United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
/"3
Research and Development
EPA-600/S2-81-022 July 1981
Project Summary
Guidelines for the Disposal of
PCBS and PCB Items by
Thermal Destruction
D G Ackerman and R Scofield
This is a resource and guidelines
report intended to aid the U.S. Environ-
mental Protection Agency Regional
Offices in evaluating facilities which
apply for approval of thermal destruc-
tion of polychlorinated biphenyls
(PCBs). Annex I incinerators and high
efficiency boilers (40 CFR 761) are
emphasized.
Information is provided on: funda-
mental combustion processes as they
relate to incineration efficiency and
the formation of organic compounds
not originally present; thermal destruc-
tion technologies that have been or
could be used for disposal of PCBs;
thermal destruction tests on PCBs and
other relevant materials; and a descrip-
tion and evaluation of sampling and
analysis methods for PCBs.
Guidelines and criteria are provided
for: interpreting those parts of the
PCB Regulations governing Annex I
incinerators and high efficiency boilers;
establishing the consistency of disposal
operations with the regulations; eval-
uating Annex I incinerators and high
efficiency boilers: Initial Reports,
Notifications, Trial Burn plans, and
Trial Burn data; issuance of approvals
and monitoring compliance; and facil-
itating coordinated Agency review of
PCB disposal operations.
This Project Summary was devel-
oped by EPA's Industrial Environmental
Research Laboratory. Research Tri-
angle Park, NC. to announce key find-
ings of the research project that is fully
documented in a separate report of the
same title (see Project Report ordering
information at back).
Introduction
Polychlorinated biphenyls (PCBs) are
derivatives of the compound biphenyl in
which from 1 to 10 hydrogen atoms
have been replaced with chlorine atoms.
PCBs are characterized by extremely
high thermal and chemical stability.
These characteristics made PCBs highly
useful in a wide variety of commercial
applications. The wide use of PCBs
coupled with a lack of recognition of
their hazards led to the present ubiquitous
distribution of PCBs in the environment.
EPA estimated that up to 1975, between
136,000 and 181,000 metric tons (300
to 400 million pounds) of PCBs had
entered the environment.
Although PCBs have low acute toxic-
ities, other adverse effects have been
found in humans, laboratory animals,
and other organisms. There is evidence
that PCBs are carcinogenic, oncogenic,
and teratogenic. PCBs are also known to
bioaccumulate and biomagnify.
In 1976, growing evidence of the
problem of PCB contamination led to an
inclusion in the Toxic Substances Con-
trol Act (TSCA) of a provision that would
require the eventual elimination of PCB
usage in the United States. The PCB
Regulations (40 CFR 761), promulgated
under TSCA do not require removal of
PCBs and PCB Items from service earlier
than would otherwise be required; but
when PCBs and PCB Items are removed
from service, disposal must be in ac-
-------
\
cordance with the PCB Regulations.
Other acts which govern the disposal of
PCBs include the Resource Conservation
and Recovery Act; the Clean Water Act;
the Marine Protection, Research, and
Sanctuaries Act; and the Occupational
Safety and Health Act.
Large amounts of PCBs are in service
and will eventually require disposal. The
Electric Power Research Institute pro-
jects a shortfall of utility- waste PCB
disposal capacity (landfill and incinerator)
in most EPA regions after January 1,
1981 (EPRI, 1980). Additionally, there
will be a smaller quantity of PCB wastes
from commercial and industrial uses
which will require disposal.
The full report is a resource and
guidelines document. It is intended to
aid EPA Regional Offices in evaluating
facilities which apply for review and/or
approval of thermal PCB destruction
activities.
Thermal Destruction
of PCBs
The. physical form of the PCB waste
(i.e., liquid or solid) is the most important
factor influencing the mechanisms of its
combustion. Other important physical
properties are viscosity, solids content,
and moisture content. The fuel's chemi-
cal properties also are important and
include its elemental composition and
heating value. The efficiency of destruc-
tion for a given fuel will depend upon
unit operating parameters: combustion
air, waste feed rate, temperature, resi-
dence time, and mixing.
