When the feedrate was reduced to 90 Ib/h, the op-
eration was satisfactory.
• International Paper — High particulate emissions
during one of the tests was the result of soot for-
mation caused by an improper control of oxygen in
the primary combustion chamber.
Summary of Results—
Based on the above discussions, there are several op-
erations where both the pilot-scale and commercial
units have failed to meet the RCRA particulate emis-
sions standard. The commercial unit now operating
at LaSalle Electric is equipped with a totally new
high efficiency scrubbing system that replaced the
poorly performing system originally employed at
Peak Oil, as discussed above. Another commercial
unit [22] is planning to incorporate a similar high ef-
ficiency scrubbing system to obtain high efficiency
over a wider range of gas flows and fines loadings.
Pretest analyses of the waste-feed for particle size
and elemental (halides, S and P), metals (heavy and
alkali), and organic species and concentrations are
necessary to identify contaminants that may cause
particulate emissions problems. As discussed in a
later subsection, treatability testing that includes
bench- and/or pilot-scale testing and thermogravime-
tric analyses (TGA) are also required to define poten-
tial particulate emissions loadings.
Characteristics of the Furnace Ash
SITE Demonstration Results—
The total PCB concentrations in the furnace ash
sampled during the normal operations of the units
were less than 1 ppm, which is below the TSCA guid-
ance level of 2 ppm, indicating effective PCB decon-
13
-------
lamination of the waste feeds containing levels up to
35.2 ppm. Additional tests were performed during
the Demode Road pilot-scale program at PCC operat-
ing temperatures lower than the normal 1,600°F and
• ranging from 900° to 1,400°F. A series of operating
conditions were imposed on the unit, including shut-
ting off the combustion air to simulate pyrolytic con-
ditions, and varying residence time in the PCC. At
900°F with no combustion air flow, two samples of
furnace ash exceeded the TSCA guidance level of 2
ppm PCBs, containing 3.396 and 2.079 ppm PCBs. At
this low PCC temperature and pyrolytic condition,
these higher total residual PCB levels in the furnace
ash may have been the result of the incomplete com-
bustion of PCBs in the feed. This is further substanti-
ated by residual levels of TCDF present in the same
furnace ash samples. (No dioxins or furans were de-
tected in any of the other Demode Road demonstra-
tion furnace ash samples.)
The concentrations of metals in the furnace ash were
similar to the concentrations in the waste feed and
indicate that the mass flow of these species remains
with the high mass flow of furnace ash that exits the
Shirco unit.
The analyses detected several volatile and semivola-
tile organic compounds in the furnace ash at average
concentrations less than 50 ppb. The compounds in-
clude: halomethanes; aromatic organics (including
benzene, toluene, chrysene, fluoranthene, phenan-
threne, and pyrene); and oxygenated hydrocarbons
(includi ng phenol, phthalates, and p-chloro-m-
cresol). Methylene chloride, acetone, and methyl-
ethyl-ketone, tetrachloroethylene and trichloro-
ethene were also detected but may be present due to
laboratory contamination.
Except for the TCDF detected during one of the De-
mode Road pilot-scale unit runs and discussed above,
no dioxins or furans were detected in the furnace ash
samples.
Case Study Results—
All of the case study data confirmed the satisfactory
results obtained during the SITE demonstrations,
except for two runs conducted during the
Florida Steel TSCA trial burns, where the PCB con-
tent in the furnace ash exceeded the 2 ppm TSCA
guidance level. It was determined that the quality of
the auxiliary fuel suffered as a result of the poor han-
dling during addition to the waste, which in turn af-
fected the efficiency of PCB removal.
Summary of Results—
In general, the concentrations of PCBs and other
organics in the furnace ash met all regulatory
standards. Several volatiles and semivolatiles were
detected, including halomethanes, aromatics, and
oxygenated hydrocarbons. Metals concentrations
approximated those in the feed and indicated that
the majority of the metals remain with the furnace
ash exiting the unit.
Mobility of Heavy Metals — Comparison
to EP Tox and Proposed TCLP Toxicity
Characteristic Standards
SITE Demonstration Results—
In order to determine whether heavy metals would
leach from the waste feed and Shirco byproducts, EP
Tox and TCLP tests were conducted on the feed, fur-
nace ash, scrubber water, and scrubber solids and
compared to their respective toxicity characteristic'
standards.
For the Peak Oil demonstration all of the results of
the EP Tox tests on the feed and the furnace ash ex-
ceeded the 5 ppm toxicity characteristic standard for
lead (24-57 ppm). Two samples of the feed exceeded
the proposed TCLP toxicity characteristic standard
of 5 ppm for lead (8.6 ppm and 35 ppm). All of the fur-
nace ash samples passed the TCLP standard. For the
other heavy metals, all of the results were below
their respective toxicity standards.
For the Demode Road demonstration all of the re-
sults were below the EP Tox and proposed TCLP tox-
icity characteristic standards — 5 ppm arsenic, 100
•ppm barium, 1 ppm cadmium, 5 ppm chromium, 5
ppm lead, 0.2 ppm mercury, 1 ppm selenium, and 5
ppm silver — except for 1 feed sample at 7.0 ppm lead
(TCLP) and 1 furnace ash sample at 6.2 ppm lead
(TCLP).
Case Study Test Results—
EP Tox tests were conducted on the furnace ash pro-
duced during the Florida Steel TSCA trial burn and
the Brio pilot-scale tests. In both cases, the results
were less than the EP Tox toxicity characteristic
standard for heavy metals. No data was provided for
the waste feed.
Summary of Results—
Despite concentrations of heavy metals in the waste-
feed and furnace ash as high as 5,900 ppm and 7,100
ppm (lead) respectively, in most cases the
concentrations of metals in the EP Tox and TCLP
leachates met their respective toxicity characteristic
standards.
Mobility of Heavy Metals — Comparison
of Feed and Furnace Ash
SITE Demonstration Results—
In order to determine whether heavy metals, particu-
larly lead, would leach from the furnace ash pro-
14
-------
duced in the Shirco unit, EP Tox and TCLP tests
were conducted to determine the mobility of heavy
metals from the furnace ash as compared to the feed.
For the Peak Oil demonstration the EP Tox results
for lead in the leachate ranged from 24 to 57 ppm for
the feed and 25 to 46 ppm for the furnace ash. The
TCLP results ranged from 2.5 to 35 ppm for the feed
and 0.008 to 0.84 ppm for the furnace ash.
For the Demode Road demonstration the initial EP
Tox analyses for lead in the leachate ranged from
0.05 to 0.67 ppm for the feed and 0.05 to 4.10 ppm for
the furnace ash. The initial TCLP analyses ranged
from 0.35 to 1.80 ppm (with one sample at 7.0 ppm)
for the feed and 0.05 to 4.10 ppm (with one sample at
6. 20 ppm) for the furnace ash. When several samples
were retested to verify the results, the concentra-
tions of lead in the EP Tox leachates (4.9 ppm feed,
3.0 ppm furnace ash) were higher than during the
initial tests, and in direct reversal to the original
data, exceeded corresponding TCLP leachate concen-
trations (2.8 ppm feed, 1.4 ppm furnace ash).
A comparison of the SITE demonstration toxicity
characteristic data indicates contrasting results.
Whereas the EP Tox results from the Peak Oil data
agreed with the results and conclusions from the De-
mode Road tests, the TCLP tests resulted in lower
concentrations of lead in the leachates of the furnace
ash as compared to the feed, indicating reduced mo-
bility of lead from the furnace ash as compared to the
feed as a result of thermal treatment. These mixed
results as compared to the Demode Road tests may be
the result of differences in the test procedures and
the alkalinity of the waste feed (the waste feed at
Peak Oil was pretreated with lime), which caused a
difference in the pH environment that is sufficient to
affect the solubility and leaching characteristics of
heavy metals, particularly lead.
Summary of Results—
The case study data did not provide any additional
information to support the SITE demonstration re-
sults. The results do not show any trend or evidence
that heavy metals, particularly lead, have reduced
mobility in the furnace ash as compared to the feed
as a result of the thermal treatment through the
Shirco unit.
Characteristics of the Scrubber Water
SITE Demonstration Results—
During the Peak Oil and Demode Road demonstra-
tions, no PCBs, dioxins, or furans were detected in
the scrubber water leaving the unit. Trace levels of
phthalates and p-chloro-m-cresol were detected at
concentrations less than 100 ppb. High levels of ben-
zene and toluene were detected during the Demode
Road demonstration but they were also present in
the scrubber makeup water as an external contami-
nant. The major concentration of contaminants was
found in the scrubber water solids associated with
the Peak Oil demonstration. Significant concentra-
tions of metals — including aluminum, calcium, iron,
lead, sodium, sulfur, and zinc at levels at-or-above 1
wt% — were detected. These high concentrations are
indications of the large flow of contaminants to the
venturi scrubber system and tend to confirm the par-
ticulate emissions problems occurring at the Peak
Oil venturi scrubber system as discussed above.
Even with the high scrubber-water blowdown-rate
with its associated high contaminant concentrations,
particulate emissions were above the RCRA stan-
dard and contained very high metals concentrations.
The scrubber water blowdown required clarification,
treatment with activated carbon, and pH adjustment
in a holding tank prior to discharge to the POTW.
Summary of Results—
The case study data did not indicate any significant
levels of PCBs, dioxins, furans, or other volatile and
semivolatile organics in the scrubber water. The
Peak Oil data, with its significant scrubber system
overloading, reemphasizes the need to pretest and
analyze the waste matrix to assess its impact on the
scrubber system and its effluents.
Operations
SITE Demonstration Results—
There were no problems associated with the opera-
tion of the Demode Road pilot-scale unit that would
impact on the ability of a commercial Shirco unit to
achieve a satisfactory level of continuous operating
performance.
The Peak Oil commercial unit, which was the first
application of a full- scale commercial unit at a Su-
perfund site, exhibited many problems, associated
mainly with feed preparation, materials handling,
and emissions control. On an overall schedule basis,
the unit remained at the site, after installation and
startup, for a period of 286 days. Based on a continu-
ous capacity of 100 ton/d and a total processed waste
feed tonnage of 7,110 tons, the unit ideally only re-
quired 71 operating days based on a 100% utilization
factor. The actual utilization factor, based on the
above, is 24%.
Preoperations testing and evaluation of alternative
feed preparations and materials handling systems
based on the physical and chemical characteristics of
the site waste matrix and the acceptable waste feed
specified for the unit are mandatory and cannot be
understated. These issues were not examined to the
extent necessary for a successful Peak Oil operation,
where the combination of an acidic, oily, clumping
15
-------
sludge, and an extraordinary high-lead-contaminant
concentration provided a serious challenge to the op-
eration of the unit.
Case Study Results—
Several of the case studies encountered problems, in-
corporated design modifications, and provided infor-
mation on overall system reliability and utilization.
• Florida Steel pilot-scale tests — Because of a low-
Btu waste feed, the primary combustion chamber
could not maintain the desired operating tempera-
ture at maximum electrical power input. Preo-
perations testing and analysis is required to de-
fine an overall heat and material balance prior to
the commitment of a commercial-scale unit.
The analysis of one run that did not meet the
TSCA DRB standard for PCBs indicated that a
low oxygen level in the secondary combustion
chamber will affect the unit's ability to meet this
standard.
• Florida Steel TCSA trial burns — Based on initial
tests conducted on the commercial-scale unit, sev-
eral modifications were implemented during the
TSCA trial burn. These included modifications to
the ash collection system, the fines ash system
that collects the material that falls through the
conveyor belt, the ash quench module, scrubber
blowdown operating procedures, and the feed-
hopper feeding mechanism. The air compressor
was also replaced with one of high capacity. De-
tails of these changes were not made available, al-
though it was indicated that the fines-ash collec-
tion system was modified to transfer the ash back
to the primary chamber via a closed loop design to
preclude exposure to PCDFs.
• Florida Steel commercial-scale cleanup — Case
study data indicate that a major design change in-
volved the replacement of the conveyor belt with a
new smaller-gauge belt that precluded substantial
"sift-through" of smaller feed particles and sand.
Attention to the physical characteristics of the
waste matrix and its impact on the unit operation
should have eliminated this problem.
The initial operating factor for the unit was ap-
proximately 50%. The onstream time continuous-
ly increased as the unit operation stabilized, with
the final month of operation sustaining a 91% uti-
lization factor. The overall project utilization rate
was 61%.
• LaSalle Electric commercial cleanup — Case
study data indicate that the commercial operation
is achieving utilization factors of 80%-90%. The
waste preparation system, including the power
screen, is performing well, although long slender
nails and spikes can pass through the screen and
into the unit. This operation on a relatively-dry
discrete soil is in sharp contrast to the poor opera-
tion of this same unit at the Peak Oil site where
an oily, sludge waste matrix caused significant
waste feed preparation and materials handling
problems.
• Brio and International Paper pilot-scale tests —
In both case studies, waste feed that was either
lumpy and clay-like, or tar-like with moisture and
clay-like adhesive qualities, caused problems in
handling and feeding the material to the pilot-
scale unit. These tests are a clear warning that the
use of a commercial-scale Shirco unit at such sites
without a careful and comprehensive waste prep-
aration and materials handling design will not be
successful.
Summary of Results—
The operation of a commercial Shirco unit design
requires strict adherence to preoperations testing
and analysis to characterize the waste matrix and
determine the required methodology for feed prepa-
ration and materials handling. With an acceptable
feed matrix, and based on recent design changes to
the current commercial units, the results indicate
that recent overall project utilization rates of more
than 60% have been achieved. Intermittent
operations over 1-3 month periods have achieved
rates up to 90%.
Optimum Operating Conditions
SITE Demonstration Results—
The Peak Oil commercial unit was being operated at
a remedial action to meet the objectives of the clean-
up at satisfactory regulatory performance standards
under optimum operating conditions. During the
tests, the high Btu feed produced an autogenous com-
bustion condition that allowed intermittent oper-
ations at specified temperatures without the input of
electrical power to the infrared heating rods in the
primary combustion chamber.
The Demode Road pilot-scale demonstration includ-
ed a series of runs that were conducted to examine
the effect of varying operating conditions on unit per-
formance and energy consumption. Highlights of the
results are as follows:
• A reduction in the PCC operating temperature
from 1,600° to 1,200°F reduced the average PCC
power usage by 48% from 0.2294 to 0.1200 kWh/lb
feed.
• A reduction in the SCC operating temperature
from 2,200° to 1,800°F reduced the average pro-
pane fuel consumption by 51% from 3,997 to 1,952
Btu/lb feed. It should be noted, however, that the
TSCA regulations require an SCC operating tem-
16
-------
perature of 2,200°F for the incineration of PCB
contaminated waste.
• The use of 3 wt% fuel oil to supplement the heat-
ing value of the feed further decreased P.CC power
usage by 26%-67% at PCC operating temperatures
of 1,600°F and 1,200°F, respectively, with accom-
panying increases in overall feedrate of 32% and
26%.
The addition of the fuel oil increased the average
HHV of the feed from 210 to 588 Btu/lb. This in-
crease in heating value is equivalent to a savings
of 0.11 kWh/lb feed. Reductions in power when
fuel oil was added to the feed were 0.07 and 0.09
kWh/lb feed, which closely approximates the cal-
culated value of 0.11 kWh/lb feed based on heating
value.
The costs for fuel oil and its attendant facilities
still must be examined for specific applications to
determine the cost effectiveness of a fuel oil addi-
tive to the feed. The cost of power, the moisture
content of the feed, the total heating value of the
feed, PCC residence time, and the overall PCC de-
sign heating input will all impact on the necessity
for and the quantity of the addition of fuel oil to
the feed.
• The results did not provide any trend or show any
change in the residual levels of the heavy metals
and organics in the furnace ash versus the levels
in the feed as the operating conditions were varied
and PCC operating temperatures maintained at
1,200° to 1,600°F. At an abnormally low PCC oper-
ating temperature of 900°F, without the input of
combustion air to simulate non-oxidizing or py-
rolytic combustion conditions, total. PCB and
TCDF concentrations in the furnace ash in-
creased. The increases may indicate that these
PCC conditions led to incomplete desorption or in-
cineration of PCB and to the production of TCDF
from the incomplete combustion of PCBs in the
feed.
Case Study Results—
Several of the pilot-scale case studies — including
Florida Steel, Brio, Times Beach, and Simulated
Creosote Pit — were conducted at varying operating
conditions. Tests were performed at PCC operating
temperatures ranging from 1, 600° to 1,800°F, SCC
operating temperatures ranging from 1,800° to
2,200°F, and PCC residence times ranging from 15 to
45 min. In general all of the results met applicable
operating and regulatory performance standards; no
data were presented on energy consumption.
Summary of Results—
The Demode Road pilot-scale SITE demonstration in-
dicates that the operating conditions of the Shirco
unit can be varied within limits to provide efficient
energy consumption and unit operation and meet all
applicable unit and regulatory performance stan-
dards. Key parameters that can be varied include
fuel oil addition to the waste feed, PCC and SCC op-
erating temperatures, combustion air flows, and
PCC residence time. Preoperations waste-feed- ma-
trix laboratory analysis and thermogravimetric
analyses — including bench- or pilot-scale testing —
will establish the recommended range of operating
parameters for the commercial unit to assure opti-
mum operation within regulatory requirements.
Environmental Regulations and
Comparison with SHIRCO Performance
Introduction
Section 121 of CERCLA (Comprehensive Environ-
mental Response, Compensation, and Liability Act)
requires that, subject to specified exceptions, remedi-
al actions must be undertaken in compliance with
applicable or relevant and appropriate requirements
(ARARs), federal laws, and more stringent promul-
gated state laws (in response to releases or threats of
releases of hazardous substances or pollutants or
contaminants as may be necessary to protect human
health and the environment).
The basic ARARs of interest are outlined in the In-
terim Guidance on Compliance with ARAR, FRL-
3249-8, Federal Register, Vol. 52, pp. 32496 et seq.
These are:
• Performance-, design-, or action-specific require-
ments. Examples include RCRA incineration
standards and Clean Water Act pretreatment
standards for discharges to POTWs. These re-
quirements are triggered by the particular reme-
dial activity selected to clean a site.
• Ambient/chemical-specific requirements. These
set health-risk-based concentration limits based
on pollutants/contaminants, e.g., emissions limits
and ambient air quality standards (NAAQS). The
most stringent ARAR must be followed.
• Location Requirements. These set restrictions on
activities because of site location and environs,
e.g., federal/state siting laws.
Superfund regulations in 40 CFR 300.68(a)(3) state
that federal, state, and local permits are not required
for Superfund-financed remedial actions or remedial
actions taken pursuant to federal action under Sec-
tion 106 of CERCLA. However, -several states, in-
cluding Connecticut, Maine, Massachusetts, New
Hampshire, Rhode Island, Vermont, New Jersey,
New York, Pennsylvania, and California, have inde-
pendent state Superfund laws that may be more
stringent than the federal laws, and thereby have
17
-------
primacy. In addition, some state and local authori-
ties — such as the California South Coast Air Qual-
ity Management District (SCAQMD) and Depart-
ment of Health Services (DOHS) — insist that all po-
tential Superfund-site incinerators must be permit-
ted like any other incinerator — in apparent dis-
agreement with the federal regulation cited above.
Deployment of a Shirco unit will therefore be affect-
ed by three main levels of regulation:
• Federal EPA incinerator, air, and water-pollution
regulations
• State incinerator, air, and water-pollution rules
• Local regulations, particularly Air Quality Man-
agement District (AQMD) requirements.
Federal EPA Regulations
ARARs—
As discussed in the interim guidance document on
compliance with ARAR (FRL -3249-8), a require-
ment under other environmental laws may either be
"applicabl e" or "relevant and appropriate" to a re-
medial action, but not both. A two- tier test may be
applied: first, to determine whether a given require-
ment is applicable; then, if it is not applicable, to de-
termine whether it is nevertheless relevant and ap-
propriate.
"Applicable requirements" means those cleanup
standards, standards of control, and other substan-
tive environmental protection requirements, crite-
ria, or limitations promulgated under federal or state
law that specifically address a hazardous substance,
pollutant, contaminant, remedial action, location, or
other circumstance at a Superfund site.
"Applicability" implies that the remedial action or
the circumstances at the site satisfy all of the juris-
dictional prerequisites for a requirement. For exam-
ple, the hazardous-waste incinerator regulations
would apply for incinerators operating at Superfund
sites containing listed or characteristic hazardous
wastes.
"Relevant and appropriate requirements" means
those cleanup standards, standards of control, and
other substantive environmental protection require-
ments, criteria, or limitations promulgated under
federal or state law that, while not "applicable" to a
hazardous substance, pollutant, contaminant,
remedial action, location, or other circumstance at a
Superfund site, address problems or situations
sufficiently similar to those encountered at the
Superfund site that their use is well suited to the
particular site. For example, if a Superfund site
contained no specifically listed or characteristic
hazardous wastes, the hazardous-waste incinerator
regulations might still be considered relevant and
appropriate.
Incinerator Regulations— ,
Resource Conservation and Recovery Act (RCRA)--
The federal hazardous-waste incinerator regulations
are considered either "applicable" or "relevant and
appropriate" to the incineration of a Superfund site
waste. These regulations establish hazardous-waste
incineration performance standards under RCRA, as
detailed in 40 CFR 264 Subpart O. These regulations
are applicable to incineration of hazardous wastes at
a Superfund site, and may be deemed relevant and
appropriate to the incineration of some wastes that
are not specifically listed in 40 CFR Part 261.
The important incinerator regulations are:
• Performance standards: Section 264.343
• Operating requirements: Section 264.345
• Monitoring and inspections: Section 264.347
• Rulemaking petitions (delisting): Sections 260.20
and 260.22
Under the current version of these regulations, an
incinerator will be required to:
• Achieve a DRE of 99.99% for each principal organ-
ic hazardous constituent (POHC) in the waste feed
•• Control HC1 emissions to the larger of 1.8 kg/h (4
Ib/h) or 1% of the stack HC1, prior to entering any
pollution control equipment
• Limit particulate emissions to less than 180
mg/dscm (0.08 gr/dscf), corrected to 7% 62
• Continuously monitor combustion temperature,
waste feedrate, and an indicator of combustion gas
velocity
• Continuously monitor CO in the stack gas
• Produce byproducts that are not hazardous and
can be delisted because they do not exhibit hazard-
ous characteristics or contain the originally-listed
hazardous constituents; or contain the originally-
listed hazardous constituents at relatively low
concentration; or contain the listed constituents in
an immobile form.
Toxic Substances Control Act (TSCA)—Incineration
of polychlorinated biphenyls (PCBs) and PCB-
contaminated materials is regulated under TSCA as
detailed in 40 CFR 761.70. These regulations estab-
lish performance standards for non-liquid PCB
waste incineration that relate to the following fac-
tors:
• Demonstrating that mass air emissions from the
incinerator are no greater than 0.0001. g PCB/kg
18
-------
of the PCBs in the feed to obtain a ORE of at least
9 .9999% (also known as "six 9s")
• Demonstrating a combustion efficiency (CE) of at
least 99.99%, where:
C(X
co2 + co
xlOO
where:
CC>2 = concentration of carbon dioxide in stack
gas
CO = concentration of carbon monoxide in stack
gas
• Measurement and recording (at intervals no long-
er than 15 min.) of the rate and quantity of PCBs
fed to the unit
• Continuously measuring and recording the tem-
perature of the incineration process (combustion
chambers)
• Monitoring and recording the concentrations of O2
(continuously), CO (continuously), and CO2 (peri-
odically at a specified frequency) in the stack
emissions, whenever PCBs are burned
• A system to automatically stop the PCB feed
whenever the monitoring operations specified for
O2, CO2, and CO fail
• Monitoring stack emissions for 0%, CO, COg, NOX,
HC1, total chlorinated organics, PCBs, and total
particulate matter, when the incinerator is first
used for the disposal of PCBs, or when the inciner-
ator has been modified in a manner that may af-
fect emissions
• Using water scrubbers to control HC1. For approv-
al, EPA requires that the HC1 removal systems
demonstrate a removal efficiency of 99%
• Measurement of the stack emissions for chlorinat-
ed dibenzodioxins and dibenzofurans
• Demonstrating particulate matter emissions lev-
els of < 180 mg/dscm (0.08 gr/dscf), when corrected
to 7% oxygen
• Demonstrating total PCB concentration in scrub-
ber water and furnace ash of < 2 ppm
Table 1 summarizes the results of the SITE demon-
strations of the Shirco technology on a commercial
transportable unit and a pilot-scale unit at the Peak
Oil and Demode Road Superfund sites, respectively.
The table compares these results to the RCRA and
TSCA performance standards for hazardous waste
incineration, as discussed above.
Water Regulations—
Provisions of the Safe Drinking Water Act also apply
to remediation of Superfund sites. CERCLA
121(d)(2)(A) and (B) explicitly mention three kinds of
surface water or groundwater standards with which
compliance is potentially required — maximum con-
taminant level goals (MCLGs), federal water quality
criteria (FWQC), and alternate concentration limits
(ACLs) where human exposure is to be limited. This
Table 1. Site Demonstration: Comparison of Results to Environmental Standards
Performance Standards
POHC
ORE, %
HCI Stack
emissions
Particulate emissions,
mg/dscm
CE, %
Total PCBs -scrubber
RCRA
Standard
> 99.99
< 1 .8 kg/h
or >99% Removal
<180
NA
NA
TSCA
Standard
> 99.9999
(PCBs)
2:99%
Removal
<180
>99.9
<2
Peak Oil
Demonstration
> 99.99(a)
(PCBs)
>99%
Removal
171-358
>99.9
0.01-0.9
Demode Road
Demonstration
>99.99
(PCBs)
< 0.001 kg/h
7-69
>99.9
0.003-3.396
water and furnace ash,
ppm
Monitoring emissions
PCB feed cutoff
Required
NA
Required
Required
Yes
Yes
Yes
Yes
(a) DRE calculation based on less-than-detectable PCB concentrations in stack gas.
(b) Venturi scrubber overloading. Unit's venturi scrubber system has been replaced with a more efficient and high impact
system.
(c) Operating at abnormally low PCC operating temperature.
19
-------
subsection of CERCLA describes these requirements
and how they may be applied to Superfund remedial
actions. The guidance is based on federal require-
ments and policies; more stringent promulgated
state requirements (such as a stricter classification
scheme for groundwater) may result in application of
even stricter standards than those specified in feder-
al regulations.
Disposal of scrubber water (blowdown) at Peak Oil
required pretreatment (n eutralization and precipi-
tation/sedimentation) prior to disposal at a POTW.
The scrubber water at Demode Road was drummed
and disposed of at an accredited treatment facility.
State and Local Regulations
In addition to the federal regulations discussed
above, the Federal Prevention of Significant Deteri-
oration (PSD) and New Source Review (NSR)
regulations promulgated under the Clean Air Act
and administered by the states will impact the
operation of the Shirco unit through the emissions
monitoring control and process requirements, or
through the permitting process in areas that require
permits to install and operate. In addition to these,
there are several local regulations that govern
incinerator operations, because incinerators are
combustion devices (emissions sources). Many of
these state and local emissions regulations are more
stringent than EPA rules, and the cognizant
regulatory agencies have primacy. Pressure should,
therefore, be anticipated from state and local
authorities in relation to NOx emissions, for
example.
There are six basic sources of potential regulations at
the state and local levels:
• Permits to construct/operate
—Best available control technology (BACT) trig-
gers for stationary sources or units
—Cumulative or offset triggers
» New Source Review
—BACT trigger levels and BACT designations
— Offset triggers
* Prevention of Significant Deterioration
—BACT controls
—Increment limitations
• General prohibitions on emissions levels
• Source-specific standards on emissions levels (cur-
rently there are none, but the mechanism exists)
• Nuisance rules
Discharge permits may also be required from the re-
gional water quality board.
The major regulatory requirements include permits
to install and operate as well as NSR/PSD reviews as
appropriate. Many states, such as California, New
Jersey, Pennsylvania, Ohio, New York, Texas, and
Virginia, require some form of NOX control and/or
monitoring for CO, unburned hydrocarbon (UHC),
and NOX. These regulations apply to all incinerators.
The required control is a BACT, reasonably avail-
able control technology (RACT), or lowest achievable
emission rate (LAER) control, and ranges from ex-
emptions for short burns of small amounts of nonha-
zardous wastes in transportable incinerators to very
stringent criteria pollutant control. Offsets may-also
be required in areas that are in nonattainment for
NO2, such as the South Coast AQMD (SCAQMD) in
California; or nonattainment for ozone, such as the
New York metropolitan area.
Most of the relevant regulations specify an emissions
rate or level that may not be exceeded or that will
trigger corrective or punitive measures. For exam-
ple, the PSD NOx trigger is 100 tons/yr for new sour-
ces. By contrast, the nuisance rules are catch-all ru-
les that seek to prevent injury or annoyance to any
considerable number of persons or to the public. Al-
though these nuisance rules do not appear aggres-
sive or overbearing, the regulatory power of the pub-
lic cannot be overstated. Public opposition can be
more effective in stalling an incineration project
than federal, state, or local regulation. Indeed, public
opposition can stall a project already approved and
permitted by the authorities. The process for grant-
ing permits to install and operate usually has provi-
sions for public input, especially for waste treatment
projects. Permitting can easily become the most ex-
pensive and time-consuming part of deploying any
incineration treatment project.
Monitoring Requirements
The operation of the Shirco unit will be required to
monitor CO and NOX emissions. The NOX require-
ments will likely come from state and local AQMD
regulatory pressure for NOX control and, in some
areas, for ozone reduction. Continuous monitoring
will likely be required. The CO requirement will
stem from the federal and state incinerator regula-
tions calling for continuous monitoring. State and lo-
cal AQMD emissions limits also exist for CO.
Incineration treatment systems are also required to
continuously monitor such variables as combustion
temperature, waste feedrate, and an indicator of
combustion gas velocity. Further, if the waste con-
tains sulfur, scrubbing and SO2 monitoring may be
20
-------
required by the air regulations. Other sampling,
analysis, equipment monitoring, and inspections
may be required as outlined in 40 CFR 264 Section
347 and 40 CFR 761.70.
Incineration systems are required to observe blow-
down discharge and furnace-ash disposal require-
ments during operation and at closure. Unless the
operator can demonstrate according to 40 CFR
261.3(d) that the furnace ash removed from the in-
cinerator is not a hazardous waste, he must manage
it in accordance with the applicable requirements of
40 CFR Sections 262 through 266. Under TSCA, if
the furnace-ash PCB concentration is > 2 ppm it still
is subject to the appropriate provisions of 40 CFR
671. Even for nonhazardous discharge, local water-
quality board regulations, as well as federal and
state regulations, will likely be enforced either as
"applicable" or as "relevant and appropriate". Fur-
ther guidance on these regulations are available in
CERCLA Section 121(d)(2)(A) and (B), as well as 40
CFR Section 35.
Waste Characteristics and Their
Impact on Performance of the
Technology [19]
Based on the evaluation of the technology per-
formance discussed previously, information on both
the physical and chemical characteristics of the
waste matrix is necessary to determine the suit-
ability of that waste for thermal processing using the
Shirco technology and the possible need for waste
preparation and pretreatment.
Preoperations waste-feed characterization and
laboratory analysis, and pilot or bench-scale testing
including thermogravimetric analyses (TGA) are
mandatory in order to define the waste feed matrix
and its impact on the Shirco unit. The unit has
pretreatment and waste-feed-preparation require-
ments, metallurgical requirements and/or limita-
tions, potential design limitations particularly in the
offgas treatment section, and operating require-
ments, all of which are impacted by the waste
matrix.
Table 2 presents a range of waste characteristics
suitable for processing in the Shirco unit. In order for
specific-site-waste matrices to conform to these
waste characteristics, preprocessing may be re-
quired. This includes sizing, classifying, screening,
dewatering, soils blending, and/or lime addition pri-
or to processing to ensure a solid/semi-solid matrix
with characteristics within the treatment range of
the unit. Pure liquids can also be processed if blended
with a suitable carrier, such as soil or vermiculite, to
form a semi-solid waste matrix. Additional materials
handling discussions are presented in a later subsec-
tion.
Table 2. Applicable Range of Waste Characteristics
Characteristics Applicable Range
Morphology
Particle size (diameter)
Moisture content
Density
Heating value
Organics (including POHCs)
Chlorine
Sulfur
Phosphorous
PH
Alkali metals
Heavy metals
Soil/solid
Semi-solid
Oily-sludge/solid
5 microns - 2 in.
0-50 wt% (no free liquids or
free-flowing sludges)
30-130lb/ft3
0-10,000 Btu/lb
0-100 wt% (determined by
preoperation testing)
0-5 wt%
0-5 wt%
0-300 ppm
5-9
0-1 wt%
0-1 wt% (determined by
preoperation testing)
Physical Characteristics
Physical characteristics of the waste feed matrix
determine the pretreatment and preparation
required to produce a waste feed that is acceptable to
the Shirco unit. Key physical characteristics include
morphology, particle size, rheology, moisture
content, density, and heating value.
Feed material can be blended to reduce free liquids
or to mitigate other incompatible characteristics
(such as unsuitable pH or low heating value), and
can be pretreated with size reduction equipment to
allow most feeds to be processed. Feedstocks may
require screening, size reduction, and mixing.
Pretreatment equipment typically represents 10% of
the system costs. ECOVA requires that feedstock
materials be tested in their laboratories in order to
identify any physical or chemical pretreatment and
preparation requirements.
Morphology, Particle Size, Rheology, and Moisture
Content—
The physical state of the waste matrix is a key
parameter in the successful operation of the Shirco
unit. Waste matrix pretreatment and preparation
activities prior to feeding the unit will be affected by
the extent to which the waste matrix is a soil, solid,
sludge, slurry, liquid, or a combination of these; its
moisture content; as well as any associated rock,
21
-------
clay, scrap metal, and other extraneous materials.
These physical characteristics will also impact on the
performance of the primary combustion chamber.
Liquids not encapsulated by the feed — that is free
liquids or free-flowing sludges — and solids sized
below 5 microns, cannot be contained on the convey-
or belt; they will pass through the belt's screen open-
ings (the belt is fabricated of woven metal strands).
High-moisture-content feeds will also require addi-
tional electrical-power input to overcome the heat
sink effect of the moisture. Liquids that are capable
of being contained for short periods can be processed,
since liquid components will be volatilized within
three minutes, after exposure to the infrared rods.
Materials greater than 2 in. or less than 5 microns
cannot be processed by the Shirco system due to
various considerations. One constraint is the height
of the bed on the conveyor, which is restricted both
by the mechanical design of the unit's feed system
and by thermodynamic considerations. On materials
greater than 2 in. or clumpy sludge-like materials,
diffusion of contaminants through the particles and
through the bed to expose contaminants to the
infrared heat is diminished. In addition, at less than
5 microns there is the possibility of very light fines
being generated that would be carried through the
system and possibly cause an overload on the
emissions control system or problems with the ash
handling system. Except for fines, the smaller the
feed material diameter, the shorter is the residence
time requirement.
The cakebreaker device and conveyor belt present
mechanical stress limitations that result in the
determination of a 100 ton/d optimal processing
design. Material feed size limitations therefore
cannot be eliminated by design changes. Conveyor
belt widths and lengths already have been design-
optimized and cannot be exceeded. Material larger
than 2 in. physically can pass by the cakebreaker,
but will not be exposed thoroughly to the heat and
thus will not be fully detoxified.
Density—
Unit capacity, as measured on a weight basis, will be
determined by the density of the feed as it relates to
the volume of waste that is excavated and the
volume of waste that can be effectively transported
through the primary combustion chamber on the
conveyor belt at a maximum bed depth of 2 in.
Heating Value—
Heating value of the feed affects both feed capacity
and energy consumption of the Shirco unit. At high
heating values, the electrical power input to the
primary combustion chamber can be reduced, or
eliminated if autogenous combustion similar to the
Peak Oil operation [1] is occurring.
Conversely, at low heating values, fuel oil can be
added to the waste to increase the heating value to a
level that will accommodate the energy balance and
maximum electrical power input of the primary
combustion chamber.
Unit Operations—
The pilot-scale unit employs manual feed prepa-
ration and handling that minimizes the effect of
unsuitable waste-matrix physical characteristics on
the unit's operation.
Based on the full-scale operations at Peak Oil [l]-and
Florida Steel [4-6], many of the operations problems
were attributable to these physical characteristics
discussed above. Improperly prepared oily-waste-
sludge at Peak Oil had a severe impact on all aspects
of materials handling at the Shirco unit. Once a
proper screening device (power screen) was installed,
the overall operation of the unit improved. Initial
operations at Florida Steel were impacted by a PCC
conveyor-belt pore-space that was larger than the
pilot-scale PCC conveyor belt, which allowed the fine
Florida sand to sieve through the belt and overload
the fines collection system. Once the belt was
replaced with a belt that matched the feed's particle
size, the overall operation of the unit improved.
Chemical Characteristics
The chemical characteristics of the waste feed define
the levels of contaminants and chemical environ-
ment that is imposed on the unit. Key chemical
characteristics include: organic and POHC species
and concentrations, halogens, sulfur and phos-
phorous, pH, alkali metals, and heavy metals. These
characteristics will impact on the Shirco unit design
and operation as follows:
• Feed preparation requirements, including pH
neutralization and chemicals stabilization.
• Combustibility of the waste feed and destruction
ofthePOHCs.
• Corrosion prevention requirements, and metal-
lurgical and refractory considerations.
• The type and efficiency of the air-pollution control
system and slagging potentials through the unit.
• Furnace-ash storage and disposal requirements,
and scrubber-water treatment and disposal re-
quirements.
Once the unit design and heat- and mass-balance
have been defined, unit operating conditions in the
22
-------
primary and secondary combustion-chambers
provide environments consistent with applicable
incineration standards to ensure (1) oxidation of
most organic contaminants to non-toxic products, (2)
acceptable stack emissions, and (3) acceptable levels
of organics, POHCs, and PICs in the furnace-ash and
scrubber-water effluent. Inorganic contaminants,
including halogens and sulfur, alkali metals, and
heavy metals are not necessarily detoxified and can
interfere with or adversely impact either the
environment or the performance of the Shirco unit.
Halogens, Sulfur, Phosphorous, and pH—
Acid gases are formed when waste feed containing
chlorine, fluorine, bromine, sulfur, and phosphorous
are thermally treated. The presence of halogens,
sulfur, and phosphorous in the waste and the
subsequent formation of acid gases result in:
• Corrosive attack of refractory and metal com-
ponents throughout the unit.
• Increased costs for acid gas neutralization.
• Possible waste pretreatment for halogen concen-
tration reduction and pH adjustment.
• Formation of chlorides (particularly with heavy
metals such as lead) causing particulate removal
problems at the air-pollution control system.
pH adjustment and the sulfur and chlorine contents
of the feedstock are important unit design and oper-
ating considerations. The high processing tempera-
tures combined with acid materials can result in the
need for expensive alloys for the belt and rollers, and
can reduce the operating life of the components in
contact with the acids. High organic chlorine loads
may exceed the air pollution control equipment de-
sign limitations. A pH of 5 to 9, and chlorine and sul-
fur contents not exceeding 5%, are recommended.
Fluorine and phosphorous concentrations of even
several hundred ppm in the feedstock will result in
eventual deterioration of the silicate-based refrac-
tory ceramics, although this is a common problem in-
herent in all thermal technologies.
Based on the above, the current Shirco unit design
places maximum limits on the halogen and sulfur
contents. As a result of problems encountered at
Peak Oil with high lead levels and particulate emis-
sions, the emissions control section of the Shirco unit
that was employed at the Peak Oil site [1] has been
replaced with a high-efficiency Calvert scrubber that
is designed to provide improved particulate-removal
efficiencies over a wider range of gas flow and fines
loading. The Shirco unit that was employed at the
Florida Steel site [6] incorporates a crossflow hori-
zontal scrubber that is similar to the original system
employed at Peak Oil. During initial trial-burn runs
[5], the high chlorine levels in the feed resulted in
high particulate emissions. The current TSCA per-
mit for the OH Materials unit specifically limits the
chlorine concentration in the feed to 0.9 wt%. Blend-
ing of the feed was employed to meet this specific
standard.
Alkali Metals—
Sodium and other alkali metals (such as potassium)
in the waste can create the following problems in the
Shirco unit:
• Deterioration of the silicate-based refractory.
• Formation of a sticky, low-melting, fine-
particulate (particularly sodium), causing possible
fouling or slagging problems at the air-pollution
control system.
Preoperations laboratory analysis and testing are re-
quired in order to assess the extent to which alkali
metals may be a problem. As in other incineration
processes using silicate-based refractories, total
alkali-metals concentration of less than 1 wt% must
be maintained, usually through feed blending.
Heavy Metals—
Heavy metals in the waste feed are not destroyed by
combustion. Although the majority of the heavy met-
als concentrates in the furnace ash, some metals,
(particularly lead) will volatilize. Depending on the
initial concentration in the feed, these metals may
cause reductions in the particulate removal efficien-
cies of conventional venturi-scrubber systems, such
as the design employed at Peak Oil [1]. As discussed
above, a high-efficiency venturi scrubber that can op-
erate over a wide range of gas flow and particulate
loading is mandatory for a transportable Shirco unit.
Preoperations testing and analysis is also required
prior to commitment at a site to define the waste ma-
trix and the effect, if any, that heavy metals concen-
trations may have on the ability of the unit to meet
applicable particulate and emissions standards, as
well as scrubber water-effluent quality-require-
ments. In general, heavy metals concentrations less
than 1 wt% can be processed in the Shirco unit.
With the majority of the heavy metals concentrated
in the furnace ash, the storage and disposal of the
furnace ash requires adherence to the RCRA hazard-
ous waste standards as defined by the EP Tox and
proposed TCLP toxicity characteristic standards.
Based on the results obtained in the Peak Oil and
Demode Road SITE demonstrations [1,2], there is no
evidence of reduced mobility of heavy metals as a re-
sult of the Shirco thermal treatment as compared to
the original waste-feed.
23
-------
Ranges of Site Characteristics Suitable
for the Technology [19,22]
Site Selection
The selection of the processing and thermal treat-
ment site is based on the following criteria:
• The site ideally needs to contain sufficient land
area to provide a concentric ring of unoccupied
space as a buffer zone between active storage,
treatment, and disposal areas, and the nearest
area of human activity. Vegetation, topography,
distance, and artificial barriers all are potential
means to screen facility activities from line-of-
sight exposure to commercial and residential
areas.
• Access roads must be available and capable of sup-
porting the 60,000-lb transportable incinerator
trailers and heavy earth-moving equipment, such
as front-end loaders.
• Accessibility to the waste feed material must be
direct and unencumbered, with adequate waste
excavation and feed preparation areas.
Climate Characteristics
The primary climatic features that can adversely af-
fect a remediation site are the amount of annual or
seasonal precipitation and the incidence of severe
storms. Copious precipitation will cause surface run-
off and water infiltration through the soil. Runoff,
that amount of rainfall that does not infiltrate the
soil, depends on such factors as the intensity and du-
ration of the precipitation, the soil moisture content,
vegetation cover, permeability of the soil, and the
slope of the site. Normally, the runoff from a 10-yr
storm (recurrence interval of only once in 10 yr) or
annual spring thaw, whichever is greater, is contain-
able by the site's natural topography. If not, berms,
dikes, and other runoff control measures must be
constructed to modify the site.
ECOVA claims that the Shirco system has no climat-
ic limitations, except those affecting the feedstock,
such as frozen ground or interference with material
excavation. This limitation can be minimized by
scheduling excavation of the material during a tem-
perate period and then covering the waste prior to
operation. The Shirco system is provided with a win-
terizing package to permit it to be operated in cold
climates.
Geological Characteristics
The main geological constraints that can render a
site unsuitable for a hazardous-waste thermal-
treatment facility are historical or predicted seismic
activity, landslide potential, volcanic or hot spring
activities, and the general load-bearing consider-
ations associated with the siting of heavy equipment
on potentially fragile geological formations.
Topographic Characteristics
The main topographic constraints are susceptibility
to flooding, erosion, and offsite drainage runoff. The
site will need sufficient area for the construction of a
runoff holding pond (or diversion to an existing hold-
ing pond) to retain surface runoff, which may contain
hazardous substances in solution. Because of the
holding pond and flood protection criteria, construc-
tion in flood plains normally is not acceptable.
The site must be graded and leveled for equipment
placement. Attention should be given to the overall
site slope, which should be compatible with the
area's natural topographical slope for drainage.
Site Area Requirements
The maximum size of a trailer-mounted, truck-
transportable incineration-system is governed main-
ly by over-the-road size and weight limitations,
which vary from state to state. In general, size re-
strictions are a length of 45 ft, width of 12 ft, height
of 12 ft, and weight of 64,000 Ib.
The main components of the Shirco 100 ton/d system
are housed in five over-the-road trailers. The prima-
ry chamber is permanently mounted on 2 trailers;
the secondary chamber and offgas handling system is
permanently mounted on 2 additional trailers; and
the fifth trailer contains the control room, laborato-
ry, and administrative offices. In addition to the
above are: a spare parts trailer, a decontamination
trailer, a feed-materials staging area, an ash han-
dling area, and water treatment facilities. A parking
area and access roads are also required. The entire
site is further defined by health and safety consider-
ations and composed of three separate zones, as fol-
lows:
# Exclusion or hot zone: This includes the actual
area of contamination and has the greatest poten-
tial for employee exposure/ The exclusion zone in-
cludes the entire incineration operation, including
the feed-preparation and feed-storage areas, the
ash conveyor and storage areas, and the emission
control system. Contaminated materials are
stored and handled in the exclusion zone.
« Contamination reduction zone: This zone sur-
rounds the exclusion zone and acts as a clean
"buffer" zone. This zone includes contamination-
reduction corridors where personnel and equip-
24
-------
ment are decontaminated prior to entering the
support zone area.
Support or clean zone: This is the zone surround-
ing and outside of the other two zones. The support
zone is a non-contaminated area where support
operations are conducted. Support operations may
include office and control-center operations.
Based on the above, the site design for the 5 main
component trailers requires a concrete pad of ap-
proximately 10,000 to 15,000 ft2, dimensioned ac-
cording to the design layout of the unit and the site's
physical constraints.
The feed-preparation-area design is preferably a con-
crete pad whose dimensions are dependent on the se-
lection (as required) of a suitable carrier material,
the selection of fuel additives to increase Btu value,
and other handling issues (dewatering, crushing,
etc.). The pad is provided with curbs around the pe-
rimeter. The joint between the curb and the slab is
sealed so the pad can be washed without allowing
contaminated soil or water to escape. If required by
state or local authorities, or by site-specific environ-
mental conditions, a building can be designed to con-
tain fugitive odors and vapors and to protect the feed
matrix, operating equipment, and instrumentation
from extreme environmental conditions.
The ash storage area is typically on a lOOxlOO-ft pad
surrounded by a 3-ft berm to prevent surface runoff.
The area is usually graded to provide drainage with
a berm containing straw filter breaks. The base of
this area is usually a 2-in. layer of asphalt to act as
the water-proof membrane. The ash is discharged
from the unit onto a conveyor. It is then conveyed to
ash bins, pulled onto a truck, and then hauled to the
storage area and placed into the appropriate covered
isolation-area.
Should the ash not automatically qualify as a de-
listed waste, the staging and sampling process would
be implemented. Estimated turnaround time for the
sample results is 24 h. As sample results become
available, the ash would be backfilled or reincinerat-
ed as appropriate.
The other equipment and ancillary facilities can be
placed on graded and graveled areas. An area of ap-
proximately 30,000 ft2 may be required to accommo-
date the above.
The complete system, exclusive of the waste site, can
be assembled on a total site area of 40,000 to 45,000
ft2, which is equivalent to an overall plot dimension
of 150x300 ft.
Site Utility Requirements
The transportable Shirco unit has been designed to
be a self-contained and stand-alone unit. It is self
supporting, but requires site preparation as dis-
cussed above. Utility requirements include the fol-
lowing:
• A continuous water supply. If city water is not al-
ready available, a well or other external supply
(such as water tank trucks) must be in place in or-
der to furnish water to the scrubber system. A re-
circulation system will be used to minimize the
scrubber water usage. The water for this purpose
is not required to be potable; however, good qual-
ity water is required, low in suspended solids, and
not brackish. The SITE demonstration results
(Appendix C) illustrate and reinforce the need for
good quality water. High calcium and magnesium
sulfates and chlorides appeared to contribute to
the excessive salts content and overloading of the
Peak Oil scrubbing system [1]. Organic contami-
nants in the tank-truck water-supply at Rose
Township [2] were also evident in the stack gas.
Sixty to seventy gpm generally is required; 10 to
30 gpm blowdown typically must be disposed of,
after suitable treatment to accommodate appro-
priate water discharge standards.
• Electrical service of 2,000-kVA, 480-V, and 3-
phases is usually taken from a local utility line to
a substation, and converted to 15-amp, 120 V, 1-
phase service. The 480-V service is used as the
power source for the PCC and other large-electric-
demand users, such as the ID fan and pumps. The
120 V service is used for ancillary systems and site
needs, as required. If electrical power is not avail-
able from a local utility line, portable diesel-
powered generators are required.
• Propane or natural gas service, equivalent to 6.2
M Btu/h.
• Water treatment chemicals, as required.
• Fuel oil for feed Btu improvement, as required.
Material Handling Required by the
Demonstrated Technology
The feed preparation section of the system is the key
to the successful operation of the Shirco unit. The
feed must be properly prepared to meet the design re-
quirements of the unit. The feed weighing and con-
veying system will be affected by the waste's phys-
ical and handling properties. Feed preparation to
25
-------
achieve the proper size and consistency is a direct
function of the matrix's characteristics. Regardless of
whether the system is designed and provided by the
unit's operator or by Shirco, preoperation analyses
and materials-handling investigations must be con-
ducted to ensure the successful application of the
myriad of materials-handling equipment and pro-
cesses to the specific-site waste-feed matrix.
The Shirco unit is designed to process a range of
soils, solids, and semi-solid sludges and slurries.
Waste can be preprocessed by dewatering, soils
blending, and/or lime addition to ensure a solid/semi-
solid matrix. Pure liquids can also be processed if
blended with a suitable carrier to form a semi-solid
matrix.
Site Excavation
Soils and Solids—
Excavation activities would normally be carried out
by bulldozers, front-end loaders, and/or other conven-
tional excavation equipment. The excavated materi-
al would be moved to the processing area either di-
rectly with front-end loaders, or via transfer truck or
conveyor. The types of excavation and transfer
equipment would depend on the type of material and
the layout of the site. This type of excavation was
conducted at Florida Steel [4] and LaSalle Electric
[11].
Semi-Solid Sludges—
Semi-solid sludges must be stabilized to a soil/solid
state by mixing with adjacent oils or other suitable
materials, such as sand and/or lime. Excavation ac-
tivities then can proceed as defined for soils and so-
lids.
Slurries and Lagoons—
The content of slurries and lagoons must be dewater-
ed, treated as a semi-solid sludge, and then excavat-
ed as a soil/solid. The excavation of a waste oil lagoon
was conducted at Peak Oil [1].
Feed Preparation
The feed preparation system, as discussed above, is a
direct function of the waste matrix characteristics
and their relationship to the requirements of the unit
for feed size and consistency.
The feed preparation system at Florida Steel [4]
(where the waste was a diverse mixture of soil/solid
constituents ranging from environmental control
dust to car bumpers and railroad ties) consisted of a
grizzly classifier, a magnetic separator, a jaw crush-
er, a roll crusher, a pugmill, a plastic shredder, a
wood chipper, a weigh belt feed, and associated con-
veyance systems.
All of the feed preparation equipment and the exca-
vated waste was covered and protected from the
weather.
The feed preparation system at Peak Oil [1] pro-
cessed the oily sludge from a dewatered lagoon that
was stabilized with lime and sand and then bulldozed
to a staging area where a power screen shredded,
screened, and aerated the waste to a consistency and
size required by the Shirco unit. The screened waste
was staged, fed to a weigh belt feeder by a front-end
loader, and conveyed to the Shirco unit's feed mod-
ule. All of the feed preparation equipment and-the
excavated waste was unprotected from the weather.
The conveyors were covered, and the PCC and SCC
systems of the Shirco unit were enclosed under a
temporary tent arrangement.
The feed preparation system at LaSalle Electric [11]
used the same power screen and equipment arrange-
ment as discussed above for Peak Oil. (The Shirco
unit at LaSalle Electric is the same unit that was
employed at Peak Oil.)
Ash Handling
When the furnace ash reaches the end of the convey-
or belt through the PCC section of the Shirco unit, it
is quenched with water sprays and is removed from
the unit through a series of screw conveyors. The ash
then is transported to the ash storage area where it is
sampled for analysis before it is placed in bulk stor-
age or reprocessed (based on the analytical results
and applicable regulatory standards). As discussed
in the following subsection describing commercial
operations at Peak Oil, the moist furnace ash, which
tends to clump and agglomerate, can cause materials
handling problems. If insufficient quenching is em-
ployed, dusting and odor problems can occur. Propri-
etary modifications were made to the OH Materials
Corp. commercial unit [4,5] that addressed these
problems. Careful control of the ash quench water
and exit temperature of the furnace ash are required
to minimize clumping and agglomeration and, at the
same time, keep dusting and odor problems under
control.
Commercial Operations
Peak Oil—
The Peak Oil waste-feed matrix (commercial-scale
unit) was a solidified sludge that was prone to
agglomeration and resulted in clogging, bridging,
and jamming of the original crusher equipment.
26
-------
Prior to the SITE demonstration test, the crusher
was replaced with a power screen that shredded,
screened, and aerated the feed to a consistency and
size that was accommodated by the Shirco feeder.
This modification improved the feed system's
reliability.
Conveyor system problems included spillage of waste
feed, waste material sticking to the conveyor belt,
and an inability to adjust the feedrate from the
conveyor to the unit's feeder system. Modifications to
the conveyor system included the addition of a
"skirt" below the conveyor to catch spillage, a
conveyor scraper that minimized sticking, and a
variable speed controller and revised motor
arrangement that provided feedrate control.
Although the overall conveyor system provided
waste feed to the Shirco unit, preoperation analyses
and materials handling investigations would have
resulted in a system design that was more adaptable
to the waste-feed matrix encountered at the Peak Oil
site.
The screw augers and their motor drives experienced
continuous clogging and overload problems. The feed
system required continuous attention by operating
personnel and the addition of "bridgebreakers" to
reduce the bridging of the agglomerating waste feed.
As is the case with the feed preparation section, the
design configuration of the feed inlet section and the
screw augers should have been specific to the waste
feed matrix. The flight pitch, height, and gear
reduction of the feed auger should have been
designed based on preoperation investigations and
waste-feed matrix analysis.
The screw augers were designed with reversing capa-
bility, and the motor drives were designed for a 50%
overload-based on adequate feed preparation. If the
feed is not properly crushed, screened, and prepared,
the augers' materials-handling efficiency decreases.
Bridging and plugging problems (particularly with
an agglomerating feed matrix) occur, causing signifi-
cant overload and eventual burnout of the motor
drives. Again there is a need for preoperation testing
and evaluation of the waste feed matrix vis-a-vis the
entire feed handling system.
The ash removal system required frequent mainten-
ance and downtime. The cooling screw and incline
screw were continually clogging and breaking, and
their motors overloaded and burned out. When the
screws were reversed to dislodge material under the
screw flights, breakage and further abuse of the mo-
tors would occur. Significant dusting and odor prob-
lems also were evident in and around the ash remov-
al system.
In addition to the design limitations discussed above,
the intermittent failure of the original feed prepara-
tion system (i.e., crusher and screen) to deliver a con-
sistently sized waste feed allowed unprepared mate-
rials to enter the unit. The unprepared feed caused
occasional jamming and blockage of the ash dis-
charge system. Plugging of the incline screw also
was caused by the buildup of ash in the discharge
chute and improper control and monitoring of the
ash quench facilities.
In early 1987 the cooling screw and incline screw de-
sign were changed; larger motors and gear reducers
were installed to further correct overload, plugging,
and motor burnout. A viable solution to future de-
signs could entail the installation of a larger-
diameter screw operating at lower rpm than the
small, high-rpm screw conveyor, which proved to be
a high-maintenance item subject to substantial wear
over a short period of time.
Another alternative, a wet system design, does not
appear to be viable; it entails substantial equipment
maintenance and environmental concerns when
dealing with an abrasive ash solution.
The dusting problems that were continually present
at the ash removal system can be minimized by care-
ful control and monitoring of the ash, quench water
flow, especially during start-up or periods of inter-
rupted ash discharge. Potential odor problems are in-
herent to the quench operation and will vary in se-
verity with the waste material. In any event, unit
and site setup should take into account these poten-
tial health problems; ash removal and storage should
be located for minimal exposure to operating person-
nel and traffic.
In addition to the feed-inlet and ash-outlet systems,
problems also occurred with conveyor belt failures,
cakebreaker failures, and belt conveyor system
maintenance.
A transportable unit moving from site to site will be
subject to metallurgical degradation if one assumes
that a single alloy will be adequate for all applica-
tions. Knowledge of the physical and chemical char-
acteristics of the feed is essential in selecting appro-
priate alloys. The original belt installed at the Peak
Oil site was provided with several test sections of
various alloys. Because of the nature of the feed ma-
terial and minimal knowledge of its chemical charac-
teristics, this approach was selected so that, if belt
failure did occur, an appropriate alloy then could be
installed. Due to the chlorine and sulfur content of
the initial feed material, certain test sections did fail
and were replaced with the standard Type-314 stain-
less steel alloy. A properly cured Type-314 stainless
steel belt provided reliable service through the com-
pletion of the project. Belt specifications and subse-
quent construction materials may require occasional
27
-------
changes due to the unique characteristics of a par-
ticular feed material.
As with the belt, metallurgical considerations for the
cakebreakers are dictated by the physical and chemi-
cal properties of the feed material and subsequent
furnace environment. Corrosion problems can be re-
solved through the selection of an appropriate alloy
for the feed material characteristics. At Peak Oil, the
original alloy was not compatible with the waste
feed. In addition, possibly due to the mechanical fail-
ures in feed screening and crushing noted earlier and
to the resultant feeding of unsized or nonspecifica-
tion waste material, the cakebreakers also may have
been subject to severe stress when these articles were
encountered, causing cakebreaker failure.
Although problems were encountered with the belt
conveyance system, it appears that the roller bearing
specifications do not require any changes. Proper at-
tention to lubricant choice and a rigorous mainten-
ance schedule are required to ensure a long roller-
bearing and belt-conveyance-system operating- life.
Florida Steel-
The Florida Steel waste matrix, whose characteris-
tics were suitable for processing by the Shirco sys-
tem, was stored in an onsite vault that was protected
from the weather. This waste matrix, along with the
extensive feed-preparation equipment onsite, result-
ed in a waste feed to the Shirco unit that met all of
the physical characteristics required for efficient
materials-handling and unit operation.
OH Materials indicated that several problems were
encountered, but it only elaborated on the conveyor
belt problem [4,6]. The conveyor belt that was origi-
nally installed had a larger pore space than the pilot-
scale unit, thus allowing the fine Florida sand to
sieve through and overload .the fines collection sys-
tem. A new, smaller-gauge conveyor belt was in-
stalled that resulted in the satisfactory operation of
the unit.
During the final month of operation, the overall op-
erating (or utilization) factor was greater than 90%,
which indicates that OH Materials had solved all of
its initial problems.
LaSalle Electric-
Discussions with the Illinois EPA indicate that the
Shirco unit is operating at an 80%-90% operator fac-
tor with few materials-handling problems. The pow-
er screen is apparently working well; on occasion,
however, large nails or spikes may pass through the
screen and cause equipment problems.
These preliminary results are an encouraging
improvement over the difficult operations and
materials-handling problems that were encountered
by this same unit at Peak Oil (exclusive of the newly
designed air-pollution-control system and possibly
other proprietary design changes).
The contrast between the Peak Oil and LaSalle
Electric waste matrices (oily sludge versus dry soil)
emphasizes the importance of the initial physical
characteristics of the site waste and the efficiency of
the feed preparation equipment in producing a
satisfactory waste feed to the Shirco unit.
Personnel Issues
Operating personnel for the Shirco unit total 13. This
includes 9 process operators; 3 supervisors including
a shift foreman, a maintenance supervisor, and an
office administrator/clerk; and a project manager.
The operations schedule consists of two 12-hour
shifts, requiring 3 operators per shift to cover the
control room, PCC, and SCC/scrubber sections.
Operating personnel are scheduled for an 8-d work
period followed by a 4-d rest period. Additional local
hires — such as laborers, operating and craft
personnel, and materials-handling personnel for soil
excavation, feed handling, and ash removal — are
site-specific and are not included in the labor profile
discussed above.
Personnel are subjected to the standard OSHA
requirements for operating moving equipment and
are required to wear the proper personal protective
equipment dictated by the specific site conditions
and contaminants.
Personnel must pass appropriate physical
examinations and have completed, and be certified,
in EPA-approved hazardous-materials training
procedures and protocols.
Tests to Evaluate Applicability and
Performance of Technology
As discussed in the preceding subsections, waste
characterization and treatability testing are neces-
sary to establish the suitability of the waste-feed and
the range of recommended operating parameters for
the commercial unit to ensure optimum performance
within regulatory requirements. The following dis-
cussion addresses the 3 test phases required for the
use of a Shirco commercial unit at a specific site.
These 3 test phases include:
• Laboratory analysis of waste feed.
• Treatability testing including bench- and/or pilot-
scale testing and thermogravimetric analyses
(TGA).
28
-------
• Technical evaluation of commercial operation and
monitoring for regulatory compliance.
Laboratory Analysis of Waste Feed
Prior to treatability testing, complete laboratory
analysis for the following key physical and chemical
properties of the waste feed matrix are recommend-
ed:
• Density — to determine .feedrate and handling re-
quirements.
• Moisture content — to determine feedrate, fuel
consumption, and handling requirements.
• High heating value - to determine feedrate and
fuel consumption.
• Non-combustible ash content — to determine vol-
ume of furnace ash requiring posttreatment han-
dling and disposal.
• Particle size analysis — to determine materials
handling requirements, feed preparation require-
ments, and particulate control.
• Flash point and viscosity" for sludges — to deter-
mine materials handling requirements.
• Elemental analysis/composition — to determine
C, H, O, N, S, P concentrations for combustion cal-
culations and unit feedrate considerations, and air
pollution control requirements.
• Elemental analysis/composition — to determine
halogen concentrations (Cl, F, Br, I) resulting in
acid gases during combustion, which require
stack- gas scrubbing facilities.
• pH — to determine handling equipment mainten-
ance and metallurgical requirements, and need
for waste preparation neutralization.
• Metal species and concentrations — to determine
alkali metals concentrations (Na, K) for equip-
ment maintenance; heavy metals concentrations
for air pollution control, scrubber water treatment
and disposal needs; and ash disposal and delisting
requirements.
• Organic species and concentrations — to deter-
mine materials handling and pollution control re-
quirements, personnel protection needs, and ash
and scrubber-water treatment and disposal re-
quirements.
• POHC species and concentrations — to determine
spiking and analytical requirements for a trial
burn or demonstration test that requires a DRE
determination.
Treatability Testing
Treatability testing that includes bench- or pilot-
scale testing and TGA analyses, will establish the
range of recommended operating parameters for the
commercial unit to assure optimum operation within
regulatory requirements including:
• Corrosion prevention data
• Feed preparation (pH neutralization, chemicals
stabilization) data
• Baseline processing conditions (residence time,
temperature, waste layer thickness)
• Furnace atmospheric requirements
• Primary energy consumption estimate
• Ash storage and disposal requirements
The pilot-scale tests and analyses will approximate
the commercial operation, and test procedures will
be similar to the commercial operation, as discussed
below.
Technical Evaluation of Commercial
Operating and Monitoring for Regulatory
Compliance
In order to verify the performance of the commercial
unit and its compliance with governmental regula-
tions on incineration, the following technical and
performance criteria and tests must be addressed.
• Physical and chemical characteristics of the feed
— including ultimate analysis, high heating val-
ue, density, moisture, ash content, and organics
and metals concentrations. This will determine
the applicability of the technology and the need
for specific waste handling and unit modifications.
Concentrations of specific physical characteristics
such as moisture content and chemical character-
istics such as organics (PCBs) and heavy metals
concentrations may be monitored during the
cleanup on a defined schedule since they can affect
the performance and operation of the unit.
• DRE levels for designated POHCs, PCBs, and the
presence of PICs in the stack gas. The regulatory
standards for POHCs are 99.99% DRE under
RCRA and for PCBs is 99.9999% DRE under
TSCA. Compliance to these standards are usually
established in a 3-d trial burn prior to or at the be-
ginning of a cleanup and are indicative of the abil-
ity of the unit to effectively destroy the hazardous
contaminants contained in the waste-feed.
29
-------
* Level of hydrogen chloride (HC1) and particulates
(including heavy metals concentrations) in the
stack gas. The RCRA standard for HC1 in the
stack gas is the larger of 1.8 kg/h (4 Ib/h) or 99
wt% HC1 removal efficiency. The RCRA standard
for particulate emissions in the stack gas is 180
mg/dscm (0.08 gr/dscf). Compliance to these stan-
dards are usually established in 3-d trial burn pri-
or to or at the beginning of a cleanup and are indi-
cative of the ability of the unit to meet air- emis-
sions control criteria.
• 02, CO, CO2, NOX, and THC concentrations in the
stack gas emissions. Concentrations are usually
monitored continuously to meet specified limits
for each species and the calculated combustion ef-
ficiency (CE) since they are indications of the per-
formance and operation of the unit.
• Level of residual heavy metals and organics, in-
cluding PCDDs, PCDFs, and PCBs, in the furnace
ash. The TSCA guidance level is 2 ppm of residual
PCBs in the furnace ash. Additional physical and
chemical characteristics of the furnace ash include
ultimate analysis, moisture, and ash content.
Standards for ash quality are usually pre-
established for the particular cleanup and must be
monitored during the trial burn and the actual
cleanup.
• Mobility of heavy metals in the furnace ash as
measured by the Extraction Procedure Toxicity
(EP Tox) and the proposed Toxicity Characteristic
Leaching Procedure (TCLP) tests. In some cases
other performance levels may be set. These stan-
dards or performance levels, if required, will be
monitored during the cleanup on a defined sched-
ule and determine furnace ash disposal require-
ments.
• Level of residual heavy metals and organic
compounds, including PCBs, PCDDs, and PCDFs,
and other physical and chemical characteristics in
the scrubber water discharged from the unit,
including pH, TOC, TDS, and TSS. Standards are
usually pre-established based on POTW
requirements and must be monitored during the
cleanup on a defined schedule.
30
-------
SECTION 4
ECONOMIC ANALYSIS
Introduction
The costs associated with the transportable Shirco
Infrared Thermal Destruction System are defined by
12 cost categories that reflect typical cleanup activi-
ties encountered on Superfund sites. Each of these
cleanup activities is defined and discussed, forming
the basis for the estimated cost analysis presented in
Table 3 for a Shirco transportable unit operation.
The costs are based on the processing of 36,500 tons
of waste-feed in a commercial unit. This quantity is
based on the amount of waste that would be pro-
cessed if the commercial unit operated at the design
capacity of 100 ton/d and a 100% operating (or utili-
zation factor) over a 365-d annual period. However,
the costs presented in Table 3 have been adjusted to
reflect real-time operations of the unit since periodic
shutdowns are required in order to respond to main-
tenance or operational problems. Costs are given
based on operating factors of 85%, 80%, 70%, 60%,
and 50%.
Results of Economic Analysis
The economic analysis is based on cost data available
from several sources [20,21]. Due to the uncertain-
ties in estimating the actual operating days per year
in which the unit will process waste at its stated ca-
pacity of 100 ton/d, a series of economic models is pre-
sented in Table 3 for operating factors ranging from
85% and 429 days onsite, to 50% and 730 days onsite.
Total costs per ton range from $182.13 to $240.79.
These costs are considered order-of-magnitude esti-
mates and have an expected accuracy within + 50%
and -30% as defined by the American Association of
Cost Engineers; however, because this is a new tech-
nology, the range of uncertainty is probably signifi-
cantly wider.
Approximately 50% of the costs associated with the
Shirco system can be reduced on a cost per ton basis
if, for a particular unit operation:
• feedrate can be increased by upgrading unit oper-
ating performance or improving initial design
• the operating (or utilization factor) can be in-
creased during the operation of the unit.
In both of these cases, the period of time that the unit
will remain at the waste site will be reduced, thus ef-
fecting cost savings.
Costs can also be reduced by effecting a basic change
in the efficiencies with which activities are executed,
thus lowering their respective costs. Productivity can
be improved in technical and administrative assis-
tance for permitting and regulatory activities, oper-
ating labor, and maintenance.
Other costs — including supplies and consumables;
utilities; and effluent/residual treatment, handling,
and disposal vary with waste feedrate and will not be
affected on a per-ton waste-feed basis. These costs,
which account for 20% to 26% of the total costs, can
be reduced by optimizing operating conditions.
The Phase II runs conducted at Rose Township [2] in-
dicate that reductions ranging from 26% to 67% in
power usage, and by as much as 51% in fuel gas con-
sumption, can be effected by adding oil to the waste
feed and reducing operating temperatures in both
the PCC and SCC. The oil addition increases the
overall waste-feed-matrix heating-value and permits
the PCC to operate at a higher feedrate and shorter
residence time, and a lower electrical-heating load.
Operations at Peak Oil [1], where the waste matrix
was primarily a waste oil sludge with a high heating
value, were autogenous in the PCC and did not re-
quire any electrical heating input.
In order to verify the feasibility, operating
parameters, and economics (power and fuel usage,
including fuel oil addition) of processing a particular
waste feed matrix, the following tests must be
conducted:
31
-------
Table 3. Estimated Costs in $/ton:Site Cleanup
Site preparation costs (estimate)
Permitting and regulatory costs
Administrative/permitting
(10% equipment costs)
Trial bums (estimate)
Site-specific permitting/Engineering (estimate)
Operations procedures/Training (estimate)
Equipment costs ($3.2M)
Startup and fixed costs
Transportation/setup (estimate)
Onsite checkout (estimate)
Initial startup/Shakedown (estimate)
Working capital (3 mo.)
Depreciation (10% of equipment costs)
Insurance and taxes (I0%of equipment costs)
Labor costs
Supplies and consumables costs
Chemicals (S2.00/ton of waste)
Oil addition (Si.OO/gal)
Utilities costs
Fuel (15.00/M Btu)
Power (S0.10/kWh)
Water (SO.80/1, 000 gal)
Effluent treatment and disposal costs and
residuals and waste shipping, handling, and
transport costs
Water ($1.00/1,000 gal) (excludes ash)
Analytical costs (SSOO/operating day)
Facility modification, repair, and replacement
costs
Maintenance (10% of equipment costs)
Contingency (10% of equipment costs)
Site demobilization costs
Decontamination/demobilization (estimate)
Totals, S/ton
Cost Element
85%
13.70
9.53
4.11
2.74
2.74
5.48
1.40
4.38
3.01
10.31
10.31
37.39
2.00
8.00
12.00
24.00
0.58
0.72
5.00
10.31
10.31
4.11
182.13
Breakdown
Unit Capacity @ 100 ton/d
Onstream Factor
80%
13.70
9.86
4.11
2.74
2.74
5.48
1.40
4.38
3.01
10.96
10.96
39.73
2.00
8.00
12.00
24.00
0.58
0.72
5.00
10.96
10.96
4.11
187.40
70%
13.70
10.64
4.11
2.74
2.74
5.48
1.40
4.38
3.01
12.52
12.52
45.40
2.00
8.00
12.00
24.00
0.58
0.72
5.00
12.52
12.52
4.11
200.09
60%
13.70
11.68
4.11
2.74
2.74
5.48
1.40
4.38
3.01
14.61
14.61
52.97
2.00
8.00
12.00
24.00
0.58
0.72
5.00
14.61
14.61
4.11
217.06
50%
13.70
13.14
4.11
2.74
2.74
5.48
1.40
4.38
3.01
17.53
17.53
63.56
2.00
8.00
12.00
24.00
0.58
0.72
5.00
17.53
17.53
4.11
240.79
These costs do not include waste excavation, feed preparation, vendor profit and ash residual disposal.
• Bench-scale tests to determine the feasibility of
any proposed pretreatment process.
* A series of waste feed analyses to determine opti-
mum operating conditions.
• A mass- and energy-balance program, allowing
the optimization of the technology by matching
the operating characteristics of the unit with the
characteristics of the waste to be incinerated.
• A series of pilot test burns by the pilot-scale
thermal-destruction unit to assure that proper op-
erating parameters can be maintained while
meeting regulatory requirements.
Additional cost information is provided in
Appendices B, C, and D and summarized in Table 4.
For comparison, the results of the economic analysis
presented in Table 3 are also provided.
32
-------
Table 4. Summary of Estimated Costs
Data Source
Brio Site
Friendswood, TX
(Shirco cost est.) [1 3]
Unit
Capacity,
tpd
150
220
Operating
Factor, %
82
82
Unit
Cost,
$/ton
I43(a)
119(3)
Lasalle Electric
Lasalle, IL
(Haztech proposal) [11] 100
Florida Steel
Indiantown, FL
(OH Materials est.) [4,6] 100
Peak Oil
Brandon, FL
(SITE Tech. Eval.
Report) [1] 100
ECOVA
Dallas, TX
(Vendor's claims) [19] 100
Economic Analyses 100
(Table 3)
60
61
80
37
85
85
80
70
60
50
300(a)
< 300(0)
197(c)
416(C)
161-257(a)
182(c)
187(C)
200(C)
217(c)
241 (c)
(a) Cost includes vendor profit, and excludes waste excavation, feed
preparation and ash disposal.
(b) Cost includes vendor profit, waste excavation and feed
preparation, and excludes ash disposal.
(c) Cost excludes vendor profit, waste excavation, and feed
preparation and ash disposal.
Basis of Economic Analysis [19-21]
A detailed discussion of each of the cost elements de-
fined in Table 3 is provided in the following:
Site Preparation Costs
The costs associated with site preparation and logis-
tics include advanced planning and management, de-
tailed site design and development, auxiliary and
temporary equipment and facilities, water condition-
ing, emergency and safety equipment, and site staff
support. Soil excavation, feedstock preparation, and
feed handling costs would normally be included but
are not being considered in this analysis due to their
site-specific variability. Total site preparation costs
are estimated at $0.5M.
Permitting and Regulatory Costs
Administrative/Permitting—
Administrative costs associated with regulatory
compliance issues for an incinerator are numerous
and varied. The costs that are being accrued under
this cost element reflect overall non-site-related reg-
ulatory activities. These activities include research-
ing national or regional permit requirements, pre-
paring initial permit applications, and supporting
the permit issuance process. Once the final permits
are issued, then recordkeeping, inspection, survey
response to permitting agencies, and additional re-
porting activities may be required.
Reporting activities include the preparation of
technical support data: the trial burn results,
sampling and analysis plan, and quality assurance
project plan by in-house engineering personnel; and
RCRA/TSCA permit forms by a senior engineering
consultant working with in-house staff. Admin-
istrative costs associated with reporting activity
cover time, travel, and per diem for .consultant and
in-house staff interfacing with federal EPA officials;
and in-house administrative and clerical staff
functions. The preparation of the final trial burn
report by in-house engineering personnel is also
included.
With the size and complexity of the unit influencing
these activities, the total administrative/permitting
costs are estimated at 10% of the equipment costs or
$0.32M. Fifty percent of these costs can be considered
time-related and will be affected by the length of
time at the site; the remaining costs are one-time
costs at a fixed $/ton basis.
Trial Burns—
Under current TSCA regulations, hazardous-waste
incineration-facility owner/operators usually are
required to perform a trial burn as the final step in
obtaining an operating permit.
In addition to the administrative and permitting
costs defined above, costs are accrued for the
execution of the TSCA trial burn to prove overall
system performance.
The costs for such a trial burn include labor and
materials for the sampling and analysis activities,
travel and per diem for the sampling team, and other
miscellaneous costs that may be attributable to the
execution of the trial burn, exclusive of admin-
istrative support.