Thermal Destruction Systems
The thermal destruction processes
described in this report can be grouped
into one of two general categories:
Annex I incinerators or high efficiency
boilers. Table 1 lists the incineration
systems which were cited most often in
the literature for disposal of hazardous
wastes. Not all of these systems are
suitable for the thermal destruction of
PCBs. PCBs have been successfully
incinerated in rotary kilns (coupled with
afterburners) and liquid injection in-
cinerators. Also, several successful
tests have been performed in cement
kilns. A test in a municipal sludge
incinerator was partially successful but
demonstrated a need for a change in the
method of feed.
In addition to incinerators, which are
designed to combust hazardous mate-
rials, conventional boilers also can be
used to destroy PCBs if proper combus-
Table 1. Incineration System Summary
Type Applicability
Reasons
Rotary kiln Liquids - Yes
Solids - Yes
Liquid
injection
Multiple
hearth
Liquids - Yes
Solids - No
Liquids - Yes
Solids - Yes
Fluidized Liquids -
bed Potential
Solids - No
Multiple Liquids - Yes
chamber So/ids - No
Catalytic Liquids - N6
combustion Solids - No
Pyrolysis
Liquids - No
So/ids - No
Starved air Liquids - No
combustion Solids - No
Molten salt
Liquids -
Potential
Solids - No
Best system for PCB Items.
Requires an afterburner.
Best system for liquid PCBs. May be used as
afterburner following a so/ids incinerator.
Design is suitable, but may require operation at
higher temperatures than currently used. Afterburner
required. Scrubber may be required.
Future consideration warranted for liquids. Hard to
remove shredded noncombustibles. Temperatures
probably too low.
Liquids may be suitable if liquid burner is installed
above grates. Insufficient air/solid mixing.
Primarily designed for gases and vapors.
High probability of forming toxic combustion products.
High probability of forming toxic combustion products.
Future consideration warranted for liquids. Not
proven commercial technology. Temperatures
probably too low. Hard to remove shredded
noncombustibles.
tion conditions are maintained. The PCB
Regulations permit, under certain con-
ditions, burning of PCB liquids with
concentrations in the range of 50 to 500
ppm in boilers with at least a 50 million
Btu/hr heat input rate. PCB liquids
below 50 ppm are not subject to the
disposal regulations. PCB liquids above
500 ppm must be disposed of in Annex I
incinerators.
Sampling Methods
For the initial use, or after substantial
modification of a facility for burning
PCBs, the stack gas may be monitored
for carbon monoxide (CO), carbon dioxide
(CO2), oxygen (O2), hydrogen (H2), nitro-
gen oxides (NOx), PCBs, organochlorines
(RCLs), hydrogen chloride (HCI), and
total particulate matter (TPM). Scrubber
effluent and solid residues also may be
monitored. Other process parameters
that may be measured are PCB and
auxiliary fuel feed rates and combustion
zone temperatures.
Carbon monoxide and oxygen must be
measured continuously and maintained
in accordance with the PCB Regulations.
Sampling for PCBs, RCL, HCI, NOX, and
TPM are not specified in the PCB Regu-
lations and are subject to the require-
ments of the cognizant EPA Regional
Administrator. During the four recent
Trial Burns, from four to five stack gas
samples (4-hour duration) were required.
Variable numbers of samples for RCLs,
HCI, TPM, and NOX also were required.
Sampling intervals for scrubber effluent
and other process influent and effluent
streams have ranged from 15- to 30-
minute intervals (Rollins, 1980; EPRI,
1980; Tennessee Eastman, 1979; and
Zelenski and Haupt, 1979).
A critical factor in all sampling activi-
ties is obtaining a representative sample
because the accuracy and precision of
an analytical result can be no greater
than the accuracy and precision of the
sampling. EPA (1980) has described in
detail (Method 5, 40 CFR 60) how to
obtain representative samples of par-
ticulate and gases in stacks and ducts.