It should be noted that these nondepreciable capital
costs only are accrued for TSCA trial burn activities;
site-specific permit and trial burn activities are
considered semivariable operating costs that accrue
under the mobilization/ demobilization cost element
breakdown discussed below.
Total costs for these trial burn activities are estima-
ted at $0.15M.
33
-------
Site-Specific Permitting and Engineering Services—
In addition to the TSCA trial burn activities dis-
cussed above, site-specific permitting and trial burn
activities may be required. Both in-house and consul-
tant technical support and engineering services may
be required to support these efforts. Total costs for
these site-specific permitting and engineering ser-
vices are estimated at $0.10M.
Operations Procedures and Training—
In order to ensure the safe, economical, and efficient
operation of theunit, operating procedures and a pro-
gram to train operators are necessary. These associ-
ated costs will accrue: the preparation of a unit
health-and- safety and operating manual; and the
development and implementation of an operator
training program, equipment decontamination pro-
cedures, and automated management and reporting
procedures. Total operations procedures and training
costs are estimated at $0.10M.
Equipment Costs
The current costs — for the design, engineering, ma-
terials and equipment procurement, fabrication, and
installation of the Shirco transportable infrared in-
cinerator on skids — are included as direct costs at
$3.20M. These costs include all the subsystems and
components installed on their respective skids and
trailers, but do not include the costs of the tractors
for the transport of the trailers. Waste preparation
equipment, ash conveyors, and auxiliary equipment
(such as an air compressor or water treatment facili-
ties) are not included.
Startup and Fixed Costs
Transportation and Setup—
The cost of transportation and setup includes disas-
sembly of the unit at its present location and trans-
port to a new location. Present Shirco designs are to-
tally skid-mounted and equipped with hydraulic lev-
elers. The trailers can be moved into place without
removing equipment, thus significantly minimizing
setup time and costs. Estimated costs are $0.20M.
Onsite Checkout—
Once the unit has been set up, it is necessary to
shake down the system to ensure that no damage oc-
curred as a result of disassembly, transport, and
reassembly. Estimated costs are $0.05M.
Initial Start-up/Shakedown—
After the incineration system has been fabricated,
and operations procedures and operator training
have been completed, the overall unit must be ini-
tially started and operated to check the mechanical
and technical integrity of the equipment and its con-
trols. The unit first must be operated without the use
of the infrared rods or the secondary combustion
chamber burners in order to check the movement of
solids through the unit in a "cold" mode. The unit
then must be operated on a nonhazardous feed ma-
trix under a "hot" mode, with the infrared rods and
the secondary combustion chamber burners in opera-
tion. Overall startup costs are estimated at $0.16M.
Working Capital—
Although the unit is a transportable system, it will
require a supply of maintenance materials attribut-
able to a nondepreciable capital cost. Maintenance
materials account for approximately one-half of the
total maintenance cost, and three-month inventories
are usually maintained.
Fuel inventory for the SCC heat source and caustic
soda solution inventory for the scrubber's acid-gas-
removal operation are also required.
Total costs for maintenance materials, fuel oil inven-
tory, and chemicals inventory are $0.11M.
Depreciation—
Because incineration is a capital-intensive waste-
treatment option, the overall costs must include an
annualized capital investment cost or depreciation.
Equipment amortization is based on a straight-line
10-yr depreciation (10% of equipment costs) at
$0.32M/yr.
Insurance and Taxes—
Depending on site location and the specific tax
strategy employed for the ownership and operation of
the unit, insurance and taxes will vary from 5% to
10% of the equipment costs on a yearly basis. For this
analysis, insurance and taxes are estimated to
represent 10% of the equipment costs of the unit at
$0.32M per year.
Labor Costs
Operating personnel for the Shirco unit totals 13 per-
sons. This includes 9 process operators and 3 supervi-
sors who cover two 12-h shifts (8-d work period, 4-d
rest period, 840 h overtime) at $25,000/yr and
$35,000/yr, respectively. It also includes a project
manager at $45,000/yr. Benefits for the above per-
sonnel are estimated at 40%xof straight-time wages,
and overtime is reimbursed at 150% of the standard
wage rate. Per diem is estimated at $100/d per per-
son, and includes lodging, meals, autos, and sched-
uled trips home.
Additional local hires as laborers, operating and
craft personnel, and materials handling personnel
for soil excavation, feed handling, and ash removal
34
-------
are site-specific and are not included in the labor
costs.
Based on the above, the total labor costs for the
operating personnel of the Shirco unit are $1.16M
per year.
Supplies and Consumable Costs
Chemicals—
The main chemical requirement is caustic soda
solution for acid gas scrubbing. The use of caustic
soda is a function of the HC1 loadings based on initial
chloride concentrations in the waste feed. Based on a
50% caustic solution, caustic requirements are 2.2
Ib/lb of HC1, or $2.00/ton of waste feed, based on 0.15
wt% chloride concentration and a chemicals cost of
$0.30/lb of caustic.
Oil Addition—
The heating value of the waste feed matrix
introduced into the unit will have a direct effect on
the unit feed capacity and electrical requirements of
the infrared heating rods. The introduction of diesel
fuel or an equivalent oil supplement will increase the
overall heating value of the waste feed matrix and
provide a means to optimize unit operations. For the
analysis, it is estimated that oil is added to the waste
feed at a rate of 3 wt% or approximately 8 gal/ton of
waste feed. Based on a cost of $1.00/gal, total costs for
oil addition are estimated at $8.00/ton of waste feed.
Utilities Costs
Variable operating-cost elements for this unit
include fuel, power, and water. They are defined as
variable operating-cost elements because they can
usually be expressed in terms of dollars per unit flow
of waste, and as such, these costs are more or less
proportional to overall facility utilization during
specific site operations.
Fuel—
The fuel requirements for the unit include natural
gas or propane fuel for the secondary combustion
chamber heating requirements. Based on SCC
heating requirements of 10 MBtu/h and fuel gas
costs of $5.00/MBtu, fuel gas costs are estimated at
$12.00/ton of waste feed.
Power—
The power requirements for the unit include the
electrical requirements for the motors that power the
pumps, fans, augers, mixers, and primary combus-
tion chamber belt drive. Also included is the
electrical requirement for the PCC infrared rods,
which supply .the initial combustion heat to the
waste feed. One of the factors affecting the electrical
requirement of these infrared rods is the heating
value of the waste matrix being incinerated. As
defined above, oil addition costs are included in the
analysis to reflect a possible increase in the heating
value of the waste matrix.
Auxiliary electrical requirements for trailer power,
site lighting, etc., are minimal and are assumed to be
included in the total power needs.
Based on the above, total power requirements are es-
timated at 1,000 kWh/h. A power cost of $0.10/kWh
is employed to reflect potentially difficult and expen-
sive extensions to power sources. Estimated costs are
$24.00/ton of waste feed.
Water-
Water use is based on an estimate of the blowdown
requirements from the scrubber system, water losses
due to evaporation, and carry-over with the stack gas
and ash residue. All other water needs are satisfied
through the internal recirculation of water from the
scrubber system. Estimated water costs are based on
water makeup requirements of 50 gpm at a cost of
$0.80/1,000 gal or $0.58/ton of waste feed.
Effluent Treatment and Disposal
Costs, and Residual and Waste Shipping,
Handling, and Transport Costs
Effluent Treatment and Residue/Water Disposal--
Costs will accrue for the disposal of ash in a suitable
landfill. Unit disposal costs for landfilling depend on
location and on whether toxic metals are present. If
toxic metals are present, secure landfilling is re-
quired, and disposal costs can exceed $100/ton of
waste feed. Ash disposal costs are not included in this
analysis.
Scrubber water blowdown after onsite settling and
pH adjustment will be routed to a municipal or re-
gional treatment facility if the wastewater meets the
treatment facility's specifications. Based on an over-
all on site and POTW treatment charge of
$1.00/1,000 gal, water disposal costs are estimated at
$0.72/ton of waste feed.
Analytical Costs
Analyses—
In order to ensure that the unit is operating efficient-
ly and meeting; environmental standards, a program
for continuously analyzing waste feed, stack gas,
35
-------
ash, and water quality is required; typical costs are
$500/d as conducted for each day of unit operation.
Facility Modification, Repair, and
Replacement Costs
Maintenance—
Maintenance materials and labor costs are extreme-
ly difficult to estimate and cannot be predicted as
functions of a few simple waste and facility design
characteristics, because a myriad of site-specific fac-
tors can dramatically affect maintenance require-
ments. Maintenance costs have been estimated at
10% of equipment costs or $0.32M/yr.
Contingency—
In any cost estimate, 10% contingencies is an accept-
able factor. Contingency costs are estimated at
$0.32M/yr.
Site Demobilization Costs
Decontamination/Demobilization—
With the completion of activities at a specific site,
the unit must be decontaminated and demobilized
before being transported to its next location. Costs
that will accrue to this cost element include the final
burnout of residual material in the system, field la-
bor and supervision, decontamination equipment
and materials, utilities, security, health and safety
activities, and site staff support. Estimated costs are
?0.15M.
References
1. USEPA, 1988. Technology Evaluation Report,
SITE Program Demonstration Test, Shirco In-
frared Thermal Incineration System, Peak Oil,
Brandon, FL. EPA/540/5-88/0020. USEPA, Risk
Reduction Engineering Laboratory, Office of Re-
search and Development, Cincinnati, OH 45268.
2. USEPA, 1989. Technology Evaluation Report,
SITE Program Demonstration Test, Shirco Pilot-
Scale Infrared Incineration System, Rose
Township Demode Road Superfund Site.
USEPA, Risk Reduction Engineering Research
Laboratory, Office of Research and
Development, Cincinnati, OH 45268.
3. Final Report: Demonstration Test - Onsite PCB
Destruction - Shirco Infrared Portable Unit at
Florida Steel Indiantown Mill Site, Indiantown,
Fla., Report Number 821-86-1, ECOVA Corp.,
Dallas, TX, Sept. 17,1986.
4. Technical Paper: Remediation of PCB-Contami-
nated Soils by Mobile Infrared Incineration, G.
J. McCartney and J. E. Burford, OH Materials
Corp., Findlay, OH, 1988.
5. Draft of USEPA TSCA Permit to OH Materials
Corp. to Dispose of Polychlorinated Biphenyls
(PCBs), OH Materials Corp., Findlay, OH, 1988.
6. Telephone communication between Sy
Rosenthal, Foster Wheeler Enviresponse, Inc.,
and George Hay, OH Materials Corp., Findlay,
OH, March 14,1989.
7. Thermal Destruction Unit Demonstration Test
Plan for LaSalle Electric Utilities, PCB
Abatement Project, Illinois EPA, Springfield, IL,
Dec. 988.
8. Remedial Investigation Report, LaSalle Electric
Utilities Site, Illinois EPA, Springfield, IL,
1988.
9. Illinois Environmental Protection Agency
Operating Approval to Westinghouse
HAZTECH, Inc. for use of Thermal Destruction
Unit at LaSalle Electric Utilities PCB
Abatement Project, Illinois EPA, Springfield, IL,
Nov. 23,1988.
10. Summary of Remedial Alternative Selection,
LaSalle Electric Utilities, LaSalle, IL, Illinois
EPA, Springfield, IL, Mar. 29,1988.
11. Telephone communication between Sy
Rosenthal, Foster Wheeler Enviresponse, Inc.
and Richard Lange, Illinois EPA, Springfield,
IL, Mar. 30,1989.
12. Demonstration Test Report: PCB Destruction
Unit, Shirco Portable Demonstration Unit -
Infrared Incineration — Twin Cities Army
Ammunition Plant, ECOVA Corp., Dallas, TX,
Oct. 23,1987.
13. Final Report: Onsite Incineration Testing at
Brio Site, Friendswood, Texas, Shireo Infrared
Systems Portable Test Unit, Report No. 846-87-
1, ECOVA Corp., Dallas, TX, Feb. 10-13,1987.
14. Final Report: Onsite Incineration Testing of
Tibbetts Road Superfund Site, Barrington, N.H.,
Shirco Infrared Systems Portable Test Unit,
Report No. 834-87-1, ECOVA, Corp., Dallas, TX,
Nov. 2,1987.
15. Final Report: Onsite Incineration Testing of
Shirco Infrared Systems Portable Pilot Test Unit
at International Paper Co. Wood Treatment
Plant, Joplin, MO, Report No. 804-86-2, ECOVA
Corp., Dallas TX, May 29,1986.
16. Summary Report: Onsite Incineration Testing of
Times Beach Dioxin Research Facility, Times
Beach, MO by Shirco Infrared Systems Portable
36
-------
Test Unit, July 8-12,1985", Report No. 815-85-1,
ECOVA Corp., Dallas, TX, Aug. 28,1985.
17. Environmental Performance Evaluation of the
Shirco Infrared Incinerator, Dioxin Destruction
Demonstration Program, Times Beach, MO.,
ERT Report No. C-3044-D-1, ECOVA Corp.,
Dallas, TX, Aug. 29,1985.
18. Results of the Shirco Infrared Incinerator
Synthetic Hazardous Waste Tests, ECOVA
Corp., Dallas, TX, May 7,1985.
19. Vendor Information from ECOVA Corp., Dallas
TX, 1989.
20. McCormick, R. J., Cost for Hazardous Waste
Incineration, Noyes Publications, NJ, 1985.
21. Mortenson, H., et al., Destruction of Dioxin-
Contaminated Solids and Liquids by Mobile
Incineration, USEPA, Hazardous Waste
Engineering Research Laboratory, Land
Pollution Control Div., Releases Control Branch,
Edison, NJ, 1987.
22. Statement of Qualifications and Experience,
Riedel Environmental Services, Inc., Portland,
OR, Dec. 1988.
23. Occupational Health and Safety Act (OSHA), 29
CFR Part 1910 - Subpart Z.
37
-------
-------
APPENDIX A
PROCESS DESCRIPTION [1,2]
The transportable Shirco Infrared Thermal
Destruction System consists of a waste preparation
system and weigh hopper, infrared primary-
combustion-chamber, supplemental propane-fired
secondary-combustion-chamber, emergency bypass
stack or diesel generator and auxiliary emergency
shutdown system, venturi/scrubber exhaust system,
and data collection and control system —all mounted
on transportable trailers. The system process flow
and the overall test-site layout as employed at the
Peak Oil site are presented schematically in Figure
A-l.
Solid waste feed material is processed by waste
preparation equipment designed to reduce the waste
to the consistency and particle sizes that can be
processed by the unit's PCC. The equipment must be
specified for each site based on the condition of the
waste. After transfer from the waste preparation
equipment, the solid waste feed is weighed and
conveyed to a hopper mounted over the furnace
conveyer belt. A feed chute on the hopper distributes
the material across the width of the conveyor belt.
The feed-hopper screw-rate and the conveyor-belt
speed-rate are used to control the feedrate and bed
depth.
The PCC is a rectangular box insulated by layers of
ceramic fiber. Combustion air is supplied to the PCC
through a series of air ports at points along the
length of the chamber. The gas flow in the
incinerator is countercurrent to the conveyed feed
material. Electric infrared heating-elements
installed above the conveyor belt heat the waste to
the designated temperature (nominally 1,600°F),
which results in desorption or incinerator of organic
contaminants from the feed. Rotary rakes gently
turn the material to ensure adequate mixing and
complete desorption. When the thermally treated
soil (now referred to as furnace ash) reaches the
discharge end of the chamber, it is cooled with a
water spray and then is discharged by a crew-
auger/conveyor to an ash hopper. Ash analyses will
determine whether the ash can be transferred to a
storage area or returned to the waste material
stockpile for reprocessing.
Exhaust gas containing the desorbed contaminants
exits the PCC into an SCC (or afterburner) where
propane-fired burners combust residual organic
compounds into CO2, CO, HC1, and H2O. The SCC is
typically operated at 2,200°F and a gas residence
time exceeding 2 s. Secondary air is supplied to
ensure adequate excess oxygen levels for complete
combustion. Exhaust gas from the SCC then is
quenched and scrubbed by a water-fed venturi-
scrubber emissions-control-system to remove
particulate matter and acid gases. An induced draft
fan transfers the gas to the exhaust stack for
discharge to the atmosphere.
The main unit controls and data collection system
are housed in a specially designed van.
An emergency bypass stack is mounted in the system
directly upstream of the venturi scrubber for the
diversion of hot process gases under emergency shut-
down conditions. An alternative emergency design
incorporates a diesel-fuel-powdered generator set
that is linked to a standby direct-drive induced-draft
fan and scrubber pump. This emergency backup
system is activated by a power failure or the loss of
the primary induced-draft fan.
The process flow concept of the Shirco trailer-
mounted pilot-scale infrared incinerator system is
used at the Rose Township Demode Road Superfund
site is essentially the same as for the transportable
facility. Figure A-2 illustrates the pilot facility, for
general information purposes only.
Typical design parameters of the primary and
secondary combustion chambers of the transportable
Shirco unit employed at the Peak Oil site are
summarized in Table A-l.
39
-------
Control
Van
Feed Hopper &
Feed Module
Ash
Discharge
Combustion Chamber
i u u n.n u M i
|—~| Combustion
Secondary
Combustion Chamber
Forced Air l~—|—»•
Blower
Combustion Air
Blower
Emergency
Bypass
Stack
Chemical Chevron
Recycle Recycle
Pumps Pumps
Makeup
Water
.Tank
Water
Conditioner
Activated
Carbon
Filter
Slowdown Water to POTW
Figure A-1. System process flow and overall test site layout -- Peak Oil.
40
-------
Fencing
Drum Storage Area
•Waste Feed
•Ash
•Slowdown Water
•Empty Drums
Shirco Pilot-Scale Infrared Incinerator System
"tf
t
t
t
&
•Waste Feed
Transfer
Drums to
Pails
Manual
Feed Feed
System
rirnary Combustion Chamber
Burner Forced
Air Blower
•"ft
SCC Vapor Outlet Duct
t
Secondary Combustion Chamber
H
Exhaust
Stack
Induced
Draft
Fan
•Control Cabinet
•Belt Speed Control
•Burner Control
•Light Panel
•Motor Control Center
•Transformer
•Electrical Service
•HEPC
Overflow
h^P
Contamination
Reduction
Zone
'Ash/
Drum
/ Propane / /
Fuel
Supply
/Water/
Drum
/ Scrubber / /
Slowdown
Drum
At Grade
Makeup Water
from Water
Supply Trailer
Figure A-2. System process flow for the pilot-scale unit.
-------
Table A-1. Transportable Shlrco Unit Design Parameters
Primary Combustion Chamber Specifications
Shell height, in.
Channel height (belt to insulation), inc.
Shell width, in.
Module width, in.
Cavity width, in.
Shell length, ft
Effective length, ft
Bed depth, in.
Residence time, min
Feed module
Zone A
Zone B
Discharge module
Installed power, kW
Zone A (3 modules, 9 heating elements per module)
Zone B (3 modules, 9 heating elements per module)
Secondary Combustion Chamber Specifications
Shell height, in.
Height, In.
Shall width, in.
Width, in.
Shell length, ft
Mixing length, ft
Residence length, ft
Total length, ft
Mixed gas residence-time, s
Operating temperatures, °F
Mixing section
Residence section
Exhaust
Auxiliary fuel capacity, MBtu/h
131
34
142
108
94
61
52
0.5-2.0
5-60
300-1,400
600-1,800
600-1,800
600-1,800
500 kVA
500 kVA
131
96
121
84
83.5
24
40
72
>2
1,400-2,600
1,800-2,600
1,600-2,600
6.8
42
-------
APPENDIX B
VENDOR'S CLAIMS FOR THE TECHNOLOGY [3]
THIS APPENDIX TO THE REPORT IS BASED
UPON CLAIMS MADE BY ECOVA EITHER IN
CONVERSATIONS OR IN WRITTEN OR PUB-
LISHED MATERIALS. THE READER IS
CAUTIONED THAT THESE CLAIMS AND
INTERPRETATIONS OF THE REGULATIONS
ARE THOSE MADE BY THE VENDOR AND ARE
NOT NECESSARILY CORRECT OR ABLE TO BE
SUBSTANTIATED BY TEST DATA. MANY OF
ECOVA'S CLAIMS ARE EVALUATED IN
SECTION 3 AGAINST THE AVAILABLE TEST
DATA.
Introduction
The Shirco infrared thermal destruction process was
first designed in 1970 for use as a municipal sludge
treatment system. One unit is still in operation in
Alaska for sewage sludge treatment. As a result of
the need for destruction of hazardous wastes, Shirco
constructed a pilot-scale infrared demonstration
system in 1984 designed to treat hazardous wastes.
The pilot-scale system then was tested at a series of
sites as highlighted in Appendix D. The first full-
scale transportable system was delivered to Haztech
in 1986 for use at the Peak Oil Superfund Site near
Tampa, Fla. Other transportable units have been
delivered to OH Materials, which has recently
completed a cleanup at Florida Steel, Indiantown,
Fla.; and to Riedel Environmental Services.
Shirco Infrared Systems, Inc. has recently filed for
bankruptcy, and ECOVA Corp. of Redmond, Wa. has
purchased a license from Shirco Infrared Systems,
Inc. to construct 2 commercial and 2 pilot-scale units.
ECOVA intends to construct, own, and operate the
infrared thermal destruction systems as part of their
overall remediation capabilities. Other licenses are
available.
Potential Applicability
The Shirco system has broad process capabilities and
can be adapted to a wide range of wastes and
material compositions. However, the system can
process only solid wastes or sludges with a minimum
particle size of 5 microns in diameter that contain a
minimal amount of free liquids. The primary
combustion chamber conveyor belt cannot contain
undersized or free flowing materials or liquids,
which will pass through the conveyor screen
openings. Waste can be preprocessed, if necessary, by
dewatering, soils blending, and/or lime addition to
ensure a solid/semi-solid matrix suitable to the
process. Pure liquids can be processed by blending
with a suitable carrier, such as soil or vermiculite, to
form a semi-solid waste matrix, or they may be
injected directly into the secondary combustion
chamber.
Large objects such as clumped material, rocks, wood,
and light metals must be shredded and processed to a
maximum particle size of 2-in. diameter because of
the clearance between the conveyor belt and the
rotary rakes that gently turn the material on the
conveyor belt. Oversized material also includes
contaminated residuals such as tyvek suits, masks,
shoes, and other stich shreddable material.
The system provided by ECOVA to treat scrubber
water also can be used to treat onsite contaminated
and decontamination water. Recovered liquids with
heating value can be injected into and burned in the
secondary chamber or added to the feed to enhance
the feedstock heating value. Spent activated carbon
from the scrubber-water treatment system can also
be charged to the PCC for thermal treatment.
Table B-l presents a range of solid/semi-solid waste
characteristics suitable for processing in the Shirco
43
-------
system. Also shown in Table B-l are the
characteristics of waste materials processed by the
pilot-scale unit during previous demonstration
, programs. For example, the waste handling
capabilities of the system range from a relatively
dry, low-organic-content soil such as that processed
at Times Beach, to a sticky, oily sludge typical of a
wood preserving waste (creosote/pentachlorophenol).
PCB-contaminated soils and sludges also have been
processed.
The Shirco system is controlled by an automatic
process system that monitors key streams and
parameters and adjusts process parameters to ensure
adherence to previously established operating
conditions. These streams are interlocked and
alarmed, allowing manual override if needed. Thus
variations in feedstock characteristics immediately
will be sensed, and appropriate adjustments made
either automatically or manually to ensure that
destruction efficiencies are maintained, and
emissions levels are not exceeded. There are no
uncontrolled emissions during upset conditions.
The transportable systems can be easily transported
to the customer's site by either truck or rail, and the
entire system can be removed from the site after the
cleanup is completed. The system can be operated
with either available electrical service or portable
electrical generation equipment.
ECOVA now plans to provide full laboratory and
pilot-scale unit operations for making site-specific
recommendations for processing, and for providing
data to support permit applications.
System Advantages
Some of the claimed advantages of the Shirco system
over conventional thermal destruction technologies
are:
• The Shirco infrared process uses no fossil fuel in
the PCC, which greatly reduces total gas flow in
the system. This results in reduced requirements
for exhaust and pollution control equipment and
reduced particulate entrainment that must be
removed.
• Gentle handling of the feed material reduces
particulate emissions as the feed material passes
through the furnace on the conveyor belt. The
conveyor belt feed ensures more uniform
exposure of waste to heat since consistent flow
through the furnace is ensured. This is in
contrast to the tumbling motion of a rotary kiln
in which the exposure (residence time) of discrete
particles is not controlled; the conveyor belt feed
also offers an improvement over the possibility
of particle carry-through in a fluid bed
incinerator. The Shirco system provides intimate
exposure of the conveyed waste feed to the heat
source by use of "cakebreakers", rotating fingers
positioned along the bed length that gently stir
the conveyed waste feed material
• The Shirco process uses a distinct type of thermal
insulation that can withstand extreme thermal
shock, thus allowing the system to be quickly
started or shut down. The unit's operating
temperature can be changed instantaneously to
respond to feed variations, rather than requiring
stepwise changes as in other processes. The
ceramic fiber insulation also reduces the weight
of the Shirco equipment, which enhances its
mobility and-reduces site foundation load
requirements.
• The Shirco system has precise process control of
the gas flowrate, residence time, and the
multizone temperature profile. As a result of the
discrete areas of heat application and zoned
temperature control, the system effectively uses
the energy value in the waste to reduce auxiliary
Table B-1. Waste Characteristics - General
Moisture, wt%
Inerts, wt%
Volatile liquids, wt%
Volatile solids, wt%
Heating value, Btu/lb
Sulfur, wt%
Chlorine, wt%
Density (Ib/ft3)
Form
Hazardous
constituent, ppm
Applicable
range
0-50
20-100
0-25
0-100
0-10,000
0-4
0-5
30-130
Solid
semi-solid
0-1,000,000
PCP/Creosote
test
10-50
30-50
5-23
15-55
2,700-6,200
0.2-1.0
0.1-3.4
55-75
Soil
oily sludge
100,000-
250,000
Dioxin tests
15-20
95
0
5
Nil
Nil
Nil
60-70
Soil
0.25
PCB tests
1-20
30-80
0-15
3-50,
0-4,500
Nil
0.2
80-120
Soil
oily sludge
75-2,800
44
-------
energy requirements; more than 90% of the
energy applied is used efficiently.
Excess air can be set from 0 to 100% to be
compatible with thermal decomposition needs.
The starved-air pyrolytic operation volatilizes
liquid wastes and frees organics from con-
taminated soils; the resulting organic gases are
destroyed in a gas-fired secondary burner. The
reduced air usage of a pyrolytic process also
reduces the necessity to handle large gas volumes,
thereby decreasing the size of the offgas process
equipment, with subsequent reductions in capital
and operating costs. A 40% reduction of
combustion-air/gas-production is achieved by the
use of infrared technology versus other
incineration systems that rely on direct fuel oil or
fuel gas combustion. Other incineration systems
generally require 150% excess air; the infrared
system typically requires 50%.
The unit can immediately be restarted after a
process interruption; there is no need to purge the
system before restart.
• Countercurrent flow of the air in the Shirco
system reduces utility requirements for heating
process air.
Some advantages of using the Shirco system for
onsite processing are:
• Avoidance of risks and liabilities involved in
transporting wastes to offsite facilities.
Reduced long-term site-contamination risks and
liability by providing effective elimination of the
hazardous wastes.
• A cost-effective alternative to other onsite
disposal options of equivalent capacity.
• Requirements for only temporary permitting, and
the capability to be quickly installed and operated
without extensive facilities and site preparation.
• If residue materials do not meet destruction
specifications they can be reburned.
It should be noted that permanent Shirco
installations also are available for ongoing control of
waste disposal problems.
System Limitations
Limitations of the system are:
• The system can only process solid wastes or
sludges with a minimum particle size of 5 microns
that contain a minimal amount of free liquids.
• The system cannot process solid wastes with a
particle size greater than 2 in. All large bulk
items, such as drums, must be shredded and sized.
• The system cannot process liquids unless they are
blended with solid carriers to form a semi-solid
feed matrix within the size constraints discussed
above.
• Preprocessing of the waste to conform with the
above sizing constraints is extremely important.
The unit's ability initially to accept the feed
matrix through the feed module and pass it along
to the conveyor belt is a critical design
consideration.
• The system is optimally designed for a nominal
commercial throughput of 100 ton/d of waste feed.
For large sites, multiple infrared systems would
be required to provide throughput comparable to a
400-toh/d rotary kiln or other larger scale
transportable incineration system. This would be
an impractical and uneconomical alternative to
the single unit.
Cost Information
ECOVA's major thermal-destruction competitors are
rotary kilns. The infrared technology has a
substantial economic advantage over kilns; capital
costs are nearly 65% less than for a rotary kiln. As
discussed above, this advantage has a limitation as
the quantity of material to be processed increases;
one 400-ton/d rotary kiln would be equivalent to four
100-ton/d infrared systems. Use of multiple Shirco
systems would preclude any capital- cost/operating-
cost advantage.
The following cost data was based on the project
analysis and computerized database provided by
ECOVA Corp. for the operation of a transportable
Shirco system.
Table B-2 presents an economic model for a current-
case (1989) ideal Shirco transportable unit operation.
The cost analysis is based on a 140-ton/d unit capable
of treating 30,000 tons of waste feed, at an 85%
utilization factor. Based on this economic model,
Figure B-l presents an analysis of thermal
treatment costs for the Shirco transportable system
based on unit capacity and total waste feed treated.
The project cost for the economic model presented in
Table B-2 is $131/ton exclusive of waste feed
excavation, feed processing, materials handling and
water and ash residual disposal costs. The economic
analysis presented in Figure B-l indicates that
treatment costs vary based on unit capacity and tons
of material processed; they range from $85 to $175
per ton of waste feed for a 220-ton/d unit, to $167 to
$267 per ton of waste feed for a 100-ton/d unit
45
-------
Table B-2. Economic Model for Shirco Transportable (Commercial) Unit
Basis
Total tonnage of waste
Unit capacity.
Operating factor:
Materials handling not included
Schedule
Mobilization/demobilization
Incineration (@ 85% operating factor)
Total toe on site
30,000 ton
140 tpd
85% utilization
30 d
252 d
282 d
Project costs
Equipment mobilization/demobilization
Daily costs (a,b)
Misc. one-time costs (c)
Equipment rental
Incinerator amortization (d)
Materials handling
Trial burn sampling and analytical (d)
Incineration analytical services (f)
Incineration (a. g)
Utilities and Supplies (h)
Electricity
Fuel gas
Chemicals
Oil addition
Water
Total Project Cost
Profit (i)
Total price, J|
Price, S/ton
120,211
155,000
515,227
not included
75,000
126,050
1,010,177
605,042
302,521
60,000
150,000
14,521
3,133,749
783,438
3,917,187
131
46
-------
Table B-2. Notes
(a) Annual costs calculated to an Equivalent Daily Rate (EDR) @ 365 d/yr.
Direct labor
Labor classification
9 operators
1 site supervisor
1 maintenance supervisor
1 clerical
1 project manager
Total
Benefits (@ est. 0.2387 x straight wage)
Overtime wages (@ 1.5 x straight wage)
Annual
salary, $
25,000
35,000
35,000
35,000
45,000
Annual
straight
wages, $
225,000
35,000
35,000
:35,000
;45,000
375,000
Annual overtime,
h
840
840
0
0
0
89,513
157,500
Total Annual labor w/overtime and benefits
Equivalent daily rate
Annual cost, $
622,013
1,704
Per diem
Lodging - 13 rooms @ $50/d
Meals - @ $25/d/person
Autos -7@$120/wk
Travel - 1 trip/person/mo @ $600/trip
Total per diem
Equivalent daily rate
237,250
118,625
43,680
93,600
493,155
1,351
Maintenance
5% of $3M (spare parts - labor provided
by operators
Equivalent daily rate
150,000
411
Facilities
Office trailer
Break trailer
Parts trailer (2)
Telephone, utility, office supplies, etc
Decon trailer
Safety supplies
Building
Total facilities
Equivalent daily rate
3,600
3,600
7,200
20,000
no materials handling
44,000
15,000
93,400
256
47
-------
Table B-2. Notes (Continued)
Project technical support permitting
2,080 h @ $35/r
Travel, expenses -52 wk @ 500/wk
Total technical support
Equivalent daily rate
Total equivalent daily rate (EDR)
Director labor
Per diem
Facilities
Maintenance
Project technical support/permitting
(b) Total daily costs for equipment mobilization/demobilization
30 d mobilization/demobilization at $4,007 EDR
(c) Miscellaneous one-time costs (estimated):
Equipment transportation
Installation subcontractor services (crane, electricity, etc.)
Grading, foundations, utility extensions
Total
(d) Incinerator amortization:
Principal
Annual interest
Term
Monthly payment
Time on site
Project cost at $54,792/mo x (282/30)
72.800
31,200
104,000
285
1,704
1,351
256
411
285
4,007
120,211
Lump sum cost, $
35,000
20,000
100,000
155,000
Annual cost. $
$3,200,000
11%
84 mo
$54,792
282 d
(e)
(0
(0)
(h)
Trial bum sampling and analytical costs (estimate)
Incineration analytical services:
500/d (estimate) x 252 d operation
Incineration operating costs:
252 d operation x $4,007 EDR
Utilities:
Electricity
Fuel gas
Chemicals
Oil addition
Water
1,000 kWh/h @ $0.10/kWh
10MBtu/h@$5.00/MBtu
$2.00/ton of waste
6.25 gal/ton waste @ $0.80/gal
50 gpm @ $0.80/1,000 gal
Total
(i) Profit
Profit margin of 20% on total project cost of $3,133,749
515,227
75,000
126,050
1,010,177
605,042
302,521
60,000
150,000
12,521
1,132,084
783,438
48
-------
o
"c
(D
I
£
I
CD
260
Shirco Thermal Treatment Costs
9' x 61' Transportable Unit
100TPD
140TPD
O 180TPD
A 220 TPD
80
Quantity of Waste Treated (Thousands Tons)
Figure B-1. Shirco thermal treatment costs - transportable (commercial) unit.
49
-------
-------
APPENDIX C
SITE DEMONSTRATION RESULTS [1,2]
Introduction
Two SITE demonstrations were performed using the
Shirco infrared thermal destruction technology: one
at the Demode Road Superfund site in Rose
Township , MI; and the other at the Peak Oil Super-
fund site in Brandon, FL. The Peak Oil Superfund
site used a transportable commercial unit to ther-
mally process approximately 7,000 tons of waste oil
sludge contaminated with PCBs and lead during a
removal action by EPA Region IV. The demonstra-
tion test was conducted from July 31 to Aug. 5, 1987.
The Demode Road site demonstration test was per
formed from Nov. 2 to 13, 1987 and utilized the
Shirco pilot-scale system to process approximately
4,000 Ib of soils contaminated with PCBs, heavy
metals, and organic compounds. The detailed results
of these demonstrations can be found in the Tech-
nology Evaluation Reports [7,8]. The result of these
two SITE demonstrations are presented in this
Appendix as Appendix C-l: Shirco Infrared Thermal
Destruction System, Peak Oil Superfund Site; and
Appendix C-2: Shirco Pilot-Scale Infrared Incinera-
tion System, Rose Township Demode Road Super-
fund site.
51
-------
-------
APPENDIX C-1
SITE DEMONSTRATION RESULTS
SHIRCO INFRARED THERMAL DESTRUCTION SYSTEM
PEAK OIL SUPERFUND SITE [1]
Introduction
Beginning in the 1950s, Peak Oil, an oil rerefiner,
operated a used-oil processing facility in Brandon,
Fla. Various waste streams from the rerefining
operation were dumped into a natural lagoon located
on the property. The lagoon quickly became
contaminated with PCBs and lead contained in the
waste, and as in the majority of Florida's delicate and
shallow aquifer systems, the result was contam-
ination of local drinking-water supplies. The U.S.
Environmental Protection Agency (EPA) added the
site to the National Priorities List (NPL) primarily
because of the contamination of groundwater by
PCBs.
Because of the existence of an imminent hazard,
EPA Region IV initiated and supervised a removal
action at the site. The Region contracted with
Haztech, Inc., an emergency removal/cleanup
contractor, to incinerate approximately 7,000 tons of
waste sludge contaminated with PCBs and lead.
In November 1986, Haztech began setting up the
transportable Shirco unit. The commercial unit was
transported to the Peak Oil Superfund site on 5
separate trailers. It can process between 100 and 200
tons of waste feed per day, depending on feedstock
characteristics. The nominal capacity of the unit is
100 ton/d. All of the trailers were designed for
overland travel by means of removable wheel
assemblies and trailer hitch gear, which allows
hauling as a truck-trailer rig over highways.
Once onsite, the wheel assemblies were removed,
and the units were field-connected together on
suitable poured-concrete pads or existing concrete
bases to assemble the complete system. The
components were connected together to form the 67-
ft-long primary combustion chamber, a 72-ft-long
secondary combustion chamber (afterburner), an
emissions control system, and a process-manage-
ment and monitoring-control center.
The SITE demonstration was conducted from July 31
to August 5,1987 during commercial operation of the
Shirco unit.
Feed Preparation
As part of the overall site remediation, the sludge
lagoon was drained of water and mixed with sand,
soil, and lime to form a conditioned waste soil
matrix. The lime, in addition to providing a binding
medium for the wet matrix , neutralized the highly
acidic wastes in the lagoon, the original site
contaminant produced as a by-product of the acid-
based oil rerefining operation .
The conditioned soil was transferred from the lagoon
to the material stockpile area by front-end loaders; a
loader then was used to transfer the waste feed to a
power screen. The gross waste feed was loaded onto a
tipping reject grid where large rocks and debris were
rejected. The bulk of the feed fell through the grid to
a belt feed hopper. The waste feed was then passed
through a shredding system and conveyed to the
vibrating power screen assembly. The shredding
system and the vibrating screens provided an
aerated and conditioned waste feed sized to less than
1 in. while rejecting larger pieces of rocks, roots, and
other materials that were not removed at the tipping
reject grid.
The prepared waste feed was then loaded into the
weigh hopper using a track loader until a
predetermined weight was attained. At that time,
waste feed to the weigh hopper was stopped and
waste was conveyed from the weigh hopper to the
PCC feed hopper by an inclined conveyor belt.
Test Procedure
The SITE program at Peak Oil was conducted from
July 31, 1987 to August 4, 1987. During this period,
EPA was present to observe the unit operation,
53
-------
collect data, and document the mechanical operating
history of the system and the problems encountered
in operating this type of commercial incineration
unit.
The overall program consisted of three replicate test
runs conducted under the normal operating
conditions of the unit. During one of these runs, a
duplicate set of samples was taken and analyzed to
satisfy rigorous quality- assurance/quality-control
(QA/QC) protocols. EPA documented all operating
conditions during the test runs and conducted
extensive sampling of the solid waste feed, stack gas,
furnace ash, scrubber liquid effluent, scrubber water
inlet, scrubber solids, and ambient air. QA/QC audit
teams from ORD observed and evaluated QA/QC
protocols for both the sampling and analytical phases
of the test program.
Results
A summary of the SITE demonstration results is
presented in Table C-l.l. Detailed results and
operating summaries are presented in the
Technology Evaluation Report [1]. The following
discussion summarizes results and conclusions from
the demonstration. Specific operating problems and
critical operating parameters are also discussed
based on the evaluation of the unit's performance
during the removal action at the site.
Characteristics of the Feed
Various waste streams from the rerefining operation
of a used oil processing facility were dumped into a
natural lagoon on the property that became
contaminated with PCBs and lead contained in the
waste. As part of the overall site remediation," the
sludge lagoon was drained of water and mixed with
sand, soil, and lime to form a conditioned waste-soil
matrix. The lime, in addition to providing a binding
medium for the wet matrix, neutralized the highly
acidic wastes in the lagoon, the original site
contaminant produced as a by-product of the acid-
based oil-rerefining operation.
The concentrations of several metals including
aluminum, iron, lead, and zinc exceeded 1,000 ppm
due to the metals pickup in the original automotive
waste oil and metals concentrations in the rerefining
operation sludge. Inorganics such as calcium,
magnesium, phosphorous, and sodium reflect the
addition of lime and sand to the original sludge
lagoon.
Total PCB concentrations ranged from 3.48 to 5.85
ppm and averaged 4.63 ppm. These low PCB
concentrations in the waste feed were the result of
mixing the original oily waste having up to 100 ppm
of PCBs with the PCB-free surrounding soil, lime,
and sand so that the resulting material could be
handled and processed as a solid waste. No dioxins or
furans were detected. Several volatile and semi-
volatile organic compounds were also detected,
including many aromatic hydrocarbons, which would
be expected in the waste oil from the automotive
combustion process.
Characteristics of the Furnace Ash
The concentrations of metals and inorganics were
similar to concentrations in the waste feed, thus
indicating that the mass flow of these species
remains with the high mass flow of furnace ash that
exits the unit.
Total PCB concentrations ranged from 0.007-0.900
ppm and averaged 0.422 ppm. These values are
below the TSCA guidance level of 2 ppm and indicate
effective PCB decontamination through the unit. No
dioxins or furans were detected. Several volatile and
semivolatile organic compounds were also detected
and can be considered as possible PICs, although
some of these compounds were identified in the
waste-feed.
Mobility of Heavy Metals - Comparison of
Feed and Furnace Ash
In order to determine whether heavy metals,
particularly lead, would leach from the furnace ash
produced in the Shirco unit, EP Tox and TCLP tests
were conducted to determine the mobility of heavy
metals from the furnace ash as compared to the feed.
The EP Tox results for lead in the leachate ranged
from 24 to 57 ppm for the feed and 25 to 46 ppm for
the furnace ash. The TCLP results ranged from 2. 5
to 35 ppm for the feed and 0.008 to 0.84 ppm for the
furnace ash.
A comparison of the EP Tox analyses conducted on
the furnace ash and the feed do not show any trend or
evidence that indicates reduced mobility of lead from
the furnace ash as compared to the feed as a result of
the thermal treatment. A comparison of the TCLP
analyses conducted on the furnace ash and the feed
indicates reduced mobility of lead from the furnace
ash as compared to the feed as a result of the thermal
treatment. The comparisons also reveal that the
concentrations of lead in the TCLP leachates from
both the feed and the furnace ash were consistently
lower than the corresponding EP Tox test levels on
the same samples.
The significant differences in results from these two
analytical techniques have been documented in a
recent Oak Ridge National Laboratory report
54
-------
Table C-1.1. Site Demonstration: Summary of Test Results
8/1/87 8/2/87
8/3/87
8/4/87
Waste feed characteristics
Moisture, wt%
Ash, wt%
HHV, Btu/lb
PCB, ppm
Pb, ppm
Chlorine, ppm
Sulfur, ppm
Chlorine (as HCI), kg/h
Sulfur (as SO2), kg/h
Stack gas
HCI, ppmv
SO2, ppmv
HCI, g/h
SO2, g/h
Particulates (@7% 02), mg/dscm
PCB, pg/h
Ash
PCB, ppm
Pb, ppm
EP Tox (Pb), mg/L, ppm
TCLP (Pb), mg/L, ppm
Operating Conditions
Waste feedrate (avg. daily), kg/h
ORE (PCB), wt%
DE (PCB), wt%
Primary combustion chamber
Exhaust temperature (avg.), °F
Residence time, min
Secondary combustion chamber
temperature(avg-), °F
Residence time, sec
Acid gas removal efficiency, wt%
HCI
SO2
16.63
69.77
2,064
5.85
5,900
< 1,000
25,300
<5
200
< 0.051
0.99
<0.8
27.40
358
57.50
0.01
7,100
25.0
0.01
3,328
99.99967
99.88
1,797
' 19
1,886
>3
>99
>99.9
16.06
69.80
1,639
3.85
4,900 :
< 1 ,000
17,800
<5
132
0.60
41.80
8.60
1070.0
211
174.50
0.240
6,000
28.0
0.01
3,287
99.99880
93.77
1,836
19
1,887
>3
>99
>99.1
14.24
72.40
1,728
5.34
5,000
< 1 ,000
18,900
<5
138
0.22
0.96
2.90
22.0
173
58.10
0.900
6,400
36.0
0.02
3,626
99.99972
83.15
1,922
18
1,889
>3
>99
>99.9
14.37
75.21
2,018
3.48
4,400
< 1,000
16,700
<5
125
0.20
0.91
2.70
20.6
171
126.20
0.540
6,200
36.0
0.01
3,600
99.99905
84.48
1,885
19
1,907
>3
>99
>99.9
(ORNL, Leaching of Metals from Alkaline Wastes by
Municipal Waste Leachate, ORNL/TM-11050, Mar.
1987). It appears that the differences in the test
procedures and alkalinity of the matrix (which is
significant here because of the lime pretreatment of
the original waste-site feed-matrix) cause a
difference in the pH environment that is sufficient to
affect the solubility and leachability of heavy metals,
particularly lead.
Mobility of Heavy Metals - Comparison to
EP Tox and Proposed TCLP Toxicity
Characteristic Standards
EP Tox and TCLP tests were conducted on the feed,
furnace ash, scrubber water, and scrubber solids. All
of the results of the EP Tox tests on the feed and the
furnace ash exceeded the 5 ppm toxicity charac
teristic standard for lead (24-57 ppm). Two samples
of the feed exceeded the proposed TCLP toxicity
characteristic standard of 5 ppm for lead (8.6 ppm
and 35 ppm). All of the furnace ash samples passed
the TCLP standard. For the other heavy metals, all
of the results were below their respective toxicity
standards.
Destruction and Removal Efficiency
(ORE) of PCBs
PCBs were analyzed in the solid waste feed, furnace
ash, scrubber effluent solids, stack gas, scrubber
liquid effliient, and scrubber water inlet.
The Shirco unit achieved a DRE for PCBs ranging
from 99.99972% to 99.99880%. The RCRA DRE
standard is 99.99%, and the TSCA DRE standard is
99.9999%.
55
-------
It should be noted that the unit was operated to
produce an ash that contained 1 ppm or less of PCB,
as mandated by the EPA Region IV On-Scene
Coordinator. This operating standard was lower than
the minimum 2 ppm TSCA standard for an allowable
PCB concentration in a product stream. PCB
concentration in the waste feed varied from 5.85 to
3.48 ppm during the tests, and therefore a unit
operation based on a DRE for PCBs was impractical
because of the difficulty in measuring extremely low
PCB concentrations in the stack emissions.
Other Organic Stack Gas Emissions
Several volatile and semivolatile organic compounds
were detected in the stack gas at concentrations
greater than 50 ppt. Low levels of several phthalate
compounds were also detected in blank samples and
may be traced to contamination from plastic
components in the process, sampling equipment, or
laboratory apparatus. Other volatile and
semivolatile organic compounds, which may
represent PICs, were detected. They include:
halomethanes; chlorinated organics; aromatic
semivolatile compounds; other volatile organics,
including benzene, toluene and ethylbenzene;
oxygenated hydrocarbons, and p-chloro-m- cresol.
Dioxins and furans were not detected in the stack gas
samples.
Acid Gas Removal—
Measured HC1 emission rates ranged from less than
0.8 to 8.6 g/h. Since the chlorine concentration in the
solid waste feed was below the 0.1% detection limit,
it is impossible to determine actual HC1 removal
efficiency. However, SO2 emissions were less than
1,100 g/h, with an average 149 kgyh SO2 feedrate
giving an average removal of SC>2 in excess of 99%.
SC>2 is more difficult to remove than HC1 in a caustic
scrubber, and the tests show that HC1 removal
should be in excess of the 99% determined for SOg
removal.
Particulate Emissions—
The particulate emissions during the first day were
358 mg/dscm. The unit was cleaned and mechanical
adjustments were made resulting in an emission rate
of 211 mg/dscm during the second day. The emissions
during the third day were 172 mg/dscm (average of
duplicate measurements). These values exceeded the
RCRA standard of 180 mg/dscm during 2 of the 4
sampling periods.
Particulate emissions were analyzed for metals. The
results indicate that the emissions control system
could not effectively control the flows of metals,
which were the principal contaminants and cause of
the system's inability to consistently meet RCRA
particulate emissions standards. In particular, the
concentrations of lead (58 wt%), sulfur (16 wt%), and
sodium (1.9 wt%) were extremely high. Based on
initial concentrations in the pretreated feed, the use
of sodium carbonate solutions in the emissions
scrubbing system, and the carryover of lead salts as
fines, the predominance of these species on the
particulate emissions becomes more likely as the
emissions control system becomes overloaded.
Analysis of Scrubber Makeup Water, Scrubber
Water, and Scrubber Solids—
No extraordinary levels of semivolatile or volatile
organics were detected in any of the streams. No
PCBs or dioxins or furans were detected.
The major concentration of contaminants was found
in the scrubber solids, as determined by an analysis
of the sludge obtained from the bottom of the water
blowdown. The significant concentrations of metals
and inorganics resulted from feed concentrations of
metals and inorganic salts that adversely affected
the emissions control system and the stack
emissions.
Operations
Introduction
A review of the Haztech, EPA Technical Assistance
Team (TAT), and EPA logbooks and progress reports,
plus discussions with unit and project personnel,
provided a summary of mechanical and operating
problems encountered in this first application of a
commercial Shirco incineration system at a
Superfund site. The startup of the unit began on Dec.
31, 1986 and continued until Oct. 13, 1987, with
7,110 tons of waste feed processed. Based on a
capacity of 100 ton/d, the overall operating or
utilization factor experienced by this transportable
Shirco infrared thermal destruction system at the
Peak Oil Superfund site was 24%. This assumes that
the unit required 71 operating days to process the
waste feed,-but that the unit remained at the site for
286 days.
Feed Preparation and Handling
Feed Preparation—
The Peak Oil waste feed matrix was a solidified
sludge that was prone to agglomeration and caused
clogging, bridging, and jamming of the original
crusher equipment. Prior to the SITE demonstration
(May 10, 1987), the crusher was replaced with a
power screen that shredded, screened, and aerated
the feed to a consistency and size that was
accommodated by the Shirco feeder.
56
-------
Conveyor—
Conveyor system problems included spillage of waste
feed, waste material sticking to the conveyor belt,
and an inability to adjust feedrate from the conveyor
to the unit's feeder system. Modifications to the
conveyor system included the addition of a "skirt"
below the conveyor to catch spillage, a conveyor
scraper that minimized sticking, and a variable
speed controller and revised motor arrangement that
provided feedrate control.
Primary Combustion Chamber Section
Feed Inlet—
The screw augers and their motor drives experienced
continuous clogging and overload problems. The feed
system required continuous attention by operating
personnel and the addition of "bridgebreakers" to
reduce the bridging of the agglomerating waste feed.
The screw augers are designed with reversing
capability, and the motor drives are designed for a
50% overload based on adequate feed preparation. If
the feed is not properly crushed, screened, and
prepared, the augers' materials -handling efficiency
will decrease; bridging and plugging problems —
particularly with an agglomerating feed matrix —
will occur, causing significant overload and eventual
burnout to the motor drives.
Ash Outlet—
The ash removal system required frequent
maintenance and unit downtime. The cooling screw
and incline screw were continually clogging and
breaking, and their motor drivers would overload
and burn out. When the screws were reversed to
dislodge material under the screw flights, breakage
and further abuse of the motors would occur. Dusting
and odor problems were evident in and around the
ash removal system.
The intermittent failure of the original feed-
preparation system (i.e, crusher and screen) to
deliver a consistently sized waste feed allowed
unprepared materials to enter the unit. The
unprepared feed caused occasional jamming and
blockage of the ash discharge system. Plugging of the
incline screw was also caused by the buildup of ash in
the discharge chute and improper control and
monitoring of the ash quench facilities.
Miscellaneous Systems—
In addition to the feed-inlet and ash-outlet system
difficulties, problems also occurred with conveyor
belt failures, cakebreaker failures, and belt-conveyor
system maintenance.
The original belt installed at the Peak Oil site was
provided with several test sections of various alloys.
Because of the nature of the feed material and
minimal knowledge of its chemical characteristics,
this approach was selected so that if belt failure did
occur, an appropriate alloy then could be installed.
Due to the chlorine and sulfur content of the initial
feed material, certain test sections failed and were
replaced with the standard Type-314 stainless steel
alloy. A properly cured Type-314 stainless steel belt
has provided reliable service through the completion
of the project.
In addition, possibly due to the mechanical failures
in feed screening and crushing noted above and to
the resultant feeding of unsized or nonspecification
waste material, the cakebreakers also may have
been subject to severe stress when these articles were
encountered, causing cakebreaker failure .
Although problems were encountered with the belt
conveyance system, it appears that the roller bearing
specifications do not require any changes. Proper
attention to lubricant choice and a rigorous
maintenance schedule are required to ensure a long
roller bearing and belt conveyance system operating
life.
Secondary Combustion Chamber Section
The only operating problem that affected the SCC
was the failure of several burner blocks. Proper
curing of the burner blocks is required prior to
achieving operating temperatures. A slow curing of
the burner blocks prior to operation may not have
been fully performed. In addition, numerous startups
and shutdowns of the unit subjected the blocks to
cooling and heating cycling that adversely affected
block life. Changes to the burner block have been
incorporated in the current design to allow for
symmetrical expansion and contraction and
minimization of stress points observed at Peak Oil,
and to move the flame front farther away from the
blocks, thus extending their life.
Emissions Control Section
Quench/Venturi System—
The original quench/venturi system design consisted
of two stainless steel quench tubes where the hot
exhaust gases from the SCC are cooled with quench
water sprays. The cooled gases enter the dual
fiberglass^reinforced-plastic (FRP) Venturis where
water injection at the venturi throats atomizes and
increases particulate precipitation as the gases
proceed into the scrubber system. The system, as
operated, was modified to a one-pass quench/venturi
flow with a venturi pressure drop exceeding 15 psi.
There were indications based on the cracking and
scorching of the FRP venturi section and warpage of
57
-------
the scrubber internals that the systems may have
been subjected to excessive process temperatures
probably caused by a failure of the quench system
and its cooling sprays. The high temperatures
exhibited by the gas exiting the quench system
probably were the result of low gas flow and
subsequent channeling of the exhaust gas stream
through only one pass of the dual quench tubes and
Venturis. Because of the channeling, the gas stream
was not exposed to the full cooling effect of the spray
nozzles, and damage to the downstream FRP systems
resulted.
The particulate precipitation effect at the Venturis
also suffered due to the channeling of the low gas
flow. In addition, the cracking of the FRP venturi
section also may have occurred because the anchor
bolts on the venturi support structure may not have
been loosened during installation of the system to
allow for thermal expansion of the quench tubes.
Compounding the loss in cooling and particulate
removal efficiency caused by the gas channeling was
the plugging of the water sprays, which reduced the
overall quench and venturi water flows and spray
coverage. This plugging may have been caused by
the excessive salts content in the quench water.
Scrubber System—
The scrubber is a horizontal cross-flow design that is
capable of scrubbing exhaust gases and meeting
regulatory requirements for acid gas removal and
particulate loading. The scrubber system at Peak
Oil, however, apparently could not control partic-
ulate emissions at the quantities and quality of the
particulates encountered. Because of the excessive
fines loadings and excessive salts content in the
scrubber water streams, the scrubber system not
only exhibited high stack particulate loadings, but
also was burdened by the significant salts buildup in
the scrubber water streams requiring higher
blowdown and fresh makeup-water rates.
Induced Draft Fan System—
Because of the particulate carry-over from the
scrubber, plating of the induced-draft fan blades
occurred, causing blade imbalance and fan vibration.
It does not appear that the design of the fan is
contributing to the problem. A water spray system
has been added at the fan to periodically wash the
blades of plated salts and minimize vibration
problems.
Emissions Control Section Redesign—
The emissions control section of the Shirco unit that
was employed at the Peak Oil site has been replaced
with a high efficiency Calvert scrubber that is
designed to provide improved particulate removal
efficiencies over a wider range of gas flow and fines
loading.
Costs
Several cost scenarios are developed, based on a
model for a 100-ton/d Shirco unit equivalent to the
unit that operated at Peak Oil. The economic
analysis concludes that the cost to operate this
commercial unit ranges from $1 96/ton at an
acceptable utilization factor of 80% to $425/ton at a
utilization factor of 37%, which reflects a corrected
actual operation of the unit at Peak Oil. The cost
analysis excludes vendor profit, waste excavation,
feed preparation and ash disposal costs.
58
-------
APPENDIX C-2
SITE DEMONSTRATION RESULTS
SHIRCO PILOT-SCALE INFRARED INCINERATION SYSTEM
ROSE TOWNSHIP DEMODE ROAD SUPERFUND SITE [2]
Introduction
The SITE program demonstration of the Shirco Pilot-
Scale Infrared Incineration System for thermal
treatment developed by Shirco Infrared Systems, Inc.
of Dallas, Tex., was conducted at the Demode Road
Superfund Site in Rose Township, Mich. The Demode
Road site is a 12-acre waste site previously used to
bury, dump, and store industrial wastes such as
paint sludges, solvents, and other wastes containing
PCBs, oils and greases, phenols, and heavy metals.
PCBs and lead were the principal contaminants in
the soil used for the test of the Infrared System. The
demonstration was conducted from November 2-13,
1987 and treated 1,799 kg (3,967 Ib) of contaminated
soil under various test conditions.
Feed Preparation
The demonstration used soil from an area of the site
that was highly contaminated with PCBs and lead,
as determined in the original remedial investiga-
tions performed at the site. Pretest sampling and
analysis further identified those sectors within the
area most highly contaminated with PCBs and lead
for excavation. Other organics and heavy metals
were also present in these sectors. Soil from these
sectors, to be used as feed for the test, was excavated
and mixed into a pile using a front-end loader, and
then screened to remove aggregate and debris
greater than 1 in. in diameter. The screened soil was
drummed and transferred to a designated zone
adjacent to the test unit. During the demonstration,
the feed material was transferred from the drums to
pails, weighed, and then manually fed to the unit
through a hopper mounted on the unit. Two drums of
soil were blended with 3-wt% fuel oil to be used for
several of the test runs to investigate the effect of
increased feed heating value on overall unit perform
ance and energy consumption at varying operating
conditions.
Test Procedure
A total of 17 test runs were conducted. Three runs
were performed under design operating conditions to
assess overall unit operation and system perform-
ance (Phase I), and 14 runs were conducted under
varying operational parameters to evaluate their
effect on system performance and energy consump-
tion (Phase II).
The Phase I runs were conducted at a 1,600°F PCC
temperature, a 2,200°F SCC temperature, and a PCC
residence time of 20 min. Each of the three runs was
sufficiently long (6-10 hours) to gather a large
enough sample of stack gas to analyze it for PCBs.
An additional run was conducted at the same opera-
ting conditions to obtain specific stack samples that
had not been successfully collected during the two
previous runs.
The Phase II runs were conducted for approximately
1 h under varied operating conditions: PCC temper-
ature 900°, 1,200°, 1,400°, and 1,600°F; SCC
temperature 1,800° and 2,200°F; PCC feed residence
time 10,15,20, and 25 minutes; and PCC combustion
air flow on-off to simulate oxidizing or non-
oxidizing/pyrblytic PCC atmosphere.
A summary of the operating program is presented in
Table C-2.1.
Results
A summary of the SITE demonstration results is
presented in Table C-2.2. Detailed results and
59
-------
Table C-2.1. Operations Summary
Primary combustion chamber
Secondary combustion chamber
Date
11-02-87
11-03-87
11-04-87
11-05-87
11-06-87
11-07-87
11-08-87
11-09-87
11-10-87
11-11-87
11-12-87
11-13-87
Time period
15:20-17:13 (d)
10:20-11:40
11:40-18:25
07:33-09:45
09:45-20:13
07:55-08:16
08:16-18:37
09:55-10:32
10:32-15:31
00:00-24:00 (e)
00:00-24:00 (e)
09:34-10:35
11:26-12:27
13:45-14:47
09:30-10:30
11:20-12:20
13:13-14:14
15:07-16:07
09:04-10:05
11:30-12:31
13:20-14:20
15:10-16:10
10:27-11:27
12:35-13:35
10:12-11:12
Test run
0
-
1
.
2
.
3
-
1-2(f)
10
14
5
6
8
15
18
7
9
11
13
16
17
19
Temperature
°F-
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,200
1,200
1,400
1,600
1,600 (b)
1,200(b)
900 (b)
1,600
1,200 (b)
1,200 (b)
1,200 (b)
900 (b)
900 (b)
1,600 (c)
Res. time
min
20
20
20
20
20
20
20
20
20
15
20
15
15
15
20
10
25
20
15
25
20
15
Waste
feedrate
Ib/h
88.89
66.53
69.09
73.11
82.86
77.41
128.57
70.23
78.68
95.41
65.81
110.77 (a)
120.00 (a)
100.33 (a)
80.00 (a)
114.10
62.95
73.85
88.29
65.00
70.00
88.29 (a)
Temperature Res time
°F s
2,200
2,200 2.60
2,200
2,200
2,200
2,200
2,200
2,200
1,800
1,800
2,200
1,800
1,800
1,800
1,800
2,200
2,200
2,200
2,200
2,200
1,800
1,800
(a) Waslo feed blended with approximately 3 wt% fuel oil.
jb) Test runs conducted under non-oxidizing atmosphere.
(c) Primary combustion chamber bed depth set at 1 in. All other test runs conducted at 1-1/2-in. bed depth.
(d) System shakedown; no formal sampling.
(o) Weekend; unit shutdown.
(0 Run 1-2 was conducted to make up for incomplete sampling runs that were to be conducted during Runs 1 and 2.
operating summaries are presented in the Technol-
ogy Evaluation Report [2]. The following discussion
presents summary results and conclusions from the
demonstration.
Characteristics of the Feed
Based on data from the previous remedial investi-
gation of the site, a specific area within the site was
identified with the highest concentrations of both
PCBs and lead, the major soil contaminants of
concern. The remedial investigation also described
the soil as a dry, brown, sandy, and silty clay topsoil,
which upon excavation proved to be an accurate
observation. Subsequent pretest sampling and
analysis of the specific area of the site identified
particular sectors with the highest contaminations of
PCBs and lead. A composite sample of all the sectors
within the area indicated a 7.8 pH, 9.0 wt% moisture,
81 wt% ash, less than 1,000 Btu/lb high heating
value, and a 0.95 g/cc density. The composite sample
contained 570 ppm of total PCBs and 580 ppm lead
(elemental lead after digestion and conversion to
inorganic form). A composite sample of the 10 sectors
chosen for excavation contained 626 ppm PCBs , 560
ppm of lead, 55 ppb of tetrachlorodibenzo-p-dioxin
(TCDD), and 4.2 ppb of tetrachlorodibenzofuran
(TCDF). Once the feed excavation was begun, it
became evident that the front-end loader could not
confine its large-scale activities to the 10 specific
sectors, and an area comprising 14 specific sectors
was excavated for the unit's feed source.
The feed characteristics of the soil were obtained
from an analysis of the composite of the grab samples
of feed taken during each of the test runs. In addition
to lead, for which concentrations ranged from 290 to
3,000 ppm and averaged 778 ppm, several other
metals were present at average concentrations
exceeding 50 ppm, including barium (591 ppm), zinc
(301 ppm), and chromium (85 ppm). Total PCB
concentrations ranged from 10.2 to 669 ppm and
averaged 272 ppm.
Several samples of the feed contained small
quantities of TCDFs ranging from 0.04 to 0.10 ppb.
Volatile and semivolatile organic compounds
including methyl ethyl ketone, trichloroethene, and
bis(2-ethylhexyl)phthalate were measured in feed
60
-------
Table C-2.2. Site Demonstration: Summary of Test Results
Operating conditions
PCC
Waste feed characteristics
Furnace ash characteristics
Temp
°F
900(a,b)
900(b)
900(b)
1,200
1,200
1,200(b)
1,200(b)
1,200(b)
1,200(a,b)
1,400
1,600
1,600
1,600
1,600
1,600
1,600(a)
1,600(a)
1,600(a,b)
Residence
time
min
20
20
25
20
15
25
20
15
15
20
20
20
20
20
10
15
15(C)
15
PCB
ppm
327
20.2
367
297
27.6
456
669
602
309
56.0
10.2
35.2
20.4
(f)
391
451
271
311
Pb
ppm
590
660
290
640
870
590
610
470
370
740
3,000
1,400
550
1,100
620
620
390
500
EPTox
(Pb)
ppm
0.29
0.67
0.32
0.05
0.20
0.12
0.20
0.18
0.21
0.07
0.15
0.20
0.23
0.14
0.25
ND
0.53
0.07
3.00(g)
TCLP
(Pb)
ppm
0.81
0.88
7.00
0.56
0.44
0.53
0.71
0.53
0.96
0.89
0.67
0.35
1.30
0.49
0.73
0.66
1.80
0.55
1.40(g)
PCB
ppm
2.079
3.396
0.168
0.115(d)
0.077
0.108(d)
0.066(d)
i
0.025(d)
0.066(d)
0.087(d)
0.037
0.112
0.003
(f)
0.045(d)
0.117(d)
0.004
0.061 (d)
(Pb)
ppm
1,000
1,400
860
1,100
1,000
1,200
1,200
2,000
1,000
1,600
1,100
1,300
1,100
420
1,700
840
1,500
800
EPTox
(Pb)
ppm
0.38
0.89
0.88
4.10
0.38
0.14
0.06
4.90(g)
(h)
0.46
ND
0.05
ND
0.13
0.28
ND
0.43
0.27
1.10
TCLP
(Pb)
ppm
2.90
6.20
3.80
1.60
3.60
0.05
4.10
2.80(g)
(h)
0.82
0.15
ND
ND
0.05
1.80
1.00
0.17
0.23
2.40
(a) Waste feed blended with 3 wt% fuel oil.
(b) Non-oxidizing atmosphere.
(c) PCC bed depth at 1 in. All other tests at 1 1/2 in.
(d) PCB levels below analytical detection limits. Total shown is sum of detectable limits indicated in analyses.
(e) ND - nondetectable value.
(f) Run was conducted to make up for incomplete semivolatile organics, PCDD/PCDF, soluble chromium, and stack gas particulate
samplings on other runs.
(g) Data from additional EP Tox and TCLP tests.
(h) ND due to broken sample container.
samples at concentrations less than 50 ppm. Methyl
ethyl ketone and triehloroethene were also detected
in solvent blanks and are attributed to analytical
laboratory contamination.
Characteristics of the of Furnace Ash
The characteristics of the furnace ash were obtained
from an analysis of the composited grab samples
taken at the conclusion of each test run. In addition
to lead, where concentrations ranged from 420 to
2,000 ppm and averaged 1,173 ppm, several other
metals were present at average concentrations
exceeding 50 ppm, including barium (1,061 ppm),
zinc (410 ppm), and chromium (8 1 ppm). Total PCB
concentrations ranged from 0.004 to 3.396 ppm. A
sample of furnace ash contained 0.07 and 0.30 ppb of
TCDF during each of two runs conducted at a 900°F
PCC operating temperature; the normal PCC
operating temperature is 1,600°F. These runs were
also conducted without the input of PCC combustion
air to simulate non-oxidizing or pyrolytic combustion
conditions . The low PCC temperature and pyrolytic
environment could have led to the incomplete
desorption or incineration of TCDF present in the
feed or to the production of TCDF from the
incomplete combustion of PCBs in the feed. Volatile
compounds (including methylene chloride, methyl
ethyl ketone, tetrachloroethene, and triehloro-
ethene) were also measured in the furnace ash
samples in concentrations ranging from 3.9 to 64
ppm, with one sample containing 980 ppm of
methylene chloride. Methyl ethyl ketone and
triehloroethene were also detected in solvent blanks,
and methylene chloride is commonly employed in
laboratory procedures; therefore these compounds
may be products of incomplete combustion and/or the
result of laboratory contamination.
Residual PCBs in Furnace Ash
During the demonstration test, 17 runs were
conducted at varying operating conditions. In
addition to the DRE levels, which are an indication
of the performance of the Shirco Pilot-Scale Infrared
Incineration System and its ability to meet RCRA
61
-------
and/or TSCA regulatory standards, the reduction of
PCB concentration from the feed to the furnace ash is
also a measure of the unit's ability to effectively
destroy PCBs and produce a furnace ash with a PCB
concentration below the TSCA guidance level of 2
ppm.
Two samples of furnace ash exceeded the TSCA
guidance level and contained 3.396 and 2.079 ppm of
total residual PCBs. The samples were produced
during two runs conducted at a 900°F-PCC operating
temperature (20 min residence time) , which is
significantly lower than the normal PCC operating
temperature of 1,600°F. These runs were also
conducted without the input of PCC combustion air
to simulate non-oxidizing or pyrolytic combustion
conditions. At this low PCC temperature and
pyrolytic condition, these higher total residual PCB
levels in the furnace ash may be the result of the
incomplete combustion of PCBs in the feed. This is
further substantiated by the residual TCDF present
in the furnace ash samples from these same two
runs, as discussed previously. The remaining runs
conducted at 1,200°, 1,400°, and 1,600°F resulted in
total residual PCB concentrations in the furnace ash
ranging from 0.003 to 0.117 ppm. A third run, which
was conducted at a 900°F PCC operating tempera-
ture but with an increased PCC residence time of 25
min resulted in a total furnace ash PCB concen-
tration of 0.168 ppm with no detectable TCDF. It is
possible that the increased residence time in the PCC
may have offset the low 900°F PCC operating
temperature, providing the additional processing
time for the satisfactory destruction of the PCBs in
the feed.
Mobility of Heavy Metals
Feed and Furnace Ash
Comparison of
In order to determine whether heavy metals,
particularly lead, would leach from the furnace ash
produced in the Shirco Pilot-Scale Infrared
Incineration System, EP Tox and TCLP tests were
conducted to determine the mobility of heavy metals
from the furnace ash as compared to the feed. The
initial EP Tox analyses for lead in the leachate
ranged from 0.05 to 0.67 ppm for the feed and 0.05 to
4.10 ppm for the furnace ash. The initial TCLP
analyses ranged from 0. 35 to 1.80 ppm (with one
sample at 7.0 ppm) for the feed and 0.05 to 4.10 ppm
(with one sample at 6.20 ppm) for the furnace ash.
A comparison of the EP Tox and TCLP analyses
conducted on the furnace ash and the feed do not
show any trend or evidence that indicate reduced
mobility of lead from the furnace ash as compared to
the feed as a result of the thermal treatment. The
comparison did reveal that the concentrations of lead
in the TCLP leachates from both the feed and the
furnace ash were consistently higher than the
corresponding EP Tox results from the same
samples.
When several samples were retested to verify the
results, the concentrations of lead in the EP Tox
leachates (4.9 ppm feed, 3.0 ppm furnace ash) were
higher than during the initial tests, and in direct
reversal to the original data, exceeded corresponding
TCLP leachate concentrations (2.8 ppm feed, 1.4 ppm
furnace ash). The results of the retest again did not
indicate reduced mobility of lead from the furnace
ash versus the feed as a result of the thermal
treatment.
Mobility of Heavy Metals - Comparison to
EP Tox and Proposed TCLP Toxicity
Characteristic Standards
EP Tox and TCLP tests were conducted on the feed,
furnace ash, scrubber water, and scrubber solids. All
of the results were below the EP Tox and proposed
TCLP toxicity characteristic standards - 5 ppm
arsenic, 100 ppm barium, 1 ppm cadmium, 5 ppm
chromium, 5 ppm lead, 0.2 ppm mercury, 1 ppm
selenium, and 5 ppm silver — except for 1 feed sample
at 7.0 ppm lead (TCLP) and 1 furnace ash sample at
6.2 ppm lead (TCLP). Despite concentrations of
heavy metals in the waste-feed and furnace as high
as 3,000 ppm and 2,000 ppm ( lead) respectively, in
most cases the concentrations of metals in the EP
Tox and TCLP leachates met their respective toxicity
characteristic standards.
Destruction and Removal Efficiency
(ORE) of PCBs
The ORE of PCBs for the first three runs (Phase I)
was greater than 99.99%. The regulatory standard
for incineration under RCRA is 99.99% ORE and
under TSCA is 99.9999% ORE. The low PCB
concentrations in the feed resulted in PCB levels in
the, stack gas that were below the analytical
detection limits for two of the runs. Therefore for
these runs, DRE is calculated based on the sum of the
detection limits of the PCB congeners. Stack gas
measurements conducted during the third run did
detect trichlorobiphenyl and tetrachlorobiphenyl
congeners, and a DRE is shown based on this
measurement. The less rigorous sampling in Phase II
of the test was not designed to allow calculation of
DRE.
62
-------
Other Organic Stack Gas and PCC
Offgas Emissions
Several volatile and semivolatile organic compounds
were detected in the stack gas at concentrations less
than 100 ppb and established standards for direct
inhalation. Low levels of several phthalate
compounds were also detected in blank samples and
may be traced to contamination from plastic
components in the process, sampling equipment, or
laboratory apparatus. Several volatile organic
compounds (including benzene and toluene) were
detected in the stack gas and the scrubber makeup
water and may be attributable to contamination
from the makeup water, although PIC formation is a
possibility. Other volatile and semivolatile organic
compounds, which probably represent PICs, were
detected. They include: halomethanes; chlorinated
species including chlorobenzene and methylene
chloride; other volatile organics including xylenes,
styrene, and ethylbenzene; oxygenated hydrocarbons
including acetone and acrolein; carbon disulfide; and
p-chloro-m-cresol. Dioxins and furans were not
detected in the stack gas samples.
The majority of the organic compounds present in the
PCC offgas samples at levels less than 500 ppb were
also present in the stack gas. The additional
destruction of organics that takes place in the SCC
and emissions scrubbing system reduced the
concentration of these organic compounds in the
corresponding stack gas samples.
Acid Gas Removal
During Phase I Runs 1-3, HC1 emissions ranged from
0.181 to 0.998 g/h, which were significantly below
the RCRA performance standard of 1,800 g/h. HC1
removal efficiencies ranged from 97.23 to 99.35 wt%.
Acid gas removal was not measured in Phase II.
Particulate Emissions
Particulate emissions were measured throughout the
demonstration and ranged from 7 to 68 mg/dscm,
well below the RCRA standard of 180 mg/dscm.
Analysis of Scrubber Makeup Water,
Scrubber Water, and Scrubber Solids
Scrubber makeup water was transported to the site
in a tank truck that may have contained some
residual contamination prior to fill up. Samples of
scrubber makeup water were taken at the end of
each run. No PCBs, dioxins, furans, or semivolatile
organic compounds were detected. Several volatile
organics (including benzene, toluene, and trichloro-
ethene) were measured at concentrations less than
15 ppm. The concentrations of heavy metals were all
less than 0.2 ppm.
Samples of the water recirculation through the
venturi scrubber system (referred to as scrubber
water) were also taken at the end of each run. PCB
concentrations were less than 200 ppt; no dioxins,
furans, or semivolatile organic compounds were
detected. Small quantities of benzene (2 ppm) and
toluene (5.7 to 11 ppm) were measured in several of
the samples and are attributable to the similar
contaminants in the scrubber makeup water. The
concentrations of heavy metals in the scrubber water
were all less than 1 ppm, except for barium, which
ranged from 0.2 to 2.2 ppm, and lead, which ranged
from 0.12 to 1.8 ppm.
Insufficient quantities of scrubber solids in the
scrubber water were available for analysis.
Overall Disposition of Metals
Total metals analyses of the feed, furnace ash, PCC
offgas and stack-gas particulates, scrubber makeup
water, scrubber water, and scrubber solids showed
that the majority of the detectable metals that
entered the unit with the feed (including lead)
remained in the furnace ash. An overall mass
balance of lead through the unit was calculated,
based on the analysis of lead in the samples, the
measured feedrate (as weighed during the runs'
operating periods), the calculated furnace ash
flowrate (based on the ultimate analysis of ash in the
feed sample), and the measured particle mass and
gas volume (obtained from the gas1 EPA Method 5
sampling trains). Phase I results indicate an average
lead mass flowrate of 28.3 g/h in the feed, 37.0 g/h in
the furnace ash, 0.206 g/h in the PCC offgas
particulates, and 0.109 g/h in the stack gas
particulates. The quantity of lead leaving the unit
with scrubber water effluent is approximately 0.204
g/h based on the maximum measured concentration
of 1.8 ppm lead in the scrubber water and an overall
approximate water flowrate of 30 gph. The PCC
offgas particulates sampled during the Phase I runs
contained an average of 5,364 ppm of lead as
compared to stack gas particulates, which contained
an average of 15,830 ppm of lead. By contrast, the
average concentration of lead in the feed was 1,550
ppm. Although the concentration of lead in the
particulate matter increases as the process flow
progresses through the unit, the actual mass flow of
lead decreases as the gas stream is cooled and treated
through the emissions control system.
For the Phase I runs, sampling and analysis
procedures were conducted to evaluate vaporous lead
concentrations in the PCC offgas and soluble
63
-------
chromium concentrations in the PCC offgas and
Stack-gas particulates. The special sampling for
vapor phase lead and soluble chromium was unable
to detect any of either metal at levels greater than
the detection limits of 2.7 and 264 ppb, respectively.
Other heavy metals, particularly barium and zinc,
with average concentrations exceeding 100 ppm in
the feed (barium 591 ppm, zinc 301 ppm) were also
present in high concentrations relative to other
heavy metals hi the furnace ash (barium 1,061 ppm,
zinc 410 ppm) and scrubber water (barium 0.8 ppm ,
zinc 0.3 ppm).
Optimum Operating Conditions
Phase II was designed to examine the effect of
varying operating conditions on energy consumption
and on residual levels of heavy metals and organics
in the furnace ash versus feed. Based on the data
obtained, an analysis was conducted to compare
energy consumption in the unit at operating
conditions that did not affect the performance of the
unit. A reduction in the PCC operating temperature
from 1,600° to 1,200°F reduced the average PCC
power usage 48%, from 0.2294 to 0.1200 kWh/lb feed.
A reduction in the SCC operating temperature from
2,200° to 1,800°F reduced the average propane fuel
consumption by 51%, from 3,997 to 1,952 Btu/lb feed.
The use of 3 wt% fuel oil to supplement the fuel value
of the feed further decreased PCC power usage by
26% and 67% at PCC operating temperatures of
1,600°F and 1,200°F, respectively, with accompany-
ing increases in overall feedrate of 32% and 26%. The
costs for fuel oil and its attendant facilities must be
examined for specific applications of the Shirco
system to determine the cost effectiveness of a fuel
oil additive to the waste feed.
As discussed in previous sections, the results did not
provide any trend in the residual levels of the heavy
metals and organics in the furnace ash versus the
levels in the feed as the operating conditions were
varied (and PCC operating temperatures maintained
at 1,200° to 1,600°F). At an abnormally low PCC
operating temperature of 900°F, without the input of
combustion air to simulate non-oxidizing or pyrolytic
combustion conditions, total PCB and TCDF concen-
trations in the furnace ash increased. The increases
may indicate that these PCC conditions led to
incomplete desorption or incineration of PCB and
TCDF and to the production of TCDF from the
incomplete combustion of PCBs in the feed.
Operations
There were no problems associated with the
operation of the Shirco Pilot-Scale Infrared Incinera-
tion System that would impact on the ability of a
commercial unit to process the waste feed at the
Demode Road Superfund site.
Specific functions for which problems may arise,
such as the feed preprocessing, screening, and
handling operations are manually performed during
pilot-scale testing and in general do not relate to any
scale-up considerations. The feed that was processed
did not present any physical and chemical properties
that would cause problems in a commercial
transportable unit. Further bench-scale tests are
important to evaluate the feasibility of any proposed
feed-pretreatment system.
An additional area of investigation focused on the
mobility of heavy metals in the furnace ash versus
the feed, as measured and compared to the EP Tox
and TCLP toxicity characteristic standards. The
results of the tests were inconclusive; there was no
evidence that the thermal treatment affected metals
leaching or mobility. Additional thermal tests are
needed to determine the effect that heavy metals
(particularly lead) will have on the furnace ash and
its ultimate storage and disposal. In general, based
on the results, the test demonstration of the pilot-
scale unit showed that the Shirco system is a viable
technology for application at the Demode Road
Superfund site.
64
-------
APPENDIX D-1
FLORIDA STEEL PILOT-SCALE TESTS [4]
Introduction
The Florida Steel Corp. conducted a feasibility study
to develop and evaluate various alternatives for
onsite treatment of PCB-contaminated soils dis-
covered at their Indiantown, FL mill site. The source
of the PCB contamination was from the use of
hydraulic fluid containing PCBs in the billet
shearing system. Leaks in the system allowed the
release of hydraulic fluid to the surrounding soils.
The purpose of the study was to aid in selection of a
method to cleanup the site. As part of this study, a
demonstration test using the Shirco pilot-scale unit
was conducted at the site on May 13-15, 1986, by
Shirco Infrared Systems of Dallas, TX. The purpose
of the test was to demonstrate the capability of the
Shirco infrared technology to detoxify the soil and
meet all the requirements of 40 CFR Part 761.
The demonstration program consisted of six tests.
Three soil mixes, with different levels of contam-
inants representative of the material stored at the
site, were used. Incinerator operating parameters
that were varied included soil residence time and
temperature of the secondary chamber.
Feed Preparation
The contaminated soil was excavated and stored in a
ground level vault. Representative samples of the
contaminated soils were stored in 55-gal drums at
the site. Material from 3 of those drums (designated
Mix 1, 2, and 3) were selected for processing in this
program.
The waste material was weighed in batches in
buckets and was manually placed on a metering belt
conveyor that fed the material to the furnace. The
conveyor is equipped with an adjustable gate that
can be used to regulate and distribute the feed.
Test Procedure
The test program was designed to evaluate the
effects of various operating conditions and waste feed
characteristics on overall system performance. Six
tests were performed in this program. A summary of
the operating parameters is given in Table D-l.l.
All runs except Tests 1 and 5 were performed with
Mix 1, which had the highest concentration of PCB.
Tests 1 and 5 were performed with Mixes 2 and 3,
respectively, which contained different concentra-
tions of organic constituents.
Feedrate to the furnace was controlled by the
metering-belt speed setting and the width of the
metering gate setting. For this program, the gate
setting remained constant to give a 1-in. bed depth;
therefore, feedrate depended on the belt speed and
other factors, such as density of the feed.
For this program, the operating temperature of the
primary chamber was anticipated at 1,650°F. Actual
operating temperatures were below that level
because the auxiliary energy requirements exceeded
the capacity of the power supply to the electrical
heating elements. The temperature of the secondary
chamber was varied from 1,900° to 2,200°F to
determine the effect of different temperature levels
on destruction effectiveness for these specific wastes.
Material residence time in the primary chamber was
established by adjustment of the belt speed. Two
residence times were used—15 and 25 min. The
minimum secondary-chamber residence-time
planned for this program was 2 s.
65
-------
Table D-1.1. Operating Parameters
Tost
Date
Time Test Begun
Timo Tost Ended
Wasta feed
Mix number
FoodratQ, lo/h
Total feed, Ib
Furnace belt speed, ft/h
Solid phase residence time, min
Process temperatures
Feed discharge, °F
Furnace Zone A, °F
Furnace Zone B, °F
FumacQ exhaust, °F
Aftofbumer, "F
Stack, «F
Furnace draft, in. H2O
Afterburner draft, in, H2O
Scrubber
Vonturi pressure drop in. H2O
Venturi water flow, gpm
Tower water flow, gpm
Slock exhaust
Average velocity, afpm
Ftow volume, dscfm
Avg. CO cone., ppm
Avg. CO2 cone., %
Avg, O2 cone., %
Process energy requirements
Furnace, Kw
Bluib feed
Afterburner, Btu/h
Btu/lbfoed
1
5-13-86
14:00
17:05
2
115.4
355.5
22.2
15
241
1.539
1,561
1,337
2,019
174
-0.006
-0.150
7.8
2.8
11.8
2,385
55
12.5
8.6
9.0
44.05
1,303
183,073
515
2
5-14-86
10:10
12:53
1
61.5
167.1
13.3
25
212
1,625
1,607
1.442
2,196
175
-0.019
-0.182
3.0
2.8
5.8
2,761
68
26.6
8.7
12.2
44.06
2,444
215,075
1,287
3
5-14-86
12:53
16:50
1
61.5
242.9
13.3
25
158
1,582
1,572
1.444
2,006
171
-0.005
-0.035
3.3
2.8
5.4
1,424
33
10.6
9.8
8.6
44.76
2,483
136,326
561
4
5-14-86
16:50
17:50
1
32.0
32.0
22.2
15
173
1,524
1,503
1,377
1,990
174
-0.007
-0.050
3.0
3.0
5.4
-
-
-
~
-
39.00
4,158
123,653
3,864
5
5-15-86
09:30
13:05
3
106.4
381.3
22.2
15
244
1,614
1,598
1,551
1,989
177
-0.018
-0.175
3.0
2.8
4.4
2,339
53
3.4
10.3
6.9
41.56
1,332
202,056
530
6
5-15-86
1318
1620
1
79.8
241.3
22.2
15
157
1,521
1,498
1.398
1,900
176
-0.024
-0.128- •
3/4
2.9
4.4
1.299
31
2.4
9.2
9.5
45.70
1,960
126,253
523
Sampling of the process was performed at 5
locations—waste feed hopper, ash hopper, scrubber
effluent, afterburner exhaust duct, and exhaust
stack. A complete set of samples was obtained for
each test, except exhaust stack samples were not
taken for Test 4.
Results
The primary objective of this program was to confirm
the ability of the Shirco infrared system process to
decontaminate polychlorinated biphenyl (PCS)
laden soils and to incinerate the PCBs with a DRE of
66
-------
Table D-1 .2. Demonstration Test Summary
Test 1
Date 5-13-86
Time Test Begun 14:00
Time Test Ended 1 7:05
Operating parameters:
Furnace:
Process power rate, MBtu/hr
Avg. residence time, min
Avg. process temp., °F
Afterburner:
Propane fuel rate, MBtu/h
Avg. process temp., °F
Avg. comb, air, acfm
Avg. oxygen, %
Avg. CO2, %
Avg. CO, ppm
Combustion efficiency, %
Particulate/HCL emissions:
Sample time, min
Stack flowrate, dscfm
Paniculate cone. @7% O2,
gr/dscf
HCL, mg/h
PCBs
Waste feedrate, Ib/h
PCB cone., ppm
PCB feedrate, g/h
0.150
15
1,531
0.183
2,015
117
8.96
8.65
12.5
99.99
-0.006
129
55
0.015
<181
115.4
76
3.98
2
5-14-86
10:10
12:53
0.150
25
1 ,603 -
0.215
2,177
135
12.20
8.67
26.60
99.97
-0.019
60
68
0.055
<136
61.5
2,790
77.83
3
5-14-86
12:53
16:50
0.153
25
1 ,573
0.136
1,993
70 ;
8.58
9.78
10.60
99.99
-0.005
94
33
0.023
<45.3
61.5 ;
2,560
71.41
4
5-14-86
16:50
17:60
0.133
15
1,523
0.124
2,007
-
9.63
9.13
7.47
99.97
-0.007
—
NA
NA
NA
32.0
2,970
43.11
5
5-15-86
09:30
13:05
0.142
15
1,610
0.202
1,980
115
6.9
10.3
3.35
99.99
-0.018
130
53
0.037
<408
106.4
400
19.30
6
5-15-86
1318
1620
0.156
15
1,471
0.126
1,883
64
9.49
9.22
2.40
99.99
-0.024
51
31
0.017
<227
79.8
2,840
102.80
Furnace ash
PCB cone.,
<2.4
<2.6
<3.4
<2.6
<2.6
<2.6
Scrubber effluent composite
PCB cone., ng/kg
<0.34
Flue gas flowrate, m3/h
PCB cone., ng/m3
PCB output, ng/h
92.75
< 3214.34
< 29.16
114.94
< 709.73
<81.6
55.93
< 946.16
<52.9
NA
NA
NA
88.22
2,416.32
213.2
51.36
< 1,719.47
< 88.31
Destruction and removal, %
> 99.999*
> 99.9999
> 99.99992
NA
99.9989*"
> 99.99991
•Required DRE not met due to limited analytical detection limit.
'Low DRE is due to periods of operation with secondary-chamber oxygen-levels approaching the permit condition of 3% excess.
67
-------
Tabta D-1.3.Waste Characterization of Drummed Soils
Moisture. wt%
Inorts, wl%
Organics, wl%
Hoating value, Btu/Ib
Density, (b/ft3
Form
Chlorine, wt%
Sutfur. wt%
PCB, pom
1
13.64
84.52
1.84
220
90
Soil
Nil
Nil
150
2
13.59
82.77
3.64
430
90
Soil
Nil
Nil
500
99.9999%, a combustion efficiency of 99.9% and
maximum particulate emissions of 0.08 gr/dscf. A
summary of the pilot-unit test results is presented in
Table D-1.2. The following discussion presents sum-
mary results and conclusions from the demon-
stration tests.
scrubber sump was drained and a sample was taken
in a 1-L glass jar. The sample was prepared by
solvent extraction and concentration. PCB
concentration was determined by GC/MS (EPA
Method 680). No PCBs were found in the composite
sample at a detection limit of 0.34 ppb.
Particulates and HCI Emissions in the
Stack Gas
The concentrations of particulates in the stack gas
ranged from 0.015-0.055 gr/dscf when corrected for
stack oxygen concentration. The results, given in
Table D-1.5 are in compliance with the RCRA and
TSCA performance standard of 0.08 gr/dscf.
The concentration of hydrochloric acid (HCI) for each
of the tests is given in Table D-1.6. The total HCI
flows were less than 0.001 Ib/h, which is significantly
below the RCRA performance standard of 4 Ib/h that
would require a 99% HCI removal efficiency.
Characteristics of the Feed
Prior to the demonstration test, samples taken from
2 of the drums of contaminated soils stored at the
Indiantown Mill were tested for physical charac-
teristics. Results of the tests are given in Table D-
1.3. These data were used to determine initial
process operating conditions. Composite samples
taken during the tests were tested for PCBs. Those
test results are given in Table D-1.4.
Characteristics of the Furnace Ash
A grab sampling procedure was used to obtain a
representative, time-averaged sample of the furnace
ash. The pilot-scale unit was equipped with an ash
sampling drawer located in the ash discharge chute.
A portion of the furnace ash that drops off the
furnace belt into the ash discharge hopper is
captured in the sampling drawer. The sampling
drawer has a capacity of approximately 50 mL. This
drawer was emptied periodically during each test
and composited in a 500 mL glass jar.
The furnace ash samples were analyzed for PCBs.
The analyses were unable to detect any PCB isomers
at levels greater than their detection limits, which
range from 0.2 to 1.8 ppm.
Characteristics of the Scrubber Water
The scrubber was operated in a recirculation mode
and make-up water was added throughout the day as
needed. At the end of each day's operation, the
Total Chlorinated Organics in the Stack
Gas
A series of two sorbent tubes containing activated
carbon were used to trap sorbents for later
determination of total-organic-halide emissions by
the use of EPA Method 450.1.
Table D-1.7 gives the results of the analysis for total
organic halogens. These results ranged from 64.6 to
1,210 ug/L.
Continuous Monitoring of Secondary
Chamber Exhaust
A continuous system was used to monitor levels of
flue-gas carbon monoxide, carbon dioxide, oxygen,
and oxides of nitrogen. The continuous monitoring
system consisted of a sample-gas conditioning
system, gas analyzers, and a data acquisition/
recording system.
The concentrations of fixed gases and nitrogen oxides
in the secondary- chamber discharge-stream were
continuously recorded. The average values and
ranges are given in Table D-1.8.
It was planned that oxygen concentration in the
exhaust gas would be maintained at 3% to insure
adequate waste/air contact in the secondary
chamber. The oxygen content was maintained at
greater than 6.5%, except for two brief periods in
Test 5 when the concentration dropped to approxi-
mately 3.2%.
68
-------
Table D-1.4. Concentration of PCBs in Waste Feed, ppm (mg/kg)
Detection Test 1*
Isomer Limit
Cl (1)-PCB 2.5
Cl (2)-PCB 2.5
Cl (3)-PCB 2.5
Cl (4)-PCB 5.0
Cl (5)-PCB 5.0
Cl (6)-PCB 5.0
Cl (7)-PCB 7.5
Cl (8)-PCB 7.5
Cl (9)-PCB 7.5
Cl (10)-PCB 12.5
Sum PCB
"Duplicate analysis
ND— Not detectable
Table D-1.5. Paniculate Emissions
Total particulate, mg
Sample volume, dscf
Grain loading, gr/dscf
Corrected loading, gr/dscf"
'Corrected for stack oxygen concentration
Table D-1.6. HCL Emissions
Impinger chloride cone., mg/L
Impinger volume, mi-
Sample volume, dscf
Gas chloride cone., mg/m3
Stack vol. flowrate, m3/h
HCL emissions, Ib/h
TableD-1.7. Total Organic Halogens
ug/sample
Sample volume, L
Concentration, iig/L
Mix 2
ND/ND
7/11
28/35
18/30
0.6/ND
ND/ND
ND/ND
ND/ND
ND/ND
ND/ND
53.6/76
Test 1
30.200
45.796
0.010
0.015
Test 1
<3
676
45.796
<2
92
<0.00
04
Test 1
165,000
26.9
64.6
Test 2
Mix 1
48
740
1,060
770
170
ND
ND
ND
ND
,ND
2,790
Test 2
115.700
48.916
0.036
0.055
Test 2
<3
571
48.916
<1
130
<0.00
03
Test 2
890
--
--
Tests
Mix 1
30
650
1,000
710
,160
5.4
ND
ND
ND
ND
2,560
Tests
44.700
36.116
0.019
0.023
Test3
<3
400
36.116
<1
69
<0.00
01
'
Tests
8,600
20.8
414
Test 4 Test 5 Test 6
Mix 1 Mix 3 Mix 1
50 3.4 27
770 71 740
1,200 200 1,100
790 110 810
160 15 160
ND ND ND
ND ND ND
ND ND ND
ND ND ND
ND ND ND
2,970 400 2,840
Test 5 Test 6
77.500 14.10
39.107 19.72
0.031 0.01
0.037 0.017
Test 5 Test 6
9 <6
493 322
39.107 19.72
4 <4
100 67
<0.00 <0.00
09 05
Test 5 Test 6
7,300 10,500
20.8 8.7
351 1,210
69
-------
The combustion efficiencies for all runs exceeded the
TSCA performance standard of 99.9%.
PCBs in the Stack Gas
A modified EPA Method-5 train was used to sample
for prganics. The various portions of the train were
individually prepared and then composited. The
composite was extracted and concentrated. PCBs
were determined by GC/MS (EPA Method 680).
The results of the PCB analyses of the stack gas
samples are given in Table D-1.9. No detectable
amounts of any isomer group were detected except
for the Test 5 samples where the concentration of
PCBs was 2.4 ug/m3.
Destruction and Removal Efficiency
(DRE) of PCBs
Table D-1.10 presents the DRE of PCBs for the tests.
The DREs were calculated using the analytical
detection limits for the samples, except for the Test 5
samples where a concentration of PCBs was detected
above the detection limit. The detection limits were
sufficient to demonstrate DREs in excess of the
99.9999% TSCA performance standard for Tests 2, 3,
4, and 6. A review of all the data for Test 1 indicates
that the TSCA DRE standard could not be
demonstrated because of the high detection limit and
the unexpectedly low concentration of PCBs in the
waste feed used in that test. The DRE for Test 5 was
99.9989%. The presence of the PCBs in Test 5 was
most likely a result of two periods of low excess
oxygen in the secondary chamber. These results
show that, for this unit, minimum permissible
oxygen levels in the secondary chamber exhaust
must be increased from that used for this program.
PCDDs and PCDFs
Composite samples of the furnace ash, scrubber
water, and the stack gas sample from Test 6 were
analyzed for PCDD and PCDF tetra-octa isomers.
The tests results showed that none of those materials
were present at the detection limits.
Operation
No operating problems were reported except that,
due to lack of sufficient power capacity, primary
chamber temperatures could not be maintained at
desired levels. It was recognized that sufficient
oxygen must be available in the secondary chamber
to assure adequate destruction of PCBs.
Although the power requirements and operational
experiences of the pilot-scale unit are not scalable to
commercial size units, the pilot-scale tests and
analyses establish the range and recommended
operating parameters for the optimum operation of
the full-scale unit.
Tablo O-1.8. Continuous Monitoring Emission Results
Parts Per Million
Percent
Combustion
Tost Efficiency, %
1 99.986
Z 99.969
3 99.989
4 99.992
5 99.997
6 99.997
Avg.
92.2
114.0
79.9
80.7
81.8
64.5
NOX
Range
(79.8-95.0)
(84.9-107.0)
(62.3-73.6))
(76.4-96.6)
(66.0-93.0)
(63.0-66.0)
Avg.
12.5
26.6
10.6
7.47
3.35
2.40
CO
Range
(9.2-17.2)
(24.4-33.5)
(10.1-12.3)
(6.76-7.84)
(2.44-4.88)
(2.4-2.9)
Avg.
8.96
12.2
8.58
9.63
6.92
9.49
02
Range
(8.05-11.7)
(8.74-11.6)
(6.5-13.0)
(8.5-10.75)
(3.22-12.2)
(6.5-11.8)
CO2
Avg.
8.65
8.67
9.78
9.13
10.3
9.22
Range
(6.0-9.10)
(8.0-8.96)
(7.76-10.9)
(8.25-9.90)
(8.46-11.7)
(8.3-10.8)
70
-------
Table D-1.9. Concentration of PCBs in Flue Gas Samples, ng/Sample
Isomer
Cl <2)-PCB
Cl (2)-PCB
Cl (3)-PCB
Cl (4)-PCB
Cl (5)-PCB
Cl (6)-PCB
Cl (7)-PCB
Cl (8)-PCB
Cl (9)-PCB
Cl (10)-PCB
Sum PCB
ND Not detectable
Table D-1. 10. Destruction
Waste feed
PCB concentrations
Ml/9
Feedrate, Ib/h
Flue gas
PCB concentration
ng/g
Flowrate, m3/h
Destruction and
removal, %
Detection
Limit
190
290
370
75
75
75
110
110
110
180
Test
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
and Removal Efficiency
Test 1
76.0
115.4
< 314.40
92.75
> 99.999
1 Test 2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Of PCBS
Test 2
2,790.00
61.50
< 709.73
114.94
> 99.999
Test 3
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Test 3 ;
2,560.00
61.50
< 946. 16
55.93
> 99.99992
Tests
ND/ND
1,800/1,900
3,200/3,300
500/800
ND/ND
ND/ND
ND/ND
ND/ND
ND/ND
ND/ND
5.5/6
Test 4 Test 5
2,970 400.0
61.5 106.4
NA 2,416.32
NA 88.22
NA 99.9989
Test 6
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Test 6
2,840.0
79.8
< 1,71 9.47
51.66
99.99991
71
-------
-------
APPENDIX D-2
FLORIDA STEEL TSCA TRIAL BURNS [5,6]
Introduction
In 1982, an environmental audit of Florida Steel
Corp.'s metal-recycling plant at Indiantown, Fla.
was performed. The site was found to be contam-
inated with PCBs, which had leaked from hydraulic
equipment used at the site. The PCB contaminated
soil consisted of approximately 14,000 tons of fill
material and 2,000 tons of sediments from the
primary settling lagoon. The identified material also
contained a significant amount of environment
control (EC) dust that was landfilled onsite prior to
being listed as a hazardous waste on Nov. 18,1980.
As part of the technology evaluation to clean up the
site, Shirco performed a trial run during the spring of
1986 with its pilot-scale unit. The trial burn was
successful and infrared incineration was chosen as
the preferred technology.
OH Materials Corp. (OHM) performed a TSCA PCB-
disposal demonstration in Sept. 1987 at the site
using its 100 ton/h transportable unit. In addition to
PCBs, the site contained significant levels of heavy
metals. (The cleanup of the soil was completed under
a Superfund CERCLA 106 order with funds provided
by Florida Steel.)
OHM complied with the majority of the criteria and
standards for PCB disposal pursuant to 40 CFR 761
in the 1987 tests, but did not meet particulate
emissions standards and the removal of PCBs in the
incinerator ash to below 2 ppm. Therefore, OHM was
not granted a TSCA approval for disposal of PCBs.
OHM performed a second demonstration at the
Florida Steel site in Indiantown, Fla. in June 1988.
Local gravel transported for the trial burn was
spiked with Askarel containing low PCB concen-
tration.The following is a description of the 1987 and
1988 TSCA tests and their results.
Feed Preparation
The site contained an assortment of different
constituents, including EC dust, furnace slag, rein-
forcing bar, car bumpers, and railroad ties. The
materials-handling system consisted of many
components in order to handle this diverse waste
stream. The system included a grizzly classifier,
magnetic separator, rock crusher, weigh belt feeder,
and associated conveyance systems.
For the 1987 TSCA test, the waste feed material was
spiked with PCBs ranging in concentration from
5,000 to 20,000 ppm. For the 1988 TSCA test, local
gravel transported for the trial burn was spiked with
Askarel containing low-PCB concentration. Initial
diagnostic tests revealed problems with particulate
emissions, attributable to the high chloride content
in the waste feed. High-PCB level Askarel was
procured and used for spiking. The high-PCB
Askarel reduced the chloride level in the waste feed
at the desired PCB concentration.
Test Procedure
The 1987 demonstration included 5 runs, 2 with site
soil spiked to 5,000 ppm PCBs and 3 runs to 20,000
ppm PCBs, feeding at a rate of 6 ton/h. Because
analysis of the soil at the Florida Steel site indicated
the presence of heavy metals, stack emission samples
were analyzed for cadmium, copper, lead, and zinc.
Operating data for the five tests runs are shown in
Table D-2.1.
73
-------
r
The 1988 demonstration included 5 diagnostic tests
prior to the TSCA runs. Test parameters are shown
in Table D-2.2. Results indicated that particulate
emission is sensitive to chloride content of the feed.
At about 7 ton/h feed, chloride content above 10,000
ppm resulted in particulate emissions over 0.08
gr/dscf. With high-PCB Askarel (having less chloride
content), the 3 TSCA test runs were conducted on
June 29-30, 1988. Operating data for these test runs
are given in Table D-2.3.
Primary-chamber residence-time ranged from 15
min in the 1988 tests to 25 min in the 1987 tests. The
15-min time was performed with a bed depth of 1 in.
nominal. The two-in. bed depth required a 23 min
residence time to achieve PCB removal. Higher
temperatures—1,570°F—were required with the 15-
min time—versus 1,220°F for the higher residence
time of 23 min.
Residence time in the secondary chamber ranged
from 5.18 to 7.14 s, well above the 2-s standard for
liquid PCBs incineration.
Results
Characteristics of the Feed
Waste feedrate for the 1988 trial burn averaged
13,835 Ib/h with an average concentration of 5,600
ppm PCBs. The 1987 demonstration averaged 11,560
Ib/h feedrate with a PCB level of 2,400 ppm, and
11,920 Ib/h at 20,333 ppm PCBs.
Chloride content in the feed varied from 0.19% to
0.79% during the test runs with feedrates ranging
from 21.8 Ib/h to 93.2 Ib/h of chlorides, respectively.
In the five diagnostic tests conducted prior to the
1988 demonstration, chloride levels ranged from 0 to
19,290 ppm. Particulate emissions of 0.316 gr/dscf
resulted from operations performed at the highest
chloride content of 19,290 ppm accompanied by a
feedrate of 143 Ib/h chlorides. A chloride level of
9.645 ppm (0.96%), produced particulate emissions of
0.072 gr/dscf, below the 0.08 gr/dscf criteria. It is
believed that the chloride content of the feed should
be restricted to 0.9% with a feed rate of 133 Ib/h of
chlorides to avoid problems with particulate
emission.
With the addition of supplemental fuel to the feed,
there is a potential for the fuel to separate out from
the soil substrate. With separation, the hazards
imposed by the oil may be two-fold: (1) Oils may
extract PCBs from the soil matrix, and oil drippings
may spread PCB-contamination in the area under
the feed conveyor and hopper, as well as in areas
within the feed staging zone. (2) The fuel oil, an
ignitable fluid, may cause flame to propagate from
the primary chamber to open areas at the job site.
Tables D-2.4 and D-2.5 summarize the waste feed
characteristics for the 1987 and 1988 test runs.
Characteristics of the Furnace Ash
Furnace ash from Run 1 of the 1987 tests contained
200 ppm PCBs. OHM believes that the inadequate
removal of PCBs from the soil feed was a result of
Run 1 being the initial operation with PCBs and that
operating conditions for treatment of PCBs were not
firmly established. Furnace ash from Run 2 of the
1988 demonstration also indicated levels of PCBs,
but at a low concentration (19 ppm). OHM studied
the problem and concluded that poor handling of the
feed auxiliary fuel influenced the quality of fuel,
Tabla D-2.1. Operating Parameters for the 1987 Test Runs
Parameters
Retention time, min.
Bod depth, in.
Primary exhaust temp., °F
Primary exit (B3) gs temperature, °F
Primary chamber pressure, in. H2O
Secondary chamber temperature, °F
Secondary exhaust, °F
Residence time, s
Excess oxygon, %
Run 1
22
2
1,347
1,159
-0.05
2,098
2,021
5.58
6.9
Run 2
25
2
1,476
1,154
-0.05
2,032
1,960
5.77
7.0
Primary Chamber
Run 3
23
2
1,550
1,221
-0.05
Secondary Chamber
1,964
1,893
5.18
7.0
Run 4
23
2
1,318
1,339
-0.05
1,935
1,853
5.45
6.9
Run 5
23
2
1,442
1,265
-0.05
2,069
1,980
5.35
6.0
74
-------
Table D-2.2. Operating Parameters for the 1988 Diagnostic Runs
Parameters
Waste feedrate, Ib/h
RGB feed level, ppm
Feed total chlorides, ppm
Chloride rate, Ib/h
Particulates, gr/dscf adjusted to 7% O2
Tabie D-2.3. Operating Parameters for 1988 Test
Parameters
Retention time, min.
Bed depth, in.
Primary exhaust temp., °F
Primary chamber (A2 temp., °F
Primary exit (82) temp., °F
Primary chamber pressure, in H2O
Secondary chamber temp., °F
Secondary exhaust temp., °F
Residence time, s
Excess oxygen, %
Table D-2.4. Feed Characteristics and ORE forthe
Parameters
Waste feedrate, Ib/h
Feed PCB level, ppm (OHM GC/MS Method 680)
PCB rate, Ib/h
PCB stack emissions, ug/h
PCB ORE
Feed total chlorides, %
Feed PCB level, ppm (EPA lab PGC/ECD)
Heavy Metals in feed, ppm
Lead
Zinc
Cadmium
Run 1
16,225
0
0
0
0.31
Runs
1987 Tests
Run 1
11,470
4,570
52.4
130
99.999999
<0.19
2,600
375
2,630
8.3
Run 2
14,115 .
8,187
19,290 ,
272.3
0.316
Run 1
1
1
1,584
1,450
1,555
-0.02
1,987:
1,914
6.11
5.01
Run 2
1 1 ,660
2,810
32.8
112
99.999999
<0.26
2,400
459
4,790
9.9
Run 3 Run 4
14,835 14,125
4,094 2,456
9,645 5,787
143.1 81.7
0.072 0.057
Primary Chamber
Run 2
15
1
1 ,566
1,434
1,605
-0.01
Secondary Chamber
1,996
1,924
6.04
4.94
Run 3 Run 4
11,798 11,817
.15,100 13,900
178 164
166 14,200
99.999999 99.999998
0.79 0.68
13,000 18,000
332 434
2,280 4,450
7.4 1 1 .3
Run 5
13,925
2,456
5,787
80.6
0.057
Run 3
15
1
1,601
1,405
1,579
-0.02
1,995
1,921
6.31
4.96
Run 5
12,144
13,900
164
26,000
99.999998
0.71
30,000
352
2,430
7.9
which in turn affected the effectiveness of PCB
removal in Run 2. Consequently, OHM instituted a
procedure to preclude recurrence of such incidents.
This corrective action is proprietary to OHM. One
solution is to closely monitor the furnace ash PCB
content during operation.
PCDDs were not detected in any of the furnace ash
from all tests in 1987 and 1988. Furnace ash from the
1987 tests contained 2,3,7,8-TCDD equivalents
below 1 ppb and are not of concern. Sift-through ash,
however, contained higher levels of PCDFs, but
operations were revised to transfer the ash back to
the primary chamber via a closed loop. Therefore,
exposure to PCDFs from sift-through ash is
precluded. PCDFs in the 1988 ash were slightly
higher, but each PCDF homolog converts to 2,3,7,8-
TCDD equivalents below levels of concern.
75
-------
r
Table D-2.5. Feed Characteristics and ORE for 1988 Tests
Parameters
Run 1
Run 2
Run 3
Waste feodrate, lb.li
Food PCB level, ppm (OHM)
(Method 680, GC/MS)
PCB rate, fcft
PCS slack emissions, 10-6 Ib/h
PCB ORE
Food total chlorides, %
Chtortdo rate, Ib/h
Food PCB level, ppm (EPA lab)
(PQC/ECD for Aroclors)
13,921
96.97
8.24
99.999992
0.54
75.2
5,700
13,856
6,966 (avg. 3 runs)
96.52
5.63
99.999994
0.36
49.9
5,20
13,728
95.63
5.11
99.999995
0.38
52.2
5,900
Results of the furnace ash characteristics for the
1987 and 1988 tests are given in Tables D-2.6 and D-
2.7.
Destruction and Removal Efficiency
(ORE) of PCBs
Stack emissions of PCBs during the 1988 tests
ranged from 0.000005 Ib/h to 0.000009 Ib/h with
feedrates of PCBs of about 95 Ib/h. DRE met the
TSCA performance standard of six 9s DRE
(99.9999%) required of incinerators. For incineration
of non-liquid PCBs, the mass air emissions standard
for PCBs is 0.001 gm PCBs out/kg PCBs in,
equivalent to the six 9s DRE. All of the 1987 test
runs complied with the emissions standard. The
actual DRE values for the 1987 and 1988 runs are
shown in Tables D-2.4 and D-2.5.
D/ox/ns and Furans Stack Emissions
PCDDs were not detected in stack emissions samples
from all tests. PCDFs however, were detected at low
levels in the stack emission samples. The 1988 tests
indicated levels of PCDFs ranging from 12.5 to 25.9
ng/m3. Conversion of PCDFs to 2,3,7,8-TCDD
equivalence transforms these numbers to levels
ranging from 1.36 to 2.76 ng/m3 of 2,3,7,8-TCDD,
well under the 10 ng/m3 level of concern. Emissions
of PCDFs from the 1987 tests were all below the level
ofconcern.
Stack emission standards do not exist for dioxins and
furans. However, RCRA performance standards are
in place under 40 CFR 264.343 for hazardous waste
incinerators. The incinerators must comply with a
DRE of 99.9999% for destruction of the principal
organic hazardous constituents (POHCs) in treating
designated dioxin-containing hazardous wastes.
With PCBs as the POHC under consideration, the six
9s DRE demonstrated for PCBs by the unit would
comply with the requirements for RCRA incinerators
treating designated dioxin-containing wastes. Tables
D-2.8 and D-2.9 provide the PCDD and PCDF
emission data for the 1987 and 1988 tests.
Other Organic Stack Gas Emissions
TSCA regulations require monitoring for total
chlorinated hydrocarbons (RC1) when initially
treating PCBs. The sampling method used for
monitoring stack emissions for volatile RCls is the
volatile organic sampling train (VOST).
Hydrocarbons other than chlorinated organics were
assessed as well. OHM collected VOST samples and
provided data for the 1987 demonstration. Therefore,
EPA did not require additional VOST monitoring
during the 1988 trial burn.
Data taken during the 1987 tests on volatile and
semivolatile organic species in the stack gas
indicated no significant levels of emissions except for
one compound, bis(2-ethylhexyl)phthalate during
Run 3. Phthalate compounds are widely used in the
plastics and elastomer industry as a plasticizer.
OHM used sealants that contain phthalates to repair
a piece of the incinerator equipment. Contamination
of the stack sample likely occurred from the sealant.
Other organic emissions were not significant when
compared to the existing health-related standards
and criteria.
Volatile and semivolatile organic compounds at
concentrations less than established standards for
direct inhalation included:
• Chlorinated methane, methylene, and other
halomethanes.
• Aromatic volatiles and semivolatiles such as
benzene, ethylbenzene, styrene, toluene, naph-
thalene, fluoranthene, phenanthrene, anthracene,
pyrene, and other benzene-related compounds.
76
-------
Table D-2.6. Furnace Ash Analysis
Parameters
PCBs in ash, ppm (OHM)
(Method 680 GC/MS)
PCBs in ash, ppm (EPA lab)
(PGC/ECD for Aroclors)
PCDDs in ash, ppb
PCDFs in ash, total, ppb
2,3,7,8-TCDD
TCDDs
PCBs in sift-through ash, ppm
(OHM & EPA labs)
PCDDs in sift-through ash, ppb
PCDFs in sift-through ash, ppb
2,3,7,8-TCDF
TCDFs
Total PCDFs
2,3,7,8-TCDD equivalent
Heavy Metals in ash, ppm
Total metals
Lead
Zinc
Cadmium
EP Tox results
Lead
Zinc
Cadmium
for the 1987 Tests
Run 1
230
200
ND
1.14
0.27
0.87
-
-
—
—
—
—
772
8,940
10.9
<0.1
<0.1
<0.1
Run 2
0.265
<2
ND
0.19
0.07
0.12
-
ND
2.00
6.6
9.06
0.702
489
5,260
9.0
<0.1
<0.1
<0.1
Run 3
ND
<2
ND
0.15
<0.04
<0.04
—
ND
3.70
14.1
23.9
1.488
494
6,240
6.0
<0.1
6.68
<0.1
Run 4
ND
<2
ND
<0.71
0.06
0.06
<2
ND
1.50
6.2
8.06
0.626
447
5,140
7.2
<0.1
6.60
<0.1
Run 5
ND
<2
ND
<0.50
<0.03
<0.03
<2
-
--
—
—
—
510
6,120
7.0
<0.1
<0.1
<0.1
ND = Not detectable
Table 0-2.7. Furnace Ash Analysis for the 1988 Tests
Parameters
PCBs in ash, ppm (OHM)
(Method 680 GC/MS)
PCBs in ash, ppm (EPA lab)
(PGC/ECD for Aroclors)
PCDDs in ash, ppb
PCDFs in ash, total, ppb
2,3,7,8-TCDF
TCDFs
2,3,7,8-TCDD equivalent
Run 1
0.946
<2
ND
3.32
0.17
0.41
0.179
Run 2
33.61
19
ND
19.36
0.825
3.10
1.09
Run 3
1.607
<2
ND
2.54
0.13
0.25
0.165
• Oxygenated hydrocarbons including phthalates,
phenol, and oxygenated benzene-related
compounds.
HCI, A/Ox and RCI Emissions
Hydrogen chloride (HCI) emissions ranged from 0.07
to 0.22 lb/h, below the TSCA criteria of 4 Ib/h for
HCI. HCI removal efficiency was greater than 99%.
NOX emissions ranged from 0.86 to 2.53 lb/h, or with
the thermal load of the secondary combustor of about
14 MBtu/h, ranged from 0.0614 to 0.181 Ib/MBtu.
This compares favorably with the 0.2 Ib/MBtu NOX
standards for steam-generating boiler units of 250-
MBtus or more, for gaseous fuels at 40 CFR 60.40,
and with standards for solid (0.50 Ib/MBtu) or liquid
fuels (0.40 Ib/MBtu).
Total chlorinated organics (RCI) were not obtained
for the 1988 test; however, the 1987 tests indicated
total RCls to range from 0.00133 to 0.015 mg/m3. The
Volatile Organic Sampling Train (VOST) was used
in sampling for RCls. In addition, the MM5 samples
(semi-VOST) were analyzed for RCls, but only traces
of 1,2,4-trichlorobenzene were detected. The highest
chlorinated hydrocarbon detected was methylene
chloride at 0.0102 mg/m3 concentration. To establish
a perspective for levels of RCI emissions, OSHA
PELs (Occupational Safety- and Health Agency,
permissible emission level) for methylene chloride
77
-------
and chloroform are 1,714 and 240 mg/m3,
respectively. The highest total RC1 was 0.015 mg/m3.
Partlculate Emissions
Particulate emission from both TSCA
demonstrations ranged from 0.02 to 0.10 gr/dscf
(adjusted for 7% oxygen), as compared to the TSCA
criteria of 0.080 gr/dscf. Particulate emissions of 0.10
gr/dscf from Run 3 of the 1987 trial burn and 0.316
gr/dscf from Run 2 of the 1988 diagnostic runs were
above the criteria. However, all test runs during the
1988 demonstration complied with the criteria.
Tables D-2.2, D-2.8, and D-2.9 provide particulate
emission data for the 1987 and 1988 tests. Factors
that contributed to the high particulate emissions
include chloride feedrate and scrubber operating
technique. For example, the 0.10 gr/dscf particulate
emission in Run 3 (1987) was caused by a scrubber
malfunction, and the 0.316 gr/dscf particulate
emission in Run 2 (1988) occurred at a chloride
concentration of 19,290 ppm. Chloride feedrate is
dependent on the waste feedrate and chloride
concentration. All three factors — chloride feedrate,
chloride content, and waste feedrate — influence
particulate emission and need to be monitored.
Scrubber operating techniques—also critical to
controlling pollutant emissions (primarily
particulates), must be maintained carefully. This
information is, however, proprietary to OHM.
Combustion Efficiency
The TSCA performance standard for PCB
incinerators of 99.9% for combustion efficiency (CE)
is related to the ratio of carbon monoxide, to carbon
dioxide. Based on the data presented in Tables D-2.8
and D-2.9, the unit met the TSCA standard for CE
during the 1987 and 1988 tests.
Mobility of Heavy Metals
Data presented for the 1987 tests in Tables D-2.4 and
D-2.6 show that the concentrations of metals in the
waste feed are similar to the concentrations in the
furnace ash and indicate that the mass flow of these
species remain with the high mass flow of furnace
ash. The EP Tox data on the furnace ash is below the
toxicity characteristic standards; there is no
comparative data on the waste feed to determine
whether the thermal treatment reduced the mobility
of heavy metals based on the results.
Scrubber Water
Scrubber water for the 1988 test contained less than
1 ppb PCBs. All aqueous waste generated during
PCB disposal activities must be below 3 ppb to be
classed as non-TSCA regulated waste. The 1987 trial
burns also resulted in scrubber solutions with PCBs
concentration less than 1 ppb. OHM complied with
Florida regulations, which includes discharge limits
of 0.001 mg/L PCBs, 0.2 mg/L lead, and 0.04 mg/L
cadmium. The EPA lab analysis was performed at a
detection limit of 2 ppm PCBs because the 3-ppb
criteria for PCBs had not been established during the
1987 demonstration.
No PCDDs or PCDFs were detected in the scrubber
water.
Scrubber water analysis data for the 1987 and 1988
tests is presented in Tables D-2.10 and D-2.11.
Operations
The OHM infrared incinerator operated without any
major problems during the 1987 and 1988 TSCA test
runs. Interruptions in the tests occurred when
problems developed with equipment breakdown and
when modifications in stack sampling equipment
were necessary to comply with EPA protocols. The
1988 tests went smoothly because OHM had seven
months of experience in treating the Florida Steel
site.
Several operating procedures and other factors were
identified based on these tests to meet TSCA
performance requirements (such as PCB in furnace
ash and particulate emission). These have been
described earlier while discussing the results of the 2
test runs.
Other operations related problems are discussed in
Appendix D-3, where information on treatment of
18,000 tons of soil at the Florida Steel site by OHM is
presented.
Several modifications including proprietary changes
to the unit were made between the 1987 and 1988
tests. These included:
1. Ash collection system, including the sift-through
ash from the feed conveyor.
2. Ash quench method.
3. Scrubber blowdown practice.
4. Feed-hopper feeding mechanism.
5. Air compressor replaced with one of higher
capacity.
78
-------
Table D-2.8. Stack Emissions Data for the
Parameters
PCBs, ng/m3
PCDDs total, ng/m3
PCDFs total, ng/m3
2,3,7,8-TCDF
Total TCDFs
HCI emission, Ib/h
HCI removal, %
Participates, gr/dscf @ 7% O2
Oxygen, %
Carbon monoxide, ppm
Carbon dioxide, %
Combustion efficiency, %
Heavy metals, mg/m3
Lead
Zinc
Cadmium
Copper
Table D-2.9. Stack Emissions Data for the
Parameters
PCBs, ng/m3
PCDDs total, ng/m3
PCDFs total, ng/m3
2,3,7,8-TCDF
Total TCDFs
2,3,7,8-TCDD equivalent
HCI emission, Ib/h
HCI removal, %
Participates, gr/dscf @ 7% O2
Oxygen, %
Carbon monoxide, ppm
Carbon dioxide, %
Combustion efficiency, %
1987 Tests
Run 1
<9.5
ND
0.86
0.54
0.86
<0.08
>99.6
0.02
11.6
3.0
5.2
99.99
2.38
3.51
0.517
0.39
1988 Tests
Run 2
<77.3
ND
1.70
0.83
1.70
<0.07
>99.8
0.03
11.7
5.0
6.3
99.99
5.74
3.82
0.89
0.598
Run 1
<360
ND*
17.2
<5
13.8
<2.19
0.088
99.88
0.053
13.5
1.2
4.8
99.99
Run 3
<772
ND
ND
ND
ND
<0.22
99.8
0.10
13.4
1.0
5.2
99.99
13.5
20.9
1.34
1.04
Run 4
<1315
ND
7.42
1.97
7.42
<0.16
99.8
0.07
13.54
1.0
5.2
99.99
10.7
13.2
1.07
0.859
Run 2
<242
ND
12.5
<5
10.0
<1.36
0.076
99.85
0.061
13.6
4.3
4.8
99.99
Run 5
< 578.4
ND
2.10
1.01
2.10
<0.12
99.9
0.08
12.7
1.0
5.2
99.99
1 1 .7
15.1
1.34
1.02
Run 3
<232
ND
25.9
<5
18.4
<2.76
0.115
99.78
0.056
13.4
1.1
5.0
99.99
:ND—<5.0 ng/m3/homolog det. limit
79
-------
r
Table D-2.10. Scrubber Water Data for the
Parameters
PCS in scrubber water, ppb
PCOD in scrubber water, ppb
PCOF in scrubber waer, ppb
Scrubber pH, avg.
Vonturi pressure drop
Scrubber water components, mg/L
Dissolved solids
Load
Zinc
Cadmium
Copper
Table D-2.11. Scrubber Water Data for the
Parameters
PCS in scrubber water, ppb CI3 detected only
PCDO in scrubber water, ppb
PCDF in scrubber water, ppb
Scrubber pH. avg.
Vonturi pressure drop, in H2O
1987 Tests
Run 1
ND
ND
ND
7.48
24.4
5,118
7.96
77.3
1.57
1.88
1988 Tests
Run 2
ND
ND
ND
7.91
16.8
5,520
7.73
86.3
2.12
2.66
Run 1
0.21
ND
ND
6.75
26.7
Run 3 Run 4
ND ND
ND ND
ND ND
8.46 8.55
25.9 25.0
7,920 7,670
9.73 1 1 .9
75.6 75.1
0.97 1.18
1.65 1.49
Run 2
0.02
ND
ND
7.06
26.1
Run 5
ND
ND
ND
8.24
27.0
6,170
9.92
51.9
0.892
1.48
Run 3
0.146
ND
ND
7.14
26.4
80
-------
APPENDIX D-3
FLORIDA STEEL COMMERCIAL CLEANUP [5-7]
Introduction
Florida Steel Corp. operated a metal recycling plant
in Indiantown, FL, which was closed in 1982 for
economic reasons. An environmental audit of the
plant site revealed contamination with PCBs that
had leaked from hydraulic equipment used there.
A complete site evaluation was completed in 1984,
which included more than 500 soil borings and 1,400
samples. The PCB-contaminated materials consisted
of approximately 14,000 tons of fill material and
2,000 tons of sediments from the primary settling
lagoon. The identified material also contained a
significant amount of environmental control (EC)
dust which was landfilled onsite prior to being listed
as a hazardous waste on Nov. 18,1980.
An above-ground vault was constructed by O.H.
Materials (OHM) during the summer of 1985 to store
and isolate the material from the environment while
disposal alternatives were evaluated. The
alternatives were evaluated during the spring of
1986, at which time Shirco performed a trial burn
with their 80-lb/h portable unit. The trial burn was
successful and infrared incineration was chosen as
the preferred technology.
Site work began in Mar. 1987 with the preparation of
the work areas for the waste storage, ash storage,
and incinerator. The site had 2 large buildings under
which all the equipment was installed (with the
exception of the water treatment system). A new
impervious concrete floor was poured in all work
areas. Also during this time, an inflatable building
(150 x 300 ft) was installed over the vault to control
dust and moisture while the waste material was
being removed.
The incinerator arrived on site in August and was
mechanically installed in 11 days. An initial series of
mechanical and electrical checkouts were performed
over the next 2 weeks. The commercial burn began in
Oct. 1987 and lasted until May 1988, during which
time 18,000 tons of PCB-contaminated soils and
sediments were burned.
OHM used a 100-ton/h incinerator purchased from
Shirco. Several modifications were incorporated into
the unit to improve its mobility/adaptability to
onsite service and safety. The largest modification
eliminated the 52-ft-tall insulated emergency stack.
The stack was eliminated for several reasons,
including problems with erecting the stack and the
potential of releasing unscrubbed gases into the
atmosphere. A new emergency backup system was
installed to include a direct-drive induced-draft fan
and scrubber pump. The emergency backup system is
activated by a power failure or the loss of the
primary ID fan.
Feed Preparation
The waste material contained an assortment of
different constituents,including EC dust, furnace
slag, reinforcing bar, car bumpers, and railroad ties.
The materials-handling system consisted of many
components to handle this diverse waste stream. The
different systems used were:
• Track Hoe
• Grizzly classifier
• Magnetic separator
• Jaw crusher
• Roll crusher
• Front-end loader
• Pubmill
• Plastic shredder
• Wood chipper
After the waste was prepared for incineration, it was
sampled for PCB and moisture analysis and
stockpiled. The waste was then fed into the feed
hopper by a front-end loader.
81
-------
Test Procedure and Results
Test procedures used in the treatment were similar
to the ones used by OHM in TSCA demonstrations of
the incinerator in Sept. 1987, and again in June
1988. These TSCA demonstrations (including
procedures and test results) are described in
Appendices D-l and D-2. No specific performance
data for the commercial remediation of the 18,000
tons of contaminated soil by OHM is available.
Operations
A problem was discovered during the checkout with
regard to the wire woven belt. The actual belt that
was installed in the full-scale unit had thicker gauge
wire than the pilot unit and thus had a larger pore
space, which allowed the fine Florida sand to sieve
through the belt. The amount of material that sieved
through the belt was greater than that which the
sieve-through collection-system could handle. This
problem was solved by installing a smaller- gauge
belt.
During the first several months of the project, many
problems were encountered that directly affected the
incinerator utilization. The utilization is a measure
of the time that waste was actually fed into the
incinerator. The equipment utilization for Oct. was
50%. The utilization increased to more than 90% for
the final month, which resulted in an overall project
utilization of 61%.
Economics
OHM performed the complete site soil treatment
(18,000 tons) at an average cost of approximately
$300/ton. The OHM scope of work included:
• Excavation of site
• Transport of processed ash to a building
• Construction of vault
• Water treatment
• Dewatering of waste ponds
• Vendor profit
• Unit mobilization and waste processing
The cost does not include ash disposal.
82
-------
APPENDIX D-4
LASALLE ELECTRIC COMMERCIAL CLEANUP [8-12]
Introduction
The LaSalle Electric Utilities (LEU) site is located in
the city of LaSalle in north-central Illinois.
LEU is a former manufacturer of electrical
equipment. Operations at the plant began prior to
World War II, and in the late 1940s the plant began
using polychlorinated biphenyls (PCBs) in the
production of capacitors. This manufacturing
practice continued until Oct. 1978. In May 1981, the
company ceased operations at the LaSalle plant after
it was ordered to do so by the Illinois Attorney
General and the Illinois Environmental Protection
Agency (IEPA). The LaSalle facility has been
abandoned since that time.
Information is limited on the waste management
practices of the company both on and off the
property. Undocumented reports allege that PCB-
contaminated waste oils were regularly applied as a
dust suppressant both on and off the property as late
as 1969. Following the regulation of PCBs, manifests
document the disposal of PCBs at all regulated
facilities.
Concentrations of PCBs in the composite soil
samples from soil on the LEU source area range from
0.20 ppm (lower detection limit) to as high as 17,000
ppm. PCB concentrations in composite soil samples
from nearby offsite areas range from 0.20 ppm to as
high as 2,600 ppm. The Remedial Investigation of
the area characterized soil off the plant property to
the degree necessary to implement remediation in
this area. The need for remediation in the areas
surrounding the plant (specifically defined as the
area outside of the security fence) was viewed as
significant, due to direct exposure to PCB at
residences and businesses.
Westinghouse/Ha/tech, Inc. (WHI) was contracted to
perform onsite remediation of the LEU offsite area
using their transportable 100 ton/d Shirco unit. The
remediation requirements were to clean up the area
to a PCB concentration of 5 ppm to a 12-in. site
depth, and 10 ppm below 12 in. The soil which is
removed was to be replaced with clean soil.
Prior to beginning the remediation work, WHI
performed a Demonstration Burn using their unit to
satisfy substantive requirements of the IEPA and
TSCA for the incineration of PCBs. The test
program used a waste feed mixture that had been
spiked with PCB material to extend the permitted
range of operating conditions for the unit. Based on
the results of these tests, IEPA issued a conditional
approval to WHI for treatment of the LEU offsite
soil. Among other things, the approval limited
incineration of soil with PCB concentrations no
greater than 50 ppm. Specific details of the
operating permit that define the regulatory
constraints on the operation of the unit are presented
in the Operations section of this discussion.
WHI began incineration of the soil in late 1988 and
expects to complete processing approximately 24,000
tons of soil by the end of 1989. WHI's bid price for the
thermal incineration work is approximately
$300/ton, whichincludes permits, utilities, sampling
and analysis and vendor profit but does not include
waste excavation, feed preparation and ash disposal.
In their first three months of operation, WHI is
experiencing unit availability in the 80 to 90%
range.
Feed Preparation
Excavated soil is stockpiled in a bermed plastic-lined
containment area adjacent to the unit. The area is
covered with a tarpaulin to minimize weather
effects, runoff, and increased moisture content.
Excavated soil is then transferred to a covered feed
83
-------
makeup area where a chopping/screening operation
is used to produce a feed stream with no individual
piece larger than 1 in. This power screen consists of
a hopper (into which a loader drops the soil), a set of
knives and two vibrating screens. The rotating
knives break up dirt clods and any other soft
material before the material reaches the screens.
Two vibrating screens are used to segregate large
particles (larger than 1 in.) from smaller ones. The
large particles are then either reprocessed,
decontaminated, or handled in another piece of
equipment, whichever is appropriate.
This waste preparation equipment is varied
according to the type of waste requiring
decontamination. For example, if large pieces of
wood are in the waste, a wood chipper is used.
Processed waste is placed in a stockpile and is moved
to the weigh hopper using a loader.
Waste is dumped into the weigh hopper (14-ton
capacity) by a loader until the hopper is filled to the
desired level. At that time, feed to the weigh hopper
will be stopped and waste will be conveyed from the
weigh hopper to the unit's feed hopper via a totally
enclosed belt conveyor. The WHI unit is also covered
by a temporary building structure.
Prior to dropping into the feed hopper, the waste
drops into an enclosed screw conveyor where fuel oil
is mixed with the waste to control fugitive dust in the
feed hopper and to increase the heating value of the
waste. The feed hopper has a live bottom consisting
of six 9-in. screws, which convey the waste forward to
an enclosed opening on the top of the incinerator.
The waste falls through the opening and forms a
layer across the width of the belt.
One concern about the power screen operation
pertains to passing of long slender items such as
nails and railroad spikes, which cause problems in
the processing downstream.
Characteristics of the Soil
The PCB-contaminated soil from the LEU offsite
areas was excavated to a 12- in. depth hi most areas,
and to larger depths (up to 5 ft) in other areas. The
criterion was to excavate to a depth of 12 in. based on
a cleanup level of 5 ppm, and below that depth as
required to achieve a cleanup level of 10 ppm.
The characteristics of the soil in the area vary from
location-to- location. At most locations, the soil
consists of silt and sand occasionally interbased with
clay. Top soil is also present in many areas.
The contaminants of concern in the LEU offsite area
are PCBs. No other materials above normal
background levels have been detected in this area.
Concentrations of PCBs in the composite soil
samples from this area range from less than 0.20
ppm to as high as 2,600 ppm. (The lower limit is the
analytical detection limit.) Additional grab samples
(from the most heavily contaminated residential
yard) revealed a hot spot containing up to 5,800 ppm
of PCBs. Concentrations typically average about 75
to 125 ppm in most yards in the area. The depths of
contamination range from 0 to 12 in. in most areas,
to as much as 5 ft at a few heavily contaminated
locations. The total volume of soil that is contam-
inated above the 5-ppm level is approximately
28,690 yd3.
Soil sampling for dioxins and furans did not detect
any tetra dioxins, including 2,3,7,8-techlorodibenzo-
p-dioxin (TCDD). However, penta, hexa, hepta, and
octa dioxin isomers as well as tetra, penta, hexa,
hepta, and octa furan isomers were detected in PCB-
contaminated areas at ppb concentrations.
PCB, dioxin, and furan data were submitted to the
Agency for Toxic Substances and Disease Registry
(ATSDR) to evaluate the degree of health concern
and the possible need for an immediate removal of
contaminated material. The resulting evaluations
indicated that detected concentrations are below
levels of concern for human health.
Sampling for additional organic contaminants
resulted in identification of polychlorinated
naphthalenes, aliphatic hydrocarbons, and poly-
nuclear aromatic compounds (anthracene, fluor-
anthene, and pyrene) directly west of factory
buildings. None of these compounds were found in
concentrations exceeding 3 ppm. Diethylhexyl-
phthalate was identified in 5 samples from this area
at a maximum concentration of 20 ppm; it has been
used as a replacement for PCBs as dieletric additive,
which may account for its presence at LEU.
Operations
Based on the Demonstration Burn discussed above,
an operating permit was issued to WHI on Nov. 23,
1988. The fully-approved commercial cleanup began
on Nov. 29, 1988, as defined by the following permit
conditions:
Findings
1. Particulate emission limit of 0.08 gr/dscf
corrected to 12% carbon dioxide.
2. Carbon monoxide limit of 500 ppm corrected to
50% excess air.
84
-------
3. Those soils found with PCBs concentration
greater than 50 ppm shall not be incinerated.
4. The hydrogen chloride (HC1) emissions shall not
be allowed to exceed 4 Ib/h.
5. The nitrogen oxides (NOx) emissions shall not be
allowed to exceed 100 ppm.
6. The sulfur content of the fuel oil additive for feed
Btu enhancement shall not exceed 0.5 wt%.
Conditions of Approval
The incinerator shall be operated in compliance with
the following conditions:
1. Soils to be treated shall be sampled with a daily
composite and analyzed for the following
parameters for the first two weeks of operation:
a. PCBs concentration in the soil
b. Soil moisture content
If any of the daily composite samples are found to
have a concentration of PCBs greater than 50
ppm, the daily composite and analysis shall be
continued for an additional two weeks. If no daily
composite samples during the two-week period
are greater than 50 ppm, then weekly composite
sampling and analysis may be conducted.
If any of the weekly composite samples are found
to exceed 50 ppm of PCBs, then daily composite
sampling and analysis shall be performed until
10 consecutive daily samples are found to be less
than 50 ppm, at which time weekly composite
sampling may then resume.
2. The secondary combustion chamber residence
time for combustion gases shall be maintained
greater than 2 seconds.
3. The carbon monoxide emissions shall be
maintained less than the following limits:
a. 500 ppm at any time
b. 100 ppm for 3 minutes
4. Combustion efficiency (CE) shall be maintained
greater than 99.9% efficient and shall be
calculated and recorded at 15-min intervals
based on the following formula related to
measure carbon monoxide and carbon dioxide.
CE = 100 x (CO2)/(CO2 + CO)
5. The incinerator and air-pollution-control
equipment shall be operated in a manner similar
to operation during the stack testing. The
incinerator shall be operated to maintain
operating parameters within the following
parameter ranges:
a. The soil feedrate into weigh hopper: 6.0 ton/h
not to exceed 12 tons in 2 h
b. The feed screw rate: 70% or less
c. The soil residence time: 15minor greater
d. The fuel oil addition: 0 to 600 Ib/h
e. Primary Zone A: at 1,200°F or greater
f. All other primary zones: at 1,400°F or greater
g. The secondary combustion chamber midpoint
temperature: 1,920°F or greater, but not to
fall below 1,820°F
h. The monitored oxygen concentration:
maintained at 4% or greater
i. The packed tower scrubbant pH:, 6.0 or
greater
j. The demister pressure drop: 3.0 in. WC or less
k. The outlet quench temperature: maintained
less than 212°F
1. The quencher blowdown: 30 gpm or greater
Waste Feed Cutoff Conditions
The incinerator's programmable controller shall be
set to stop the contaminated-soil feed augers when
the incinerator is operating outside the following
parameter limits:
1. All primary combustion chamber zones except Al
shall be maintained greater than 1,400°F.
2. The PCC static pressure shall be maintained at
0.01 in. WC or less with 5 s delay.
3. The SCC midpoint temperature shall be
maintained greater than 1,820°F.
4. The oxygen concentration leaving the SCC shall
be maintained greater than 3.0%, with a 3-min
delay.
5. The carbon monoxide concentration shall not
exceed the following:
a. 400 ppm, with a 30 s delay
b. 100 ppm for 3 min
6. The Calvert scrubber pressure-drop shall be
maintained greater than 30 in. WC with a 5-min
delay.
85
-------
7. The stack temperature shall be maintained less
than 200oF, with a 5-min delay.
Emission Rates
Based on the average of the 4 emission tests and the
allowable rates established in this approval, the
emissions from the treatment of soil are expected to
be as follows:
Actual
M\
Atoved
fb/1i
ton/yr
Particulates
0.40
1.81
7.93
Contaminant Emissions
CO THC NOV
0.02 0.01 1.57
0.16 0.09 2.54
0.70 0.39 11.1
SO,
6.27
27.50
HCI
0.11
4.0
17.5
86
-------
APPENDIX D-5
TWIN CITIES PILOT-SCALE TESTS [13]
Introduction
Contracted by both the Federal Cartridge Co. and
Honeywell, Shirco Infrared Systems (now ECOVA
Corp.) performed demonstration tests with the pilot-
scale Shirco unit at the Twin Cities Army Ammuni-
tion Plant (TCAAP) located in New Brighton, MN.
The purpose of these tests was to demonstrate the
Shirco technology capability relative to the decon-
tamination of polychlorinated biphenyl (PCB) laden
soils. The data obtained from the testing was subse-
quently used to evaluate site-remediation tech-
niques. In addition, triplicate sampling and analysis
data obtained from a PCB-spiked soil and performed
in accordance with TSCA guidelines was used for op-
erating permit approval. During the entire PCB-
spiked-soil incineration portion of the test program,
representatives of the USEPA Office of Toxic Sub-
stances and Midwest Research Institute monitored
the operation.
ECOVA first demonstrated the pilot-scale unit dur-
ing the week of Jan. 20, 1987. High particulate emis-
sions, failure of the Continuous Emissions Monitor
(GEM) and lack of feed shutdown interlocks negated
those trials. A second demonstration test was per-
formed at the TCAAP on May 27, 1987. This in-
volved completion of 3 consecutive successful runs,
each of 2 h duration.
In the May 1987 tests, the pilot-scale unit satisfacto-
rily demonstrated the ability to meet the prescribed
non-liquid PCB incineration TSCA performance
standard of 0.001 g PCB stack emissions per kg PCB
introduced into the incinerator. This standard repre-
sents a 99.9999% destruction and removal efficiency.
In addition, the pilot-scale unit destroyed PCBs in
soil to a level below 2 ppm per resolvable gas chroma-
tographic peak.
Feed Preparation
Site and test material preparation was performed by
Federal Cartridge Corp. personnel. The soil was ob-
tained from the Federal Cartridge Corp. and Honey-
well sites for the Jan. 1987 tests and for the May
1987 tests, from the Federal Cartridge Corp. site. In
each case, the soil was spiked with PCBs. For the
Jan. 1987 tests, the PCB content ranged from 48 to
28,000 ppm, whereas PCB concentration was ap-
proximately 45,000 ppm for the May 1987 tests. The
waste feed was manually introduced into a feed hop-
per onto a flighted metering conveyor located at the
end of the furnace The metering belt is synchronized
with the furnace belt to control the material fee-
drate. The feed hopper is mounted above the furnace
belt. There is an adjustable guillotine-type gate at
the discharge end of the metering section. This gate
assures that an amount of material no greater than
that which can pass through a preset slot size can en-
ter the furnace. The slot size is adjusted by the height
of the gate above the conveyor belt and was set at 1
1/2 in. for the tests.
Test Procedure
A total of 7 tests were performed in Jan. 1987. Table
D-5.1 presents the operating data for these tests.
The TSCA demonstration trial burns conducted on
May 27, 1987 were triplicate 2-hour tests at planned
feedrates of 100 Ib/h of site soil. During the first test,
the M5 sampling train indicated potential problems
with particulate emissions. The source of the prob-
lem was traced to the high solid feedrate. Therefore,
the feedrate was reduced about 10%, and a fourth
test was planned, repeating only the M5 sampling
87
-------
Tab!o D-5.1. Operational Data for the January 1987 Tests
Tost
No
1
2
3
4
5
6
7
Primary
Chamber
Temp., °F
1,600
1,600
1,600
1.600
1,600
1.600
1,600
Residence Time
Primary Chamber
min
15
15
15
15
15
15
15
Secondary
Combustor
Temp., °F
2,175
2,175
2,200
2,200
2,200
2,150
2,200
Residence Time
Secondary
Chamber s
2.04
2.14
1.73
1.82
1.89
1.85
2.12
Solids
Feedrate
Ib/h
95.8
84.0
83.2
83.2
129.0
25.0
106.0
PCB Level
Feed
ppm
48
1,070
28,000
18,000
24,000
8,600
418
PCB
Feedrate
Ib/h
0.002
0.041
1.058
0.680
1.406
0.491
0.418
procedure. Table D-5.2 provides the operating pa-
rameters for these tests.
Discharges from the pilot-scale unit included stack
emissions, furnace ash, and scrubber water. ECOVA
collected and analyzed samples from the stack, as
well as solids and liquids generated. Split samples of
the solids and liquids were collected by EPA for ana-
lysis.
Results
A summary of the Jan. and May 1987 test results is
presented in Table D-5.3. The following discussion
presents summary results and conclusions from the
tests.
Characteristics of the Feed
During each soil test run, a representative time-
averaged feed soil sample was obtained using a grab
sample technique. An approximate 100-mL grab
sample was collected at the feed hopper at 30-min in-
tervals throughout each run. The samples were com-
posited in a specially cleaned 1-L amber glass jar
with a Teflon-lined cap. Table D-5.3 summarizes the
PCB concentrations in the feed soil for the Jan. and
May 1987 tests.
Characteristics of the Furnace Ash
During each test run, a representative time-
averaged furnace ash sample was obtained using a
grab sample technique. The pilot-scale unit is
equipped with an ash sampling drawer located di-
rectly above the ash discharge chute. A portion of the
furnace ash that drops off the incinerator conveyor
belt into the ash discharge hopper is captured in the
sampling drawer. The Jan. 1987 tests showed less
than 1 ppm PCB in the furnace ash for Tests 1, 2, and
6, and PCB levels of 0.003, 0.002, 0.0003, and 0.005
ppm for Tests 3, 4, 5, and 7, respectively. Furnace-
ash PCB concentrations for the May Tests 1, 2, and 3
were 0.048, 0.017, and 0.038 ppm, respectively.
These values are well below the 2-ppm TSCA guid-
ance level.
Results of the furnace ash analysis from the Jan.
1987 trials indicated detectable levels of dioxins and
furans. This data is shown in Table D-5.4. Samples
for dioxin and furan analysis in the furnace ash were
Tabta D-S.2. Operational Data for the May1987 Tests
Primary Residence Time
Tost Chamber Primary Chamber
No Temp., °F min
1 1,600
2 1,600
3 1,600
4 1,600
15
15
15
15
Secondary
Combustor
Temp., °F
2,200
2,200
2,200
2,200
Residence Time
Secondary
Chamber s
2.42
2.53
2.50
2.50
Solids
Feedrate
Ib/h
94.0
78.2
76.0
76.0
PCB Level
Feed
ppm
45,000
45,000
45,000
45,000
PCB
Feedrate
Ib/h
4.22
3.52
3.26
3.26
88
-------
composited for the May 1987 tests and are presented
in Table D-5.5.
Destruction and Removal Efficiency
(ORE) of PCBs
Table D-5.6 presents the PCB ORE data for the Jan.
tests. For incineration of nonliquid PCBs, the TSCA
performance standard for PCBs is 0.001 g PCBs
out/kg PCBs in. Tests 6 and 7 failed to comply with
the emissions standard.
The DREs for all PCB-contaminated soil tests
performed in January were in excess of 99.9999%,
with two notable exceptions. The DRE for the
Federal Cartridge Co. native soil test performed on
Jan. 21, 1987, was calculated at 99.9994%. ECOVA
suggests that sample contamination may be the
reason for this calculation result. ECOVA believes
that the actual DRE was above 99.9999%, since
wastes spiked with 1,000 and 30,000 ppm PCBs on
that and subsequent days resulted in DREs greater
than 99.9999%. The processing of the Honeywell
waste resulted in analyzed PCB DREs of 99.9998%
and 99.9970% for the waste feeds containing 8,700
ppm and 418 ppm PCBs, respectively. A review of the
operating conditions and history suggests no reason
why this lower DRE should have occurred. A
comparison with feedrates, stack flows, and sample
volumes finds that these parameters were
essentially the same as for all other exposures.
However, a review of the laboratory data finds that
the total number of nanograms of PCB caught during
the 8,600 and 418 ppm PCB process operations was
12,939 and 16,208, respectively. This compares to
between 80 and 623 for all the previous five sampling
tests. Thus, there is not an obvious process operation
explanation for the lower DRE and the explanation
may be found in the sampling or analysis procedures.
For the May tests, soil feed to the unit was planned
at a rate of 100 Ib/h. Midway into Test 1, ECOVA
personnel observed that the desired feedrate was ex-
cessive for the particular feed stream. Process condi-
tions were difficult to stabilize. The feedrate was low-
ered to about 90 Ib/h and operations continued. PCB
flowrates and emissions are presented along with ap-
propriate destruction and removal efficiency (DREs)
in Table D-5.7. All test runs complied with the 0.001
gm PCBs out/kg PCBs TSCA standard.
The DREs for all 3 test runs were significantly great-
er than 99.9999%. The PCB emission rate for the
first test run was somewhat higher than for the fol-
lowing 2 runs. This may also be attributable to the
higher feedrate used during the first half of this run.
Other Organic Stack Gas Emissions
Dioxins and furans concentrations in the stack emis-
sions for the Jan. tests are presented in Table D-5.8,
and for the May tests in Table D-5.9.
PCDDs and PCDFs were not detected in the stack
emission samples taken during the May tests.
Sampling for the total chlorinated organics was per-
formed during the Jan. testing, but was omitted dur-
ing the May testing. Table D-5.10 presents a sum-
mary of the RC1 analysis performed on the Jan. sam-
ples. Because of the equivalent magnitude of PCB
concentration for the two test periods, these results
should be very indicative of what was present during
the May tests. In summary, the RC1 levels observed
are extremely small indicating efficient destruction.
Acid Gas Removal
Tables D-5.10 and D-5.11 present the HC1 concentra-
tions in the stack gases for the Jan. and May 1987
tests. As shown, the HC1 concentrations observed in
the Jan. 1987 tests ranged from 0.00026 to 0.0094
Ib/h and from 0.014 to 0.022 Ib/h for the May tests.
These concentrations are significantly below the
RCRA performance standard of 4 Ib/h. The HC1 re-
moval efficiency in all tests was in excess of 99%.
Paniculate Emissions
Tables D-5.10 and D-5.11 present the particulate
emissions data for the Jan. and May 1987 tests. In
Jan. 1987, particulate emissions for four soil-process
tests (the Federal-Cartridge nominal-48 and 2,070
ppm PCB, and the Honeywell nominal-8,700 and
418-ppm-PCB) were below the RCRA standard of
0.08 gr/dscf. However, for the nominal-30,000 ppm-
PCB triplicate-emissions-sampling tests, the parti-
culate emissions (corrected to 7% oxygen) were ei-
ther above or close to the limit. This was due to a
plugging of the wet-gas-scrubber venturi and tower-
scrubbing-liquid nozzles, which greatly reduced the
efficiency of the scrubbing process. The plugging
originated from corrosion scale on the walls of the
piping. This scale subsequently released from the
walls and collected at the nozzles. Leaks in the pip-
ing, found at the conclusion of the three tests,
prompted a replacement of the metal piping with
Qest plastic piping prior to the Honeywell soil test-
ing. The particulate emissions decreased after the
piping change.
89
-------
r
TablQ D-S.3. Demonstration Test Results Summary
Date
Test
1
1/21/87
Test
2
1/21/87
Test
3
1/22/87
Test
4
1/22/87
Test
5
1/22/87
Test
6
1/23/87
Operating parameters:
Waste teedrate, kg/h 43.49 38.14 37.77 37.77 58.57 56.75
PCS concentration, g/kg 0.048 1.076 28 18 24 8-6
PCB feedrate, kg/h 0.002 0.041 1-058 °-680 1-406 °-491
Auxiliary fuoi feedrate, kg/h 4.56 3.43 5-39 3-79 5-18 5-46
Avg. SCO residence time, s 2.04 2.14 1-73 1.82 1.89 1.85
Avg. combostion air flow, acm/min 0.40 o.40 °-40 °-40 °-40 °-40
Avg. oxygen, % 7.6 8.4 8.8 8.8 8.4 8.4
Avg. carbon dioxide, % 8.4 8.o 8-6 8-6 9-° 8-6
Avg. carbon monoxide, ppm 1 2 3 2 8 1
Combustion efficiency. % 99.9988 99.9975 99.9953 99.9977 99.9911 99.9988
Avrj, scrubber water (low, gpm 10 10 10 10 10 10
Avg. scrubber water, pH 7.5 3.0 5-° 7-° 7-5 5-°
Particulate/HCI emissions:
Stack gas (towrate, dscm/min 1.726 1.733 2-066 1-868 1-924 1-811
Particulata concentration, mg/dscm 53.33 5322 237.85 235.12 209.51 146.25
Chtormo, g/min 0.00197 n.ooi98 °-0346 °-0691 °-0684 °-0054
HCI removal, % 88.74 99.42 99.64 98.98 99.51 99.89
PCS omissions:
PCB feedrate, g/min 0.348 Q.680 17-67 11-33 23-43 8-18
PC8 output rate, g/min 2.0x10-7 1 OxiO-7 7.78x10-7 6.43x10-7 2.55x10-7 1.63x10-5
PCB ORE, % 99.9994 99^999985 99.999996 99.999994 99.9999989 99.9998
PCDD/PCDF emissions:
Total PCDD emissions, ng/dscm 0.46 NA 67.45 5.31 0.95 NA
Total PCDF emissions, ng/dscm 15.15 NA 117.60 43.00 22.20 NA
Continued)
In the May test, the particulate concentrations
ranged from 0.0522 to 0.0950 gr/dscf (adjusted for 7%
oxygen), as compared to the RCRA standard of 0.080
gr/dscf. When operating the pilot-scale unit at the
feedrate of 100 Ib/h, as planned, ECOVA operators
had difficulties stabilizing the processing conditions,
resulting in the high particulate emissions in Test 1.
The gas flow through the scrubbing system during
the first particulate test exceeded the scrubbing ca-
pacity for the particulate loading. Subsequently re-
ducing the feedrate to 90 Ib/h produced acceptable
particulate emissions (i.e., 0.08 gr/dscf). The soil fee-
drate averaged 85 Ib/h for the final 3 particulates
sampling tests.
AfOx Emissions
Table D-5.10 and D-5.11 present the NOX emissions
data for the Jan. and May 1987 tests. For the May
tests, NOX emissions ranged from 0.07 to 0.09 Ib/h, or
with the thermal rating of the secondary combustor
of 390,000 Btu/h, ranged from 0.18 to 0.23 Ib/MBtu.
This compares marginally with the 0.2 Ib/MBtu NOX
standards for steam-generating boiler units of 250
MBtus or more, for gaseous fuels at 40 CFR 60.40,
but compares favorably with standards for solid (0.50
Ib/MBtu) or liquid fuels (0.40 Ib/MBtu).
90
-------
Table D-5.3. (Continued)
Date
Test
7
1/23/87
Test
8
5/27/87
Test
9
5/27/87
Test
10
5/27/87
Test
11
5/27/87
Operating parameters:
Waste feedrate, kg/h 48.12 42.68 35.64 34.37 34.37
PCB concentration, g/kg 0.418 45.00 45.00 43.00 43.00
PCB feedrate, kg/h 0.020 1.92 1.60 1.48 1.48
Auxiliary fuel feedrate, kg/h 4.35 6.79 6.62 6.78 6.78
Avg. SCO residence time, s 2.12 2.42 2.53 2.50 2.50
Avg. combustion air flow, acm/min 0.40 0.40 0.40 0.40 0.40
Avg. oxygen, % 8.8 9.4 9.3 9.1 8.8
Avg. carbon dioxide, % 8.2 7.9 7.9 7.8 8.4
Avg. carbon monoxide, ppm 1 4 4 7 0.0
Combustion efficiency, % 99.9988 99.9949 99.9949 99.9910 100
Avg. scrubber water flow, gpm 10 10 10 10 10
Avg. scrubber water, pH 8.0 9.5 8.0 7.5 7.5
Particulate/HCI emissions:
Stack gas flowrate, dscm/min 1.726 3.821 3.170 3.141 3.056
Particulate concentration, mg/dscm 121.12 268.94 153.33 154.13 169.50
Chlorine, g/min 0.0017 0.166 0.144 0.0136 0.106
HCI removal, % 99.12 99.13 99.10 99.12 99.32
PCB emissions:
PCB feedrate, g/min 0.33 32 26.67 24.67
PCB output rate, g/min 1.00x10-5 6.48x10-6 4.23x10-6 3.53x10-6
PCB ORE, % 99.997 99.999980 99.999984 99.999986
PCDD/PCDF emissions:
Total PCDD emissions, ng/dscm 21.4 ND ND ND
Total PCDF emissions, ng/dscm 1,534.0 ND ND ND
ND = not detected
Table D-5.4. Dioxins and Furans in Furnace Ash - January
1987 Tests
Test
No.
1
2
3
4
5
6
7
2,3,7,8-TCDD
ppb
<0.26
<0.05
<0.05
<1.0
—
<0.06
Total TCDDs
ppb
<0.8
<0.05
<0.05
<1.0
—
<0.06
Total PCDDs
ppb
220.8
17
27
116
—
2.4
Test
No.
1
2
3
4
5
6
7
2,3,7,8-TCDFs
ppb
< 0.0054
<0.04
<0.04
<0.80
—
<0.05
Total TCDFs
ppb
< 0.0054
<0.04
<0.04
<0.80
—
<0.05
Total PCDFs
ppb
0.0034
0.5
3.4
26
—
<1.06
Table D-5.5. Dioxins and Furans in Furnace Ashand Scrubber
Water - May 1987 Tests
Chemicals
Furnace
Ash, ppb
2,3,7,8-TCDD, ng/dscm
Total TCDDs
Total PCDDs
0.02
0.02
ND
Scrubber Water,
ppt
0.00046
0.00046
ND
2,3,7,8-TCDF, ng/dscm
Total TCDFs
Total PCDFs
0.02
0.02
ND
0.00050
0.00050
ND
ND Not detected
Combustion Efficiency
In accordance with TSCA performance standards,
the combustion efficiency for each of the 3 test runs
of May 1987 was calculated using the contractor
CEM values for CO and CO2. The calculated values
91
-------
r
Tabla D-S.B. PCB ORE For the January 1987 Test
PCB ORE
Tost
No.
1
2
3
4
5
6
7
%
99.9994
99.999985
99.999996
99,999994
99.9999989
99.9998
99.997
gPCB
Out/
kg PCB In
0.000575
0.000147
0.000043
0.00005
0.000011
0.001993
0.0303
PCB Mass
Rate In
mg/h
20.88 x 103
40.8x103
1, 060.2 x 103
679.8 x 103
1, 405.8 X 103
490.8x103
19.8 x 103
5
PCB
Mass
Rate Out
mg/h
0.012
0.060
0.046
0.038
0.015
0.978
0.600
Table D-5.9. Dioxins and Furans in Stack Emissions - May
1987 Tests
Chemicals Test 1 Test 2 Test 3
2,3,7,8-TCDD
Total TCDDs
Total PCDDs
2,3,7,8-TCDF
Total TCDFs
Total PCDFs
Values in ng/dscm
ND = Not detected
0.33
0.33
ND
0.57
0.57
ND
0.11
0.11
ND
0.61
0.61
ND
0.19
0.19
ND
0.77
0.77
ND
Tabla D-5.7. PCB ORE For the May 1987 Tests
PCB ORE
Tost
No.
1
2
3
%
99.999980
99.999984
99.999986
gPCB
Out/
kg PCB In
0.000203
0.000159
0.000143
PCB Mass
Rate In
mg/h
1,920 x 106
1.600X 106
1,480 x 106
roo
Mass
Rate Out
mg/h
0.389
0.254
0.212
Table O-5.8. Dioxins and Furans in Tack Emissions - January
1987 Tests
Tost
No.
1
2
3
4
5
2,3,7,8-TCDD
<0.11
< 0.056
< 0.01 29
<0.011
Total TCDDs
<0.52
0.95
0.65
0.48
Total PCDDs
<1.34
68.09
5.40
0.95
<0.53
10.56
20.9
Test
No.
i
2
3
4
5
6
7
2,3.7,8-TCDFs
1.34
—
3.6
1.8
1.4
—
113
Total TCDFs
6.88
—
15.6
8.80
6.04
—
609
Total PCDFs
15.1
_
114.3
41.3
20.8
—
1,420
Values in ng/dscm
were significantly higher than the TSCA perfor-
mance standard of 99.9%.
Scrubber Water
PCB levels in the scrubber water for the Jan. and
May tests are presented in Table D-5.12.
Analyses of the dioxins and furans in the scrubber
water for the Jan. tests are presented in Table D-
Table D-5.10.Stack Emissions - January Tests
Test Particulates HCI
No.
1
2
3
4
5
6
7
Table D-5.11
Test
No.
1
2
3
4
gf/dscf Ib/h
0.0233 0.00026
0.0232 0.00027
0.1039 0.0047
0.1246 0.0094
0.0915 0.0093
0.0639 0.0074
0.0529 0.0022
NOX
ppm
103
100
99
92
96
109
133
RCI
mg/m3
46.79
46.57
45.50
31.17
37.38
38.28
37.39
.Stack Emissions - May Tests
Particulates
gr/dscf
0.0950
0.0528
0.0522
0.0559
HCI
Ib/h
0.022
0.019
0.018
0.014
NOX
ppm
102
98
98
—
Table D-5.12.Scrubber Water PCB Levels
Jan. Tests May Tests
Test
No.
1
2
3
4
5
6
7
Scrubber water
PCB level,
ppb
—
<15
5.37"
5.37*
5.37*
<100
0.148
Test
No.
1
2
3
4
Scrubber water
PCB level,
ppb
9.7
1.6
9.9
—
The scrubber water samples for Tests 3, 4, and 5 were composited
and analyzed, giving one result.
5.13. May tests used composite samples of scrubber
water for dioxins and furans analysis. Results are
given in Table D-5.5.
Operations
Since the Twin Cities tests were performed using the
nominal 80 to 100 Ib/h pilot-scale unit, operational
92
-------
experiences are not scalable to the large commercial
units. Two problems that were encountered are dis-
cussed below.
Table D-5.13.Dioxins and Furans in Scrubber Water - January
1987 Tests
Test
No.
1
2
3
4
5
6
7
2,3,7,8-TCDD
< 0.0054
< 0.0038"
< 0.0038*
< 0.0038*
—
< 0.0089
t<
Total TCDDs
< 0.0054
< 0.0038*
< 0.0038*
< 0.0038*
—
< 0.0089
Total PCDDs
< 0.0034
< 0.0282*
< 0.0282*8.09
< 0.0282*
—
0.021
Test
No.
1
2
3
4
5
6
7
2,3,7,8-TCDFs
< 0.0043
< 0.003*
< 0.003*
< 0.003*
—
< 0.0071
Total TCDFs
:
< 0.0043
< 0.003*
< 0.003*
< 0.003*
—
< 0.0071
Total PCDFs
< 0.0025
45*
45*
45*
—
< 0.0071
' Scrubber water samples for Tests 3, 4, and 5 were composited and
analyzed.giving one result.
Values in ppb
Feed rates greater than 100 Ib/h apparently caused
instability in the test operations. The high feedrate
resulted in an increase in the flue gas velocity be-
cause a greater quantity of fuel and combustion air is
required to destroy the PCBs, producing a higher vol-
ume of combustion gases. Data revealed high stack-
gas velocity as a potential indicator of process insta-
bility. This also resulted in high particulate emis-
sions since the capacity of the scrubber system was
exceeded.
In the Jan. testing, excess particulate emissions were
caused by plugging of the wet-gas-scrubber venturi
and tower-scrubbing-liquid nozzles, which greatly
reduced the efficiency of the scrubbing process. The
plugging originated from corrosion scale on the walls
of the piping. This scale subsequently released from
the walls and collected at the nozzles. Leaks in the
piping, found at the conclusion of the 3 tests,
prompted a replacement of the metal piping with
plastic piping prior to the further soil testing. The
particulate emissions decreased after the piping
change.
93
-------
-------
APPENDIX D-6
BRIO PILOT-SCALE TESTS [14]
The Shirco pilot-scale unit, contracted by the Brio
task force, was in operation at the Brio refinery site
in Houston, TX. between Feb. 9 and 14,1987. The ob-
jectives of these thermal treatment tests on excavat-
ed pit material were as follows:
• To determine the incinerator-ash chemical compo-
sition.
• To demonstrate that the incinerator feed system
can reliably provide a continuous, blended feed to
the incinerator and deposit this feed material in a
uniform manner on the incinerator belt.
• To demonstrate that the incinerator can meet the
RCRA performance standard of 99.99% destruc-
tion and removal efficiency for POHCs.
• To provide design information and economic data
required to evaluate the feasibility of incinerating
certain Brio-Site pit wastes.
The actual series of test burns was performed from
Feb. 10 through 13, 1987. Shirco Infrared Systems,
Inc. personnel operated the unit and prepared the
test matrix.
Feed Preparation
The feed for the test was evacuated from the Brio on-
site pits using a backhoe. Materials from four differ-
ent pits were obtained and packed in 55-gal drums.
System operators and feed preparation personnel re-
ported that overall the consistency of the feed was a
tacky soil that had a clay content. The feed also con-
tained large pieces of tar chunks. To produce the re-
quired feed size of less than 1/2 in., manual screening
and delumping was necessary. Large pieces of tar
chunks shattered when struck. Some material that
was more clay-like and contained tarry chunks re-
quired a great deal of effort to prepare through a 1/2-
in. screen. The clay had to be pressed through the
screen and the tarry chunks had to be broken by im-
pact and passed through the screen. Although lime
was not needed for acid neutralization, a small per-
centage of lime or other materials like fly ash would
be useful to reduce the tacky nature of the feed.
To demonstrate that the process can meet RCRA per-
formance standards for DRE of POHCs, it was neces-
sary to spike the feed material with carbon tetrachlo-
ride, CC14. The method employed to accomplish spik-
ing involved placing a preweighed amount of feed,
about 50 Ib, in a cement box and adding to the feed a
predetermined amount of carbon tetrachloride CC14,
diluted with hexane. The feed then was quickly
mixed using a garden hoe, immediately shoveled into
plastic 5-gal buckets, and sealed.
The feed material was screened through a 1/2-in.-
mesh screen before introduction into the feed hopper.
Test Procedure
A total of 8 test runs were conducted, which repre-
sented feed material from 4 different pits (J, I, M,
and B) with each material tested at residence times
of 12 and 18 min in the primary combustion cham-
ber. A knife gate near the entrance region of the con-
veyor belt was used to control the size of the material
entering the furnace and to set the height of the feed
material on the belt, which in turn controls the fee-
drate. The beginning of each run was started by feed-
ing unspiked material until the system was stabi-
lized. The startup of the actual test began when the
feed of spiked material began. The test ended when
the last of the spiked feed was discharged off the belt
in the furnace.
The PCC of the unit consists of two zones (A and B),
which can be individually controlled for tempera- '
ture. Throughout the test program, the PCC tem-
perature was controlled through combustion air ad-
dition and auxiliary electric power to between 1,550°
and 1,600°F in Zone A and a nominal 1,600°F in Zone
B. The primary combustion chamber exhaust tem-
perature was maintained between 1,600° and
1,700°F for the Pits J and I materials. However, for
95
-------
the Pits M and B materials, the PCC exhaust and the
Zone A temperatures were decreased. This is the re-
sult of a higher combustible content in the Pits J and
I waste as compared to the lower Btu content of the
Pit M and B material. The scrubber stack tempera-
ture was maintained at a level between 175° and
181°F throughout the tests. The secondary combus-
tion chamber was maintained between 2,150° and
2,200°F during all testing. Table D-6.1 presents the
process conditions for each of the 8 tests conducted at
the Brio site.
For each test burn, a complete analysis was conduct-
ed on the feed, ash, scrubber makeup water (before
test run), scrubber water (after test run), scrubber in-
let gas, and scrubber exhaust gas (stack) samples.
Results
Characteristics of the Feed
As described earlier, material from Pit J (excluding
rocks) would break apart easily. The waste material
from Pits B, I, and M was much more clay-like and
included some tarry chunks.
All test material was spiked with carbon tetrachlo-
ride (CC14) to determine the destruction capability of
the thermal treatment. CCU is the fourth hardest
compound on the difficult-to-destroy-by-thermal-
treatment scale, based on a hierarchy established by
the EPA. It has a higher rating than any other com-
pound found on the Brio site. CC14 was injected into
the soil and was mixed for several minutes to homog-
enize it.
In addition to CC14, the feed material contained sev-
eral different contaminants. Table D-6.2 provides
ranges of some of the typical contaminants found in
the feed material from the four pits.
Characteristics of the Furnace Ash
The analysis of the furnace ash for all test runs indi-
cated the destruction of all potential problem com-
pounds to levels below the level of concern and often
below the minimum level of detection. Also, analysis
was done for chlorinated pesticides/PCBs, organo-
phosphorous pesticides, phenoxy herbicides, metals,
cyanide, sulfide, fluoride, dioxins, and furans. None
of these chemical pollutants were detected. The EP
Tox test for leachable metals showed all metals to be
below the toxicity characteristic standards. Sulfides
combined in the furnace ash ranged from 170 to 9350
ppm.The principal contaminant, CCU, which ranged
from 4 to 128 mg/kg in the feed material, was ana-
lyzed at less than 0.005 mg/kg in the furnace ash for
all tests.
Table D-6.3 presents the weights and volumes of feed
material during the test program. From this data,
weight and volume reduction percentages were cal-
culated. These two percentages were defined as the
percentage of the initial value removed. The table
shows that a nominal weight reduction ranging from
38% to 51% was accomplished for each test. Pits I and
J had weight reduction ranging from 38% to 45%,
while both Pits M and B resulted in 51% reductions.
The volume reductions were similar in all tests. All
12-min PCC residence-time-cases resulted in a nomi-
nal 55% volume reduction, whereas the 18-min cases
were 45%. This suggests some increase in ash
particle size with extended thermal exposure. In
summary, these tests indicate that mass and volume
may both be reduced by approximately 50% through
thermal processing.
Tablo D-6.1. Brio Site Process Conditions
Tost
NoJPit
1AJ
2/J
3/J
4fl
5/M
6,M
7IB
8/B
CCL4
Spike
mL
30
30
30
30
30
30
30
30
Res.
Time
min
12
18
18
18
12
18
18
12
Feed-
Rate
Ib/h
71.9
'58.2
50.2
67.3
43.2
33.0
41.7
42.3
Temp.
Zone A
°F
1,596
1,596
1,601
1,600
1,510
1,565
1,540
1,531
Temp.
Zone B
°F
1,612
1,613
1,608
1,586
1,599
1,612
1,612
1,612
Temp.
sec
°F
2,159
2,224
2,204
2,200
2,235
2,209
2,195
2,216
Primary
Exhaust
°F
1,727
1,611
1,601
1,601
1,189
1,258
1,234
1,232
Power
kWh
15.3
16.5
13.8
16.8
21.9
19.6
23.0
21.6
Fuel
Ib/h
10.9
11.7
9.7
8.9
9.7
8.3
10.3
9.5
Knife
Gate
in.
1 1/8
1 1/4
1 3/8
1 1/8
1 1/8
1 1/8
1 1/4
1 1/4
96
-------
Table D-6.2 Typical Contaminants in Brio Site Feed Material Using the stack gas flowrate and the volume of gas
Compund Concentration, mg/kg sample caught in the MM5 sampling train, the
• maximum amount of CCLi that could have passed
Acetone < 1 to 120 through the stack was calculated. All these values
Anthracene < 20 to 32 were transformed into hourly flowrates using the
Benzene <0.50 to 3.6 total test time. Table D-6.5 presents these calcu-
Carbon disulfide < 0.50 to 5.7 lations in a tabular form, along with the resulting
Carbon tetrachloride* 4 to 128 destruction arid removal efficiency, DRE.
Chlorobenzene < 0.50 to 31
Chloroform < 0.50 to 43 The results presented in Table D-6.5 show that the
1,2-Dichloroethane 8.3 to 106 DRE of CC14 was at minimum, greater than
1,1-Dichloroethane < 0.50 to 31 99.9997% for the 8 tests performed.
1,1-Dichloroethylene < 0.50 to 16
Ethyl benzene 7.4 to 160 The DRE results were based on the total amount of
Napthalene < 20 to 140 CC14 added to the feed. The reason for this approach,
Nitrosodiphenylamine < 20 to 100 as opposed to a DRE based on feed sample analysis, is
Phenanthrene < 20 to 416 that when adding a liquid chemical into a material
Styrene < 0.50 to 140 with a high clay content (such as the test material),
1,1,2,2-Tetrachloroethane < 0.50 to 39 one cannot achieve a homogeneous mixture capable
Tetrachloroethylene < o.so to 28 of supporting testable grab samples. For example., in
Tetrachloromethane 4 to 128 * grab"fee(J samPling testing program, the composite
T . naot sample could potentially contain a small or
''.''.'"',", large concentration of the trace compound producing
i,2trans-D,chloroethylene < o.so to 2.6 data for DRE purposes that would not be comparable
1,1,1-Trichloroethane < 0.50 to 1.5
1,1,2-Trichloroethane < 0.50 to 132
Trichloroethylene < 0.50 to 33 Other Organic Contaminants—
Vinyl chloride < 1.0 to 3.0 Analyses of the scrubber inlet and the stack gases
for polychlorinated dibenzo-p-dioxins and dibenzo
'Spiking chemical furans did not detect these compounds at levels
greater than the detection limit of <0.75 ug/mL of
Scrubber Inlet/Stack Gas Data concentrated extract.
Particulate Emissions-
Table D-6.4 summarizes the exhaust-stack particu-
late loadings for each run. The levels are all below
the 0.08 gr/dscf RCRA performance requirement.
Continuous Emissions Monitoring—
Also summarized in Table D-6.4 are the emissions of
Os, SO2, CO, and NOX as measured by the
continuous monitoring equipment at the scrubber
inlet.
Destruction and Removal Efficiency (DRE) of CCU--
Table 0-6.3. Weight and Volume Reduction of Waste Feed Materials
In the analysis of the charcoal tubes in the MM5
train for volatile organics, a quantity of methyl
chloride was detected. This was the solvent used to
clean the train prior to the test. The only other
compounds found were toluene, methyl bromide,
tetrachloroethane, chloroform, and trichloroethylene
— all at levels less than established standards for
direct inhalation.
Primary Chamber
Residence Time
mm
Run
Initial
Weight
Ib
Ash
Weight
Ib
Weight
Reduction
Initial
Volume
Ash
Volume
Volume
Reduction
12
18
18
12
12/18
18
12
J-1
J-2
1-1
I-2
M-1,2
B-1
B-2
260.5
162
138
113
302
100
155
160
101
78
62
147.5
77
75
38.6
37.7
43.5
45.1
51.2
23.0
51.6
5.788
3.6
2.97
3.63
7.02
2.22
3.44
2.588
1.956
1.6
1.6
2.99
1.15
1.61
55.3
45.66
45.86
55.67
57.42
48.25
53.19
97
-------
r
Table D-6.4. Brio Site Stack Gas Analyses
Analysis
Test Number
ParticulatQ(a) gr/dscf
SO3, IWh
SO2, ib/h
CO, ppm
NOX, Ib/h
HCI,mofl-
DRE.%
0.015
0.12
0.34
0.0
0.022
< 0.0678
> 99.9998
0.022
0.29
0.20
0.0
0.02
< 1 .095
> 99.9997
0.027
0.10
0.11
0.0
0.025
< 1.380
> 99.9998
0.034
0.076
0.055
0.0
0.026
< 1.323
> 99.9997
0.007
0.014
0.077
0.0
0.027
< 1.034
> 99.9997
0.006
0.005
0.001
0.0
0.025
< 0.61 9
> 99.9997
0.018
0.013
0.004
0.0
0.030
< 1.031
> 99.9997
0.016
0.010
0.001
0.0
0.022
< 0.953
> 99.9997
(aJCorrccted to 7% O2
Scrubber Effluent/Makeup Water
Scrubber water was analyzed for priority pollutants,
cyanide, total organic carbon, and chlorides. The
organic portion of the priority pollutants were all
lower than the detection limit, with the exceptions of
methyl bromide (0.01 mg/L) di-n-butyl phthalate
(0.39 mg/L), methylene chloride (0.012 mg/L),
methyl chloride (0.010 mg/L), 2-ethylhexyl phthalate
(0.04 mg/L), and phenol (0.13 mg/L). The high
chloride levels are due to the high concentration of
chlorinated hydrocarbons in the feed. Chloride exists
as sodium chloride in the scrubber water. Carbon
tetrachloride levels in both the makeup and the
scrubber were all <5.0 ug/L indicating virtually no
carbon tetrachloride in the water.
The metal and organic concentrations in both the
plant-supplied makeup water and the scrubber
effluent were essentially the same in each case.
Thus, it was concluded that there was not significant
addition of metals to the scrubbing liquid during
thermal processing.
Full Scale System Economic Analysis
An economic analysis was performed to determine
treatment costs for the Brio site if a commercial-scale
unit were used. I/sing a site size of 125,000 tons and
Pit J waste data, 2 commercial-size units were
considered. A nominal 150 ton/d unit will use a 9 x 61
ft primary chamber. The waste-treatment cost data
for the 150-ton/d system is shown in Table D-6.6,
which shows a minimum treatment cost of $143/ton.
Similarly, the cost for a 9 x 85-ft primary chamber
unit designed to treat 220 ton/d is estimated at
?119/ton. These costs do not include costs for feed
excavation, feed preparation, ash disposal, interest,
and taxes. The estimates are accurate to + 25%.
Both units were assumed to operate 50 wk/yr, 6 d/wk
giving a utilization factor of 82.42% (300 d/yr).
Table D-6.6. Onsite Mobile Incineration Service Estimated
Economic Model
Equipment: 1 Shirco mobile system
Capacity: 150 ton/d
Yearly throughput: 45,000 tons, (assuming 50 wk/yr, 6 d/wk
operation)
Direct operating costs: $ 76.03/ton
Equipment cost: 29.86/ton
Profit, taxes, and opportunity cost: 37.06/ton
Total minimum cost $l42.95/ton
Operation
The Shirco unit used at the Brio site was a pilot-scale
unit with minimal operational problems. The main
problems related to the feed system. Since a pretest
feed-preparation study was not performed, the
equipment available for screening, delumping, and
mixing was not adequate for the task. It was found
that, in order to produce a desired feed size, manual
screening and delumping was necessary. A hardwire
screen was needed to breakup the lumps and remove
larger rocks.
It is concluded that all materials at the site will
require delumping and screening prior to
incineration. For the wastes in 3 pits, mixing with
lime, kiln dust, fly ash, or dry soil is recommended to
minimize the sticky nature and simplify materials
handling. To facilitate system design a materials
preparation test is recommended.
98
-------
Table D-6.5. Destruction and Removal Efficiency of CCI4
Test
Pit
Stack flow, DSCFH
DSCF
Sample volume, DSCF
DSCFH
Sampling date
Sampling interval, time
min
Total waste fed, Ib
Total CCI4 in feed, g
Total CCI4 in sample, g
Total CCI4 in stack, g
CCI4 emission rate, g/h
CCI4 feed rate, g/h
ORE, %
(CCI4 feedrate
ORE =
1
J
5,163
9,035
65.035
37.1628
2/10/87
15:37-17:22
105
150
48
< 7.1 08x1 0'7
< 9.875x1 0-5
< 5.643x1 0-5
27.4286
> 99.9998
- CCI4 emission rate) x
2
J
5,542
10,899
59.121
30.0615
2/11/87
10:02-12:00
118
105
48
<7.131X10'7
< 1.31 5x1 0-"»
< 6.684x1 0-5
24.4067
> 99.9997
100%
3
I
6,063
10,913
54.924
30.5133
2/11/87
14:16-16:04
108
63
48
< 7.282x1 0-7
<1. 4469x1 0'4
< 8.038x1 0-5
26.667
> 99.9998
4
I
5,164
6,283
33.488
27.5243
2/11/87
18:45-19:58
73
63
48
< 7.2327x1 0-7
<1. 357x1 0'4
<1. 115x1 0-4
39.452
> 99.9997
5
M
4,426
8,778
44.259
22.3154
2/12/87
10:50-12:49
119
75
48
< 7.263x1 0-7
<1. 440X10-4
< 7.263x1 0-5
24.2017
> 99.9997
6
M
4,883
15,056
72.527
23.5222
2/12/87
15:02-18:07
185
75
48
< 7. 166x1 0-7
<1. 4877x1 0'4
< 4.825x1 0-5
15.5675
> 99.9997
7
B
4,710
9,028
43.208
22.5433
2/13/87 '
10:21-12:16
115
60
48
< 7.44x1 0-7
<1.55x10-4
< 8.1 11x10-5
25.0434
> 99.9997
8
B
4,872
10,150
49.154
23.5939
2/13/87
13:43-15:48
125
75
48
< 7.383x1 0-7
<1. 525x1 0'4
< 7.31 8x1 0-5
23.0400
> 99.9997
CCI4 feedrate
Legend: DSCFH - Dry standard cubic feed per hour
DSFC - Dry standard cubic feed
-------
-------
APPENDIX D-7
TIBBETTS ROAD PILOT-SCALE TESTS [15]
Introduction
The Environmental Protection Agency (EPA)
contracted the O.H. Materials Corp. in Sept. 1986 to
use a small-scale mobile hazardous-waste incin-
erator to detoxify contaminated soil at the Tibbetts
Road Hazardous Waste Site in Harrington, NH. The
soil was contaminated with dioxin, polychlorinated
biphenyls (PCBs), herbicide, and solvents. Sub-
sequently, Shirco Infrared Systems, Inc. was
contracted by O.H. Materials to perform the detoxi-
fication in its pilot-scale unit. The objectives of the
program were as follows:
• To decontaminate approximately 5 yd3 of exca-
vated soil, including a percentage of rocks, wood,
and plastic. The soil contained volatile and semi-
volatile organic pollutants, including poly-
chlorinated biphenyls (PCBs) and dioxins and
furans.
• To determine "the furnace-ash chemical composi-
tion.
• To demonstrate that the unit can meet the RCRA
and TSCA performance standards for the
Destruction and Removal Efficiency (DRE) of the
designated priority pollutants.
Field incineration activities at the site were
conducted on a 24 h/d basis for the period from Nov. 6
through 14, 1986. Continuous measurements of
selected stack-gas parameters were conducted dur-
ing this entire period to ensure the efficient opera-
tion of the incinerator. In addition, 3 discrete test
runs were conducted to document the DRE and
process stream concentrations of selected hazardous
organic species of interest.
Feed Preparation
The feed to be processed was excavated from the
Tibbetts Road site prior to the test and stored in a
waste dumpster, which was sealed until the test
date. Prior to use as feed for the Shirco unit, the
waste material was screened through a 1-in. hard-
ware cloth to remove large rocks, sticks, and pieces of
plastic. After screening, the material was placed into
plastic 5-gal buckets and sealed. The buckets were
weighed and placed by the feed hopper for feeding.
The contaminated soil was fed to the unit by a feed
operator. The 5-gal bucket was transported to the top
of the primary-chamber feed module and manually
fed to the feed hopper in 5-lb increments. As needed,
the material would be spread and more material
added such that the feedrate remained constant. The
rate of soil feed to the furnace was set by adjusting
the feed hopper gate opening to 1 in. and the belt
speed to a 20-min residence time. Feeding continued
in this manner throughout the project test period.
Test Procedure
Operating parameters for the unit while incin-
erating Tibbetts Road waste (including the three
emission test runs) are given in Table D-7.1.
For the demonstration test at the Tibbetts Road site,
PCC Zones A and B were controlled at a set point
temperature of 1600°F. The PCC temperature was
controlled between 1,500° and 1,600°F combustion
air and auxiliary electric power. The SCC tempera-
ture was maintained between 2,200° and 2,350°F.
A comprehensive sampling and analytical program
was conducted. The goals of this sampling and
analytical program were to:
• Determine the DRE of the principal organic
hazardous constituent (POHC), PCBs, as defined
by Region I;
• Characterize each of the 4 process streams for
selected hazardous constituents, including PCBs
and chlorinated dioxins/furans; and
101
-------
Tablo D-7.1.Operation Summary
Primary Chamber
Secondary chamber
Date
11/05/86
11/06/86
11/06/86
11/06/86
11/07/86
11/07/86
11/07/86
11/08/86
11/09/86
11/09/86
11/09/86
11/11/86
11/12/86
11/12/86
11/13/86
11/13/86
11/14/86
Time
Period,h
13:23-16:20
02:40-09:40
09:40-13:50
13:50-24:00
00:00-10:16
10:16-13:31
13:31-24:00
00:00-24:00
00:00-07:40
07:40-09:30
09:30-13:51
15:50-24:00
00:00-06:00
17:25-24:00
00:00-10:00
11:08-24:00
00:00-16:30
Emissions Residence
Test No. Time, min
20
20
1 19.5
20
2 20
20
20
20
25
3 25
30
30
20
20
20
20
Temp.,
°F
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
1,600
.1,600
Feed-Rate,lb/h
25.42
51.21
48.19
42.93
37.00
61.92
47.62
34.53
46.65
53.57
45.16
34.02
25.71
59.67
59.67
80.76
21.16
Temp., Residence
°F Time, s
2,300
2,300
2,350
2,350
2,350
2,350 2.24
2,350 2.13
2,380 2.08
2,375
2,380
2,375 2.01
2,350
2,350
2,400
2,350
2,350
2,325
* Monitor the combustion efficiency of the system
throughout the entire test period to ensure
Operations appropriate for the destruction of the
designated organic species of interest.
During the entire time the incineration system was
in operation, the monitoring for gases such as carbon
monoxide (CO), carbon dioxide (CC>2), oxygen (Oz),
and oxides of nitrogen (NOS) was performed.
Samples of the waste feed, furnace ash and scrubber
water were taken during each of the 3 stack-
emissions sampling periods. These samples were
later analyzed in the laboratory for organics and
chlorides. Stack gas sampling over three 4-h periods
was performed, and analyses for organics, partic-
ulatcs, HC1, and RC1 were conducted.
Results
The following discussion presents the results of the
analysis of the waste feed, furnace ash and scrubber
water during the 3 emissions sampling tests for
PCBs, PCDDs, and PCDFs. Also reported are the
results of the stack gas analysis for PCBs, PCDDs,
PCDFs, particulates, HC1, RC1, and fixed gases (CO,
COg, Oa, NOX). Based upon the feed and stack gas
concentrations of the defined priority pollutants, the
destruction and removal efficiencies are also
presented.
Characteristics of the Feed
Results from the PCB analyses of the waste feed are
summarized in TableD-7.2. As indicated in the table,
the contaminated material originally contained an
average of 700 ppm of total PCB. The distribution of
the congeners and the predominance of the
hexachloro- and heptachloro-biphenyls in the waste
suggested high-boiling Aroclors, such as 1260.
Table D-7.2. PCB Concentrations in Waste Feed, ppm
Isomer Group Test 1 Test 2 Test 3
Cl,
CI2
CI3
CU
CI5
CI6
CI7
CI8
CI9
Cl,,
-PCB
-PCB
-PCB
-PCB
-PCB
-PCB
-PCB
-PCB
-PCB
3 -PCB
NDa
NDa
NDa
2.4
61
280
190
40
2.8
NDa
NDa
NDa
NDa
2.3
74
310
210
45
2.5
NDa
NDa
NDa
NDa
31.
86
430
300
59
3.8
NDa
NDa = 3.6 ppm
102
-------
Results of dioxin and furan analyses on the waste
feed are summarized in Table D-7.3, and indicate the
presence of chlorinated furan and tetrachlorodi-
benzo-p-dioxin species in all samples. Concentrations
of the chlorinated furans ranged from 333 ppt to 2.0
ppb. The concentrations of total
Table D-7.3. PCDD/PCDF Concentrations in Waste Feed, ppb
Isomer Group Test 1 Test 2 Test 3
Table D-7.5. Concentrations of PCDD/PCDF in Furnace Ash,
ppb
PCDF
PCDD
Tetra
Penta
Hexa
Hepta
Octa
Tetra
Penta
Hexa
Hepta
Octa
0.081
0.19
0.69
1.3
1.6
0.88
<0.015
<0.019
<0.15
<0.20
0.081
0.48
0.97
1.6
2.0
3.8
<1.2
<0.11
<0.15
<0.20
0.033
<0.20
0.38
0.71
0.91
1.2
<0.84
< 0.094
0.15
0.20
TCDD ranged from 0.88 to 3.8 ppb. Additional
analyses specific for the determination of 2,3,7,8-
TCDD were not conducted as part of this project.
Previous data collected by EPA Region I confirmed
the fact that this isomer was not present in the soils
at the site.
The "worst case" TCDD equivalents for the waste
feed samples ranged from 0.92-4.04 ppb.
Characteristics of the Furnace Ash
PCB and PCDD/PCDF analysis results of the furnace
ash are given in TablesD-7.4 and D-7.5. The total
PCB concentrations in the furnace ash samples from
each run ranged from 5.7 to 16.4 ppb. PCDD/PCDF
analyses of the furnace ash samples indicate that no
detectable levels of these species were present. The
TCDD equivalents for these samples averaged 0.006
ppb.
Table D-7.4. Concentrations of PCBs in Furnace Ash, ppb
Isomer Group Test 1 Test 2 ' Test 3
cu
CI5
CI6
CI7
CI8
Clg
-PCB
-PCB
- PCB
-PCB
- PCB
-PCB
NDb
NDb
3.6
2.1
NDb
NDb
NDb
1.2
8.3
4.6
NDb
NDb
NDb
1.5
9.3
5.6
NDb
NDb
Isomer Group
Test 1
Test 2
Tests
PCDF
PCDD
Tetra
Penta
Hexa
Hepta
Octa
Tetra
Penta
Hexa
Hepta
Octa
< 0.0011
< 0.011
< 0.0039
< 0.0032
< 0.0085
< 0.0041
< 0.0059
< 0.0023
<0.10
<0.20
< 0.001 2
<0.012
< 0.0039
< 0.0085
< 0.0075
< 0.0022
< 0.0053
<0.0031
< 0.0044
<0.10
< 0.001 8
< 0.011
< 0.0043
< 0.022
< 0.0075
< 0.0033
< 0.0096
< 0.0095
<0.10
<0.50
Destruction and Removal Efficiency
(ORE)
Concentrations of PCBs and PCDF/PCDDs in the
stack gas are presented in Tables D-7.6 and D-7.7.
The calculated destruction and removal efficiencies
for total PCBs as well as individual isomer groups
are presented in Table D-7.8. In those instances
where there were no detectable levels of PCBs in the
stack gas samples, the DREs were calculated using
the associated analytical detection limits. The
concentrations of PCBs detected in the stack gas
samples from Test 2 resulted in a DRE of 99.99981%.
DREs calculated for the individual isomer groups
detected in the stack gas sample ranged from
> 99.99% to 99.99981%.
Table D-7.6. Concentrations of PCBs in Stack Gas, ng/m3
Isomer Group Test 1 Test 2 Test 3
CI4 - PCB -
CI5 - PCB -
CI6 - PCB -
CI7 - PCB -
CI8 - PCB -
CI9 - PCB -
NDd 40 ng/m3 NDg 30 ng/m3
NDe 20 ng/m3 NDh 45 ng/m3
NDf 25 ng/m3
NDd
27.8
122.4
74.5
NDg
NDh
NDe
NDe
NDe
NDf
NDf
NDh
NDb = 13 ppb
The analytical detection limits were sufficient to
demonstrate DREs in excess of 99.9999% for Test 3.
The "greater than" values for these calculations,
which do not meet the requirement of 99.9999%, are
the result of PCB feed concentrations less than 100
ppm.
103
-------
Table D-7.7. Concentrations of PCDD/PCDF in Stack Gas,
ng/m3
isonw Group Test 1
PCDF
PCOD
Tetra —
Penta —
Hexa —
Hepta —
Octa —
Tetra —
Penta —
Hexa —
Hepta —
Octa —
Testa
0.111
< 0.097
< 0.062
< 0.1 22
< 0.478
< 0.032
< 0.1 33
< 0.1 07
< 0.1 78
< 0.367
Tests
<0.115
< 0.287
< 0.089
< 0.289
<1.09
< 0.045
<0.112
< 0.097
< 0.420
<1.06
Tablo D-7.8. Destruction and Removal Efficiencies of PCBs
and PCDDs/PCDFs, %
Isomer Group Test 1
Test 2
Tests
PCB
Tola! PCBs
PCDF
PCDD
cu
Clg —
CI6
C17
CI8
CI9
-
Tetra —
Penta —
Hexa —
Hepta —
Octa —
Tetra -
Penta —
Hexa —
Hepta -
Octa —
> 99.9909
99.99979
9.99979
99.99981
> 99.99967
> 99.99904
99.99981
99.26441
> 99.89201
> 99.96560
> 99.5904
> 99.871 90
> 99.99545
—
—
—
—
> 99.996
> 99.99988
> 99.99997
> 99.99994
> 99.999975
> 99.991
> 99.99992
>. 98.33534
—
> 99.88724
> 99.79883
> 99.42652
> 99.9821 5
—
—
—
—
A general review of the incinerator operating
conditions was conducted in an attempt to resolve
the variation in DREs for Tests 2 and 3. Since the
SCO temperature was in excess of 2,200°F during
both of these runs, the review was centered around
the gas residence time and turbulence in the
chamber. The calculated gas residence-time for both
of the test runs was determined to be in excess of the
2-s period required by TSCA. The degree of
turbulence, or mixing, in the secondary chamber was
evaluated by the calculation of a Reynolds Number
for the combustion gases. Turbulent flow exists at
Reynolds Numbers in excess of 2,300. Below this
number, laminar or transition flow prevails and
mixing occurs only by diffusion. The Reynolds
Number for Test 2 was calculated to be 2,200,
indicating transition flow, which may be responsible
for the DREs below the 99.9999% level. Values
calculated for Test 3, during which the incinerator
did achieve the required DRE, are in excess of 2,400.
Valid data from Test I were not available due to a
non-quantitative transfer of the sample extract
during the final concentration step in the laboratory.
As indicated in Table D-7.7, detectable levels of
TCDF were present in the sample extract from Test 2
at a level of 0.111 ng/m3. The level of TCDF in the
stack gas corresponds to a DRE of 99.26%. The
remaining DRE values were calculated using the
analytical detection limits for each isomer group.
These detection limits were not appropriate for the
determination of the required level of destruction
due to the low levels of these constituents in the
waste feed.
The calculated DRE for TCDF is not consistent with
the level of destruction demonstrated for PCBs. A
comparison of the heats of combustion, the general
measure of incinerability currently used by the EPA,
for TCDF and PCBs indicates that these compounds
should behave similarly under identical process
conditions. The fact that the calculated DREs for
these two constituents are so profoundly different
suggests that the TCDF in the stack gas may be a
product of incomplete combustion (PIC) related to
the low turbulence condition that was present in Test
2. This can be remedied by simple modifications to
the design of the SCC to produce a more turbulent
atmosphere for the complete oxidation of organic
material.
Other Stack Emissions
Additional analyses, including: fixed gases; total
particulate; hydrochloric acid; oxides of nitrogen and
total chlorinated organics, were also conducted on
the stack gas stream. The averaged results of these
analyses are presented in Table D-7.9.
The calculated combustion efficiencies for all the test
runs were determined to be greater than 99.9%. The
associated concentrations of carbon monoxide in the
flue gas stream ranged from 2.1 ppm to a high of 8.0
ppm.
The particulate concentration values reported in
Table D-7.9 have been corrected to 7% O2.
Particulate emissions ranged from 0.040 to 0.050
gr/dscf and are in compliance with the RCRA
performance standard of 0.08 gr/dscf.
104
-------
Table D-7.9. tack Gas Composition
Table D-7.10. PCDF/PCDD Concentrations in Scrubber Efflu-
ent (ppb)
Parameter
Particulate .emissions
gr/dscf @ 7% O2
HCL emissions, Ib/h
RCL emissions, mg/m3
Fixed gas composition
Oxygen, %
Carbon dioxide, %
Nitrogen oxides, ppm
Average carbon monoxide,
ppm
Combustion efficiency, %
Test 1
0.044
0.002
2.0
-
8.4
8.4
163.1
4.2
99.99
Test 2
0.040
0.003
< 5.2
-
8.7
8.3
175.2
4.2
99.99
Test3
0.050
0.015
< 2.8
-
10.3
8.0
209.8
3.0
99.999
Isomer Group
PCDF Tetra
Penta
Hexa
Hepta
Octa
PCDD Tetra
Penta
Hexa
Hepta
Octa
Test 1
<0.40
<1.8
<0.76
<2.9
<8.8
<0.76
<2.0
<2.1
<4.0
<6.9
Test 2 Test 3
<0.26 <1.0
<1.1 <1.6
<0.72 <1.6
< 1 .6 < 6.0
<5.2 <7.6
<0.32 < 0.038
<1.3 <1.8
<1.2 <2.4
<2.6 <3.8
<3.6 <8.4
Hydrochloric acid (HCL) emissions from the system
were determined to be less than 4 Ib/h, and in
compliance with the RCRA performance standards
for hazardous waste incinerators.
The average stack gas concentrations of nitrogen
oxides (NOX) and total chlorinated organic (RCL)
were determined to be 182.7 ppm and <3.3 mg/m3
respectively.
Scrubber Water
Analyses of scrubber water from each run indicate
that no detectable levels of PCBs were concentrated
in this process stream. The average analytical
detection limit for these analyses was 18 ppb.
Results of analysis for PCDDs and PCDFs are
presented in Table D-7.10.
No detectable levels of PCDF/PCDD species were
present in the scrubber effluent samples collected
during the program. The TCDD equivalents for these
samples ranged from 0.774 to 1.47 ppt.
Operations
No major problems are reported in the unit operation
at the Tibbetts Road incineration. Since a pilot-scale
unit was used, operating experience is not applicable
to a commercial unit.
105
-------
-------
APPENDIX D-8
INTERNATIONAL PAPER PILOT-SCALE TESTS [16]
Introduction
During the period of Nov. 15 to 22, 1985, tests were
performed at the International Paper Co., Wood
Treatment Facility in Joplin, MO, for the purpose of
determining the ability and the operating conditions
required of the Shirco pilot-scale unit to meet the
EPA emissions and soil decontamination standards
for incineration of their creosote pit waste.
Wood preserving processes had been performed at
this plant, which used creosote and later
pentachlorophenol. Prior to RCRA, settling ponds
were used for waste water treatment. Nine settling
ponds comprised the water treatment operation. As a
result of the RCRA amendment, specifically Federal
Regulation .40 CFR 261.32, the presence of
pentachlorophenol and creosote designated the ponds
as hazardous waste sites. Consequently, the
International Paper Co. planned to clean up the site.
In an effort to acquire data to enable them to perform
the most cost-effective and permanent cleanup, the
pilot incineration test program was run on the site. A
total of 7 test runs were made over a 4-day period,
which included thermal processing and
accumulation of emissions and soil samples.
The primary objectives of the test program were to,
confirm the ability of the Shirco technology to
decontaminate creosote and pentachlorophenol
(PCP) laden soil and to incinerate the PGP at a
verified Destruction and Removal Efficiency (DRE)
of 99.9999%, and other Principal Organic Hazardous
Constituents (POHCs) at a DRE of 99.99% or
greater.
Feed Preparation
The waste materials processed during the test
program were pre-specified combinations of the
waste in Ponds 1 through 7 and the dewatered sludge
from the current wastewater treatment process.
Based on the results of a chemical analysis, test
blends were defined from a combination of the
individual pond wastes.
The goal of the International Paper Co. was to
prepare a blend, or a minimal number of blends,
which would maintain a steady and cost-effective
thermal process during the site cleanup. Conse-
quently, the 3 blends were chosen that would be
expected to demonstrate the realistic range of
operatingconditions. It was found that Pond 6
contained the highest levels of priority pollutants.
Test Mix 1 coupled this pond with the much-lower
contaminant level of Pond 2. The combination of
Ponds 4 and 5 suggested a median pollutant range.
Finally, the sludge from Ponds 1, 3, and 7 would have
a lower pollutant concentration. In order to decrease
the moisture content of the waste, a portion of the
dewatered sludge from the current process was
mixed with the waste. The proportions of each pond
waste comprising the blend were determined by the
International Paper Co. based on their projections of
relative percentages of each pond. Waste from each
pond was mixed in proportions of pond sludge and
pond dirt that eventually must be decontaminated.
The test mixes were blended from the following pond
waste combination:
Mix 1 — 1 part Pond 6, plus 1 part Pond 2, plus 2/3
part dewatered sludge.
Mix 2 — 1 part Pond 5, plus 1 part Pond 4, plus 2/3
part dewatered sludge.
Mix 3 — 4 parts Pond 7, plus 1 part Pond 3, plus 1
part Pond 1, plus 2 parts dewatered sludge.
In order to accomodate the feed system on the pilot-
scale unit, the above mixes also considered the waste
consistency and its ability to be fed to the unit with
minimum difficulty. The 3 blends were similar in
moisture content and adhesive qualities. However,
as the mix number increased, the adhesive
characteristics also increased. The viscosity of all the
mixes were high in that none would flow or slump.
107
-------
The Mix 1 blending was performed by combining the
percentages of Pond 6, Pond 2, and dewatered sludge.
However, large rocks and sludge lumps were not
removed as required. Subsequently, the blended mix
was classified using a hand-operated finger de-
lumper and a 3/8-in. hardware screen. The laborers,
attired in protective gear, removed the rocks, oper-
ated the delumper, and forced the sludge through the
screen by hand. Sludge was then put in 5-gal plastic
buckets, awaiting weighing and feeding to the
furnace.
Mix 2 and 3 blending was performed in essentially
the same manner. The components for the mixes
were staged in barrels on a slab approximately 50 ft
from the waste-water treatment building. The
laborers first acquired the defined proportions for
each mix from the staged barrels and transferred
them to the mixing area. Then small quantities of
each source components were alternately passed
through the hand delumper, which discharged into a
55-gal drum. Rocks were removed when found
during this delumping process. Then the mix in the
barrel was forced through a 3/8 inch hardware cloth
screen that removed rocks, broke sludge lumps, and
further homogenized the mix. Five-gal plastic pails
were filled with the discharge from the screening.
These pails were staged, as were those for Mix 1, for
weighing an'd subsequent thermal processing. Only
the amount of feed needed for testing was prepared
each day.
When needed for feed to the furnace, a pail of waste
was weighed on a platform scale. The scale was set
with the pail tare weight. The weight of the material
in the pail was recorded on the operation data log,
along with the time that feeding from that pail was
initiated.
Material to be processed was manually dumped
through a feed hopper onto a metering conveyor
located at the end of the furnace. The metering belt
was synchronized with the furnace conveyor to
control the material feedrate.
The feed metering conveyor for this furnace was
designed for non-adhering contaminated soil. How-
ever, with adequate preparation and monitoring, the
first sludge/soil mix fed to the furnace in a steady
manner. The second sludge mix tended to be more
adhesive and required constant attention to prevent
bridging and subsequent feed stoppage. A laborer
constantly monitored the feed and ended its passing
through the gate, as required. The third waste mix
tended to be more tar or batter-like. This presented
enough of a rate inconsistency to require the
cancellation of the emissions sampling during its
processing. Otherwise, the operation of the entire
system proceeded without difficulty throughout the
test program.
Test Procedure
Seven test runs were conducted. The process data for
the tests are given in Table D-8.1.
Previous testing performed on similar wastes had
suggested that the creosote and pentachlorophenol
contaminated waste could be decontaminated
successfully at a nominal PCC temperature of
l,600oF. Consequently, this temperature was also
used during this test program. The PCC Zone A,
(drying and initial volatilization) and Zone B (high
temperature volatilization and oxidation) were both
controlled at a setpoint temperature of l,650o and
l,600oF, respectively. The PCC nominal residence
times were chosen based on the furnace effective
length of 66.5 in.
The SCC temperature was chosen for each test
condition based on EPA guidelines and the results of
previous programs. The dependency of DRE on
process temperature was also examined. The
temperature during a specific test was adjusted
using fuel and input air flow. The starved air
combustion products from the primary chamber
provided additional fuel to the secondary chamber.
The feedrate of contaminated soil to the PCC was
controlled by the furnace- belt speed-setting and the
gap opening of the feed-conveyor guillotine-gate. The
speed of the feed conveyor and furnace conveyor belts
are synchronized. Both are driven by the same drive
jnotor and are geared accordingly. The guillotine-to- '
belt gap was 1.0 in. for Test 1 and 0.75 in. for Test 2.
It is estimated that the bulk density of the
contaminated soil was 70 Ib/ft3. The resulting feed-
rates for the 30- and 15-min residence times were 46
and 70 Ib/h respectively. However, the feed rate on
the second day was limited to 48.1 Ib/h to eliminate
potential clogging of the feed inlet.
Results
A summary of the demonstration tests at Inter-
national Paper at the Joplin, MO, site are presented
in Table D-8.2. The following discussion presents
summary results and conclusions. Specific operating
problems are also discussed.
Characteristics of the Feed
The 3 feed mixes contained lumps and rocks that
tended to jam at the metering gate. All mixes had
108
-------
Table D-8.1. Thermal Process Test Data Summary
I— 1
o '
CD
Date
11/18/85
11/19/85
11/20/85
1/20/85
1/21/85
1/21/85
1/21/85
Emissions
Test No.
1
2
3
4
5
6
7
Feed
Mix
1
1
1
1
2
3
Dewatered
Sludge
Primary
Chamber
Residence Time
min
26.8
25.66
15.02
26.1
15.05
15.13
15.07
Temperature, °F
Average
Feedrate
Ib/h
40.00
33.98
69.90
49.18
46.48
54.56
Not
Measured
Primary
Zone A
1,600
1,685
1,700
1,635
1,680
1,620
1,600
Primary
ZoneB
1,616
1,665
1,615
1,615
1,610
1 ,580
1,500
Secondary
Exhaust
2,130
2,200
2,195
1,800
2,000
1,980
2,125
Secondary
Chamber
Failed T/C
Failed T/C
2,140
1,755
1,950
1 ,930
2,070
Primary
• Chamber
Power Rate
kW
5.30
5.71
7.52
7.30
6.53
9.76 "
18.75
Secondary
Chamber Fuel
Ib Propane/h
6.00
3.27
6.66
2.14
2.63
8.87
N/A
T/C = Thermocouple
-------
Table D-8.2. Test Results Summary
Test No.
Average ORE, %"
Peolaehtorophenol ORE, %
Naphthalene (ORE), %
Paniculate emissions, gr/dscf
Average CO emissions, ppm
Ash organic concentration, ppb""
> 99.99906
> 99.99996
99.94076
0.020
114
73
> 99.99972
> 99.99998
99.99135
0.016
28
ND @20
> 99.99960
> 99.99999
99.99049
0.147
35
ND @30
> 99.99972
> 99.99998
99.99872
0.017
15
ND @30
>99.99914
> 99.99998
99.99872
0.070
18
ND @30
NC
NC
NC
NC
NC
ND @35
* AvoragQ ORE (or all organic constituents except naphthalene.
*Sum ol all organic constituents remaining in ash (i.e., none detected at 20 ppb)
NC • Stack sampling not conducted during Test 6.
adhesive tendencies. The second sludge mix tended
to be more adhesive and required constant attention
to prevent bridging and subsequent feed stoppage.
The third waste mix tended to be more tar or batter-
like.
The feed material was analyzed for moisture,
combustibles, and contaminant content, along with
density and heating value. Prior to the testing, a
brief analysis was performed on 3 approximate waste
compositions. These data are presented on Table D-
8.3 and became the basis for the initial process
operation settings.
Table D-8.3. Pretest Waste Analysis Data (% on as-received
basis)
Pond Nos.
(Composing
Waste)
5&2
3&7
4&6
Moisture"
24.9
26.3
29.9
% Volatile
25.3
18.0
25.8
High
Heating Value
(HHV) Btu/lb
4,500
2,800
6,200
% Moisture includes all weight lost by drying at 103°C.
Table D-8.4 presents an organic analysis of soil
samples for each test run for both hazardous
constituents and other organic compounds identified
in the samples by GC/MS. The highest concentration
of hazardous constituents consisted of benzo-
(a)anthracene (470-1,300 ppm), carbazole (1,700-
4,500 ppm), chrysene (720-2,200 ppm), fluoranthene
(110-14,000 ppm), naphthalene (91-2,600 ppm),
pentachlorophenol (4,600-12,000 ppm), and phenan-
threne (240-22,000 ppm). Analyses were not
conducted on the waste feed samples from Test 7.
Characteristics of the Furnace Ash
The residual organic concentration of each con-
stituent identified in the waste feed was nonde-
tectable in the furnace ash (detection limit ranging
from 20 to 40 ppb) for each run, with the exception of
the biphenyl (20 ppb) and naphthalene (53 ppb)
compounds in Test 1. It is not clear whether the
presence of these compounds is process related or due
to laboratory interferences. Nonetheless, the
reported concentrations were well below the level
required for ash delisting (i.e., approximately 1,000
ppb for naphthalene).
Organic Destruction and Removal
Efficiency (ORE)
The incinerator destruction and removal efficiency
for each constituent identified in the waste feed is
given in Table D-8.5 for Tests 1-5. Stack gas
sampling was not performed during Test 6 due to
sampling equipment problems.
The DREs for each test run exceeded RGRA perform-
ance standards of 99.99% for pentachlorophenol and
for all other POHCs with the exception of naph-
thalene. The DRE for naphthalene fell short of the
99.99% standard during Tests 1 and 5.
Naphthalene is a natural contaminant of XAD-2
resins and as such, should not be used to assess
system performance. The DRE for pentachlo-
rophenol, for instance, which is more difficult to
destroy than naphthalene, exceeded 99.9999%
during each test.
Particulate Emissions
For the first 5 tests, with the exception of Test 3,
particulate emissions ranged from 0.016 to 0.07
gr/dscf corrected to 7% C>2, as compared to the RCRA
standard of 0.08 gr/dscf. Particulate emissions
reported for Test 3 were 0.147 gr/dscf. The excessive
emissions were a result of soot formation caused by
an improper control of oxygen in the PCC. The stack
sampling contractor's oxygen monitor was not
functioning throughout the entire test program, and
Shirco operators were forced to set incinerator air
flow conditions purely by "ear". Given the fact that
incinerator operating conditions were adjusted
without the aid of flue gas (>2 monitoring, the overall
results were considered satisfactory.
110
-------
Table D-8.4.Waste Feed Analysis, ppm
Constituent
Test No. 3
Acenaphthene
Acenaphthylene
Anthracene
Benzo (a) anthracene
Biphenyl
Carbazole
Chrysene
Dibenzofuran
Dibenzothiophene
Fluoranthene
Fluorene
1 -Methylnaphthalene
2-Methylnaphthalene
Naphthalene
Pentachlorophenol
Phenanthrene
Pyrene
2,300
1,800
6,700
690*
ND
1,700
ND
1,200
480
5,500
2,200
850
1,100
129*
8,600
8,000
5,800
1,700
ND
4,600
650*
ND
1,700
870
1,100
750
4,100
2,000
1,700
2,800
1,500
6,800
7,500
4,000
1,700
ND,
4,600
470*
430
2,700
720*
760
900
4,000
2,400
2,500
4,100
2,600
12,000
8,000
2,200
4,200
ND
11,000
1,300
1,300
, 5,400
2,200
2,800
: ND
14,000
4,600
2,100
4,400
2,500
11,000
22,000
7,400
ND
ND
ND
ND
ND
ND
ND
ND
430
ND
ND
3,000
5,600
600"
6,800
2,100*
ND
40*
ND*
44*
ND
51
ND
ND
22
44
100*
40*
320
580
91*
4,600
240
87*
Detection limit (ppm)
Acids
Base Neutral
150
760
140
690
140
1,800
250
1,300
120
7,400
15
360
ND - None detected
Trace concentrations detected blow the average reporting (detection) limit.
Stack gas sampling was not performed during Test 6
due to sampling equipment problems, nor for Test 7,
which was conducted solely to determine furnace ash
quality.
Operations
The pilot-scale unit setup, dismantling, and decon-
tamination proceeded smoothly and in a timely
manner. Set-up was completed in 5 h. The
dismantling, decontamination, and packing for
transport were completed in 12 working hours with
the exception of difficulties encountered with the
feed system and 1 SCC thermocouple, which required
replacement. Otherwise, all of the pilot-scale unit
equipment operated well throughout the week.
The difficulty with the feed system was a direct
result of its mismatch with the feed material. Rocks
tended to jam at the metering gate, and at times in
the rotary airlock. The metering gate in the feed
conveyor allows material on the moving belt below it
to pass under and therefore spreads, levels, and
meters the feed into the furnace. This effect was
expected to work well on an expected drier feed
material. However, heavy rains during the previous
week, coupled with a more cohesive and adhesive
material than expected, did not allow unattended
feeding. With feed mix No. 2 (and to a greater extent
feed mix No. 3), bridging at the gate and adhesion to
the rotary airlock rotors was experienced. A
conventional Shirco leveling-screw and belt-
conveyor-type spreading/leveling system without a
rotary airlock should handle the type of waste
experienced without any problems.
Ill
-------
Tablo D-8.5. Flue Gas Destruction and Removal Efficiencies
Test No.
Samples vol., clscf
Stack flow, dscfm
Wasto feed, Ib/h
Constituent
Accnaphthene
AccnaplithylcnQ
Anlhraceno
Bonzo(a)anlhracene
Biphenyl
Carbazote
Chiysene
DibenzoJuran
Dibenzothiopene
FJuoranthona
Ftuorene
1 -Methylnaphthalene
2-Methylnaphthalene
Naphthalene
Pontachtorophonol
Phonanthreoe
Pyrcrne
Constituent not detected
1
42.94
115.38
40.0
> 99.99980
> 99.99985
> 99.99995
> 99.99948
•
> 99.99979
*
99.99926
> 99.99926
99.99968
> 99.99984
99.99686
99.99435
99.94076
> 99.99996
99.99956
99.99969
in waste feed.
2
116.10
80.68
34.0
ORE (%)
*
> 99.99994
> 99.99998
> 99.99983
*
> 99.99994
> 99.99988
99.99980
> 99.99986
> 99.99997
> 99.99995
99.99905
99.99807
99.99135
> 99.99998
99.99996
> 99.99997
3
122.22
106.44
69.9
-
> 99.99996
> 99.99998
> 99.99986
> 99.99985
> 99.99998
> 99.99991
99.99978
> 99.99993
> 99.99998
> 99.99997
99.99822
99.99682
99.99049
> 99.99999
99.99998
> 99.99997
4
86.48
119.26
49.2
-
> 99.99996
> 99.99998
> 99.99998
> 99.99846
> 99.99997
> 99.99993
99.99986
*
99.99997
> 99.99996
99.99905
99.99918
99.99872
> 99.99998
99.99996
> 99.99998
5
107.89
1 1 1 .43
46.5
«
ft
*
*
*
ft
ft
> 99.99973
*
ft
99.99824
99.99785
99.98482
> 99.99998
> 99.9999
ft
112
-------
APPENDIX D-9
TIMES BEACH PILOT-SCALE TESTS [17,18]
Introduction
The Times Beach contamination originated at the
Northeast Pharmaceutical and Chemical Co. plant
in Verona, MO, where 2,3,7,8-tetrachlorodibenzo-p-
dioxin was an unwanted byproduct from the
manufacture of disinfectants. The company paid a
waste-oil hauler to remove toxic sludge containing
the chemical. He mixed the fluid with waste oil that
he later sprayed on various sites around Missouri to
keep dust down. Those sites included four horse
arenas, gravel roads in Times Beach, a trailer park,
and his own farm.
The contaminated soil was later used as fill on
residential property in various places, where some
hot spots have shown contamination levels as high as
90 ppb. Rain eventually spread the fill onto other
property and into waterways. At least 150,000 tons
of dioxin-laced soil in Times Beach requires
treatment.
The Missouri Dept. of Natural Resources (MDNR)
decided in 1983 to set up a field test facility, where
any company that thought it had a good method for
treating dioxin could demonstrate their technology.
For a $16,500 fee to cover its costs, MDNR offered
developers an area at the site containing controlled-
quality contaminated soil that the agency would
sample before and after the treatment attempt.
During the period of July 5 through 12, 1985, the
Shirco Infrared Systems pilot-scale unit was onsite
at the Times Beach Dioxin Research Facility to*
demonstrate the Shirco technology capability to
successfully decontaminate soil laden with 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD). Equipment set-
up, preliminary operation, test operation, decontam-
ination, and disassembly was included in this period.
The test operation of the unit was conducted on July
10 and 11. The MDNR Environmental Div.
coordinated the site preparation. Shirco Infrared
Systems, Inc. prepared the test protocol and operated
the unit.
Test Procedure
The testing was planned so that adequate material
quantity would be processed and adequate samples of
emissions and soil be taken to demonstrate that the
soil and emissions decontamination and destruction
efficiency levels would be reached. Operations were
evaluated at 30-min and 15-min primary-
combustion-chamber residence times. A 7-h
emissions-and-soil-sample duration accompanied the
30-min residence time, and a 2-h and 22-min
duration was used for the 15-min residence time. The
unit feedrate averaged 47.68 Ib/h at a 1-in. bed depth
during the 30-min residence time exposure. The
feedrate during the 15-min residence time test
averaged 48.12 Ib/hr with a 0.75 in. bed depth.
Another important process, operating parameter was
temperature. Over the residence length of the PCC,
temperature was controlled in 2 equal length zones.
During the 30-min residence time test, the feed end
zone was maintained at a nominal temperature of
1,560°F and the discharge end zone was maintained
at a nominal 1,550°F. For the 15-min residence time
test, the respective temperatures were both 1,490°F.
The SCC was heated by a propane burner (compared
to the electric resistance heating elements used in
the PCC), and its temperature was maintained above
2,200°F during both tests. The nominal SCC
temperatures were 2,250° and 2,235°F, respectively.
The temperature of the exhaust gas leaving the wet
gas scrubber and discharging into the atmosphere
was nominally 165°F over the entire test duration.
Stack gas sampling of the pilot-scale unit during the
Times Beach dioxin destruction program was
conducted using the Modified Method 5 train.
Samples of the contaminated Times Beach soil were
collected from the feed hopper to the incinerator in
conjunction with each of the test runs. Three to four
113
-------
grab samples were collected during each run and
composited to a single sample for subsequent
analysis. Samples of furnace ash were collected
during the course of each test run by means of an
access port located in the ash hopper. Four to five
grab samples were collected during each run and
composited to a single sample for subsequent
analysis. Samples of the scrubber effluent water
were collected at 30-min intervals throughout each
test run. The samples from the recirculating system
were collected from the blowdown tank and
composited to provide a single, one-liter sample for
each run.
Results
Table D-9.1 provides a summary of the test results.
The contaminated soil samples used in the two tests
contained 230 and 155 ppb of 2,3,7,8-tetrachlo-
rodibenzo-p-dioxin (2,3,7,8-TCDD). No 2,3,7,8-TCDD
was detected in the furnace ash from either run at
detection limits of 0.038 ppb and 0.033 ppb.
Similarly, no 2,3,7,8-TCDD was detected in the stack
gas from either run at detection limits of 0.002 and
0.003 ng/m3. Based on these detection limits, the
demonstrated destruction and removal efficiency
(DRE) of the Shirco pilot-scale unit during the Times
Beach demonstration was >99.999996 and
> 99.999989%. No 2,3,7,8-TCDD was detected in the
scrubber slurry samples at a detection limit of 1 ppb.
Table D-9.1. Summary of Results
RCRA
Performance
Standard
30-min
Residence
15-min
Residence
Composite feed soil
2,3,7,8-TCDD concentration
Composts furnace ash
2,3,7,8-TCDD concentration
Paniculate emissions at 7% O2
Gas phase DRE of 2,3,7,8-TCDD
<1 ppb
0.08 gr/dscf
> 99.9999%
227 ppb
Not detected
at 38 ppt
0.001 gr/dscf
> 99.999996%
156 ppb
Not detected
at 33 ppt
0.0002 gr/dsc
> 99.999989%
114
-------
APPENDIX D-10
SIMULATED CREOSOTE PIT PILOT-SCALE TESTS [16]
During the week of Apr. 8, 1985, the Shirco pilot-
scale unit was used to incinerate a simulated
creosote-pit waste. The simulated material was 22%
creosote, 1% pentachlorophenol, 8% water, and 69%
soil. The simulated creosote feed analysis is
presented in Table D-10.1. The material was fed to
the PCC, which was maintained at 1,600° to 1,800°F;
the SCC was maintained at 1,800° to 2,200°F. The
operating conditions are shown in Table D-10.2. The
PCC was operated with no combustion air and no
added auxiliary electrical power.
Table D-10.1 .Simulated Creosote Feed Analysis, wt%
Test
No.
1
2
3
4
5
6
7
Creosote
24.04
20.65
24.54
22.95
22.20
21.55
24.52
Penta-
chlorophenol
1.29
0.80
0.89
0.81
0.85
0.96
0.92
Water
7.06
7.93
7.41
7.67
7.71
7.67
7.39
Inert
(dry soil)
67.61
70.62
67.16
68.57
69.24
69.82
67.17
Resulting particulate emissions rates were between
0.007 and 0.012 gr/dscf corrected to 7% oxygen; this
concentration is significantly below the RCRA
performance standard of 0.08 gr/dscf. The calculated
destruction and removal efficiencies (DREs) of the
principal organic hazardous constituents (POHC)
were at or above the RCRA performance standard of
99.99%, except where the POHCs were below the
detectable limit. The pentachlorophenol was
identified as being the most difficult feed component
of the POHCs to destruct. The test results indicated
that the pentachlorophenol was below the detection
limit in both the stack gas and furnace ash analyses.
The resulting DREs calculated at the detection limit
were greater than 99.99% in every case and as high
as 99.99986%. The furnace ash analysis and DREs
for the 15 identified POHCs are reported in Tables D-
10.3 and D-10.4.
Table D-10.2.Simulated Creosote Waste Incineration Operating Conditions
Bed
Test Thickness
No. in.
1 1.0
2 1.0
3 0.5
4 1.5
•• 1.0
1.0
7 1.0
Solid Phase
Residence Time
min
25
45
25
25
15
25
25
Solid
Feed rate
Ib/h
42.4
20.4
40.6
58.3
121.0
56.4
61.9
Temp.
Zone A
°F
1,625
1,632
1,634
1,635
1,612
1,615
1,818
Temp.
Zone B
°F
1,672
1,606
1,691
1,867
1,725
1,658
1,883
Temp.
Bed Chamber
°F
2,200
2,166
2,195
2,210
2,189
1,810
2,220
115
-------
Tablo D-10.3. Furnace Ash Analysis (ppm)
Run No.
POHCs:
1
1 Pcntachtorophenol ND ND ND ND ND ND
2 Pbonoi ND ND ND ND ND ND
3 2,4-Dimethylphenol ND ND ND ND ND ND
4 lrKteno(1,2,3"CD)Pyrene ND ND ND 0.547 0.03 0.019
S Bonzo (B) & (K) Fluoroanthene 0.057 ND ND 0.56 0.12 ND
6 Benzo (A) Pyrene 0.032 ND ND 0.472 0.137 ND
7 Bonz (A) Anthracene/Chrysene 0.495 ND 0.045 ND ND 1.213
8 Naplhatene 2.36 0.036 3.447 1.438 0.248 0.454
9 Acenaphthene 0.014 ND ND 0.006 ND ND
10 AceoaphUiytene 0.017 ND ND 0.011 ND ND
11 Fluoreno ND ND ND 0.005 ND ND
12 Anthracene/Phenanthrene 10.79 0.092 2.425 7.037 0.403 4.5
13 Fluoranthene 5.992 0.0763 0.711 6.795 0.362 5.058
14 Bonzo (GHI) Perytene ND ND ND 0.08 0.026 0.008
15 PyrOOQ 2.133 0.013 0.173 2.458 0.164 1.348
NO - Not detectable
116
-------
Table D-10.4 Destruction and Removal Efficiencies, %
Run No. •
POHCs:
1 Pentachlorophenol
2 Phenol
3 2,4-Dimethylphenol
4 Indeno (1,2,3-CD) pyrene
5 Benzo (B) & (K) fluoroanthene
6 Benzo (A) pyrene
7 Benz (A) anthracene/chrysene
8 Napthalene
9 Acenaphthene
10 Acenaphthylene
1 1 Fluorene
1 2 Anthracene/phenanthrene
13 Fluoranthene
14 Benzo (GHI) perylene
15 Pyrene
1
> 99.99893
99:99119
99.99731
> 99.85641
99.99976
> 99.97370
99.99987
99.99678
99.99990
99.99925
99.99992
99.99977
99.99968
> 99.38346
99.99984
2
> 99.99614
> 99.99532
99.99776
> 99.92262
99.99987
> 99.98583
99.99993
99.99929
99.99996
99.99948
99.99993
99.99987
99.99986
> 99.66776
99.99993
3
> 99.99834
> 99.99775
99.99968
> 99.96295
99.99994
> 99.99321
99.99997
99.99966
99.99998
99.99975
99.99996
99.99994
99.99993
> 99.84092
99.99996
4
> 99.99986
> 99.99820
99.99978
> 99.97037
99.99995
> 99.99457
99.99992
99.99886
99,99983
99.99955
99.99974
99.99877
99.99750
> 99.87280
99.99951
5
> 99.99945
> 99.99929
> 99.99989
> 99.98833
> 99.99998
> 99.99786
> 99.99999
99.99976
99.99995
99.99990
99.99993
99.99984
99.99972
> 99.94991
99.99995
6
> 99.99902
> 99.99858
> 99.99979
> 99.97662
> 99.99996
> 99.99571
> 99.99998
99.99961
99.99995
> 99.99988
99.99995
99.99990
99.99983
> 99.89962
99.99995
7
> 99.99703
99.99801
99.99971
> 99.97330
> 99.99995
> 99.99511
99.99984
99.99596
99.99924
99.99859
99.99880
99.99623
99.99675
> 99.88535
99.99823
> - DREs were calculated at detection' limits.
-------
-------
REFERENCES FOR APPENDICES
1. USEPA, 1988. Technology Evaluation Report,
SITE Program Demonstration Test, Shirco
Infrared Thermal Incineration System, Peak
Oil, Brandon, FL EPA/540/5-88/0020. EPA Risk
Reduction Engineering Laboratory, Office of
Research and Development, Cincinnati, OH
45268.
2. USEPA, 1989. Technology Evaluation Report,
SITE Program Demonstration Test, Shirco
Pilot-Scale Infrared Incineration System, Rose
Township Demode Road Superfund Site. EPA
Risk Reduction Engineering Laboratory, Office
of Research and Development, Cincinnati, OH
45268.
3. Vendor information from ECOVA Corp., Dallas,
TX, 1989.
4. Final Report: Demonstration Test - Onsite PCB
Destruction - Shirco Infrared Portable Unit at
Florida Steel Indiantown Mill Site, Indiantown,
FL Report Number 821-86-1, ECOVA Corp.,
Dallas, Tex., Sept. 17,1986.
5. Technical Paper: Remediation of PCB-
Contaminated Soils by Mobile Infrared
Incineration. G. J. McCartney and J. E. Burford,
OH Materials Corp., Findlay, OH, 1988.
6. Draft of USEPA TSCA Permit to OH Materials
Corporation to Dispose Polychlorinated
Biphenyls (PCBs), OH Materials, Findlay, OH,
1988.
7. Telephone communication between Sy
Rosenthal, Foster Wheeler Enviresponse, Inc.
and George Hay, OH Materials Corp., Findlay,
OH, Mar. 14,1989.
8. Thermal Destruction Unit Demonstration Test
Plan for LaSalle Electric Utilities, PCB Abate
ment Project, Illinois EPA, Springfield, IL, Dec.
1988.
9. Remedial Investigation Report, LaSalle Electric
Utilities Site, Illinois EPA, Springfield, IL,
1988.
10. Illinois Environmental Protection Agency
Operating Approval to Westinghouse
HAZTECH, Inc. for use of Thermal Destruction
Unit at LaSalle Electric Utilities PCB
Abatement Project, Illinois EPA, Springfield,
IL, Nov. 23,1988.
11. Summary of Remedial Alternative Selection,
LaSalle Electric Utilities, LaSalle, IL, Illinois
EPA, Springfield, IL, Mar. 29,1988.
12. Telephone communication between Sy
Rosenthal, Foster Wheeler Enviresponse, Inc.,
and Richard Lange, Illinois EPA, Springfield,
IL, Mar. 30,1989.
13. Demonstration Test Report: PCB Destruction
Unit, Shirco Portable Demonstration Unit -
Infrared Incineration - Twin Cities Army Am-
munition Plant, ECOVA Corp., Dallas, TX.,
Oct. 23,1987.
14. Final Report: Onsite Incineration Testing at
Brio Site, Friendswood, TX, Shirco Infrared
Systems Portable Test Unit, Report No. 846-87-
1, ECOVA Corp., Dallas, TX, Feb. 10-13,1987.
15. Final Report: Onsite Incineration Testing of
Tibbetts Road Superfund Site, Barrington, NH,
Shirco Infrared Systems Portable Test Unit,
Report No. 834-87-1, ECOVA Corp., Dallas, TX,
Nov. 2,1987.
16. Final Report: Onsite Incineration Testing of
Shirco Infrared Systems Portable Pilot Test
119
-------
Unit at International Paper Co. Wood 18.
Treatment Plant, Joplin, MO, Report No. 804-
86-2, ECOVA Corp., Dallas, TX, May 29,1986.
17. Summary Report: Onsite Incineration Testing
of Times Beach Dioxin Research Facility, Times
Beach, MO by Shirco Infrared Systems Portable 19.
Test Unit, July 8-12,1985, Report No. 815-85-1,
ECOVA Corp., Dallas, TX., Aug. 28,1985.
Environmental Performance Evaluation of the
Shirco Infrared Incinerator, Dioxin Destruction
Demonstration Program, Times Beach, MO,
ERT Report No. C-3044-D-1, ECOVA Corp.,
Dallas, TX, Aug. 29,1985,
Results of the Shirco Infrared Incinerator
Synthetic Hazardous Waste Tests, ECOVA
Corp., Dallas, TX, May 7,1985.
120
------- |