Many of the sampling methods described
in this section are based on modified
Method 5 sampling trains. Methods for
sampling aqueous streams are described
in: (Grant, 1978; and "Standard Meth-
ods," 1975). The American Society for
Testing and Materials (1978) gives a
procedure for obtaining representative
samples of bulk solids. EPA recently
published a manual (EPA, 1980) for
-------
sampling and analysis for hazardous
materials.
Gas sampling methods (stack or
ambient air) can be grouped into three
general categories: 1) liquid absorption
systems, 2) liquid phase absorbents
coated on solid supports, and 3) solid
adsorbents. For sampling PCBs in stack
gas from Annex I incinerators, use of a
Florisil-trap-modified Method 5 train is
recommended, as described in EPA's
interim sampling and analysis manual
(Beard and Schaum 1978). For other
RCLs and organics, use of a Method 5
train modified with an XAD-2 trap is
recommended. Tenax adsorption tubes
have also been used. If ambient air
monitoring for PCBs is to be performed
during a PCB Trial Burn, it is recom-
mended that a high volume sampler
incorporating polyurethane foam be
used.
There are suitable commercial instru-
ments for continuous monitoring of 02,
CO, and C02. The reference method for
NO, is Method 7 (40 CFR 60), but there
are equivalent near-continuous instru-
mental methods. Hydrogen chloride and
TPM are sampled by a modified Method
5 technique.
Liquids are often taken by single "grab
samples"; however, these samples may
not be representative of a stream during
process variations, and thus composite
sampling is appropriate. Modified Mine
Safety Appliances Model 5 samplers
have been used when personnel or
work space monitoring for PCBs or RCLs
is necessary.
Analytical Methods
Data on airborne PCB levels and on
PCB emissions from combustion sources
are limited, partly because the assess-
ment and interpretation of PCB emission
data is complicated. One of the compli-
cating factors is the fact that there are
209 possible PCB isomers, and most
PCBs were made and marketed as
mixtures. Interpretation is also compli-
cated by the fact that the combustion
process alters the relative amounts of
isomers, and thus ambient air.samples
tend to be enriched in higher volatility
PCBs and deficient in the lower volatility
PCBs. Interferences from compounds
(e.g., pesticides) which exhibit analytical
behavior similar to PCBs also can make
the interpretation of results difficult.
The most commonly used analytical
method for determining PCBs is gas
(chromatography (GC) which separates
compounds in the vapor phase. After
separation, various detectors are used
to measure compounds. The most widely
used detector is the electron capture
detector (ECD) although flame ionization
detectors (FID) and mass spectrometry
(MS) can also be used. Gas chroma-
tography-mass spectrometry (GC/MS)
is generally used for confirmatory pur-
poses.
For quantifying PCBs, three general
methods can be used individually or in
combination. Pattern recognition is the
most common method and involves
comparing the areas under a multipeak
chromatographic elution pattern gener-
ated by a sample to the areas under the
elution pattern generated by a PCB
mixture of known concentration. An-
other method, derivatization, involves
converting all PCB isomers in a sample
to the fully chlorinated isomer, deca-
chlorobiphenyl (DCB), and then using
ECD or GC/MS to measure the amount
of DCB present. Quantification also can
be achieved by the measurement of
individual isomeric components using
GC/ECD or GC/MS.
Several different analytical methods
for the detection of PCBs are described
in the literature, but it is recommended
that the methodology specif ied in EPA's
interim manual (Beard and Schaum,
1978) be employed for analysis of
samples taken during PCB Trial Burns.
Results are reported in terms of nano-
grams (ng) DCB per dry standard cubic
meter of combustion effluent sampled.
To verify the presence of PCBs, GC/MS
and pattern matching with Aroclor
mixtures is used. The precision of the
DCB analysis is stated to be 10 to 15
percent, and recovery of PCBs through
the entire sampling and analysis proce-
dure is stated to be 85 to 95 percent.
Combustion of PCBs
The purpose of incinerating PCB-
containing wastes is to destroy the
PCBs so effectively that any emissions
of undestroyed PCBs to the environ-
ment will be at such low concentrations
that adverse environmental and health
impacts are not expected to occur. Five
combustion process parameters have
major impacts on the destruction and/
or formation of polycyclic organic matter
(POM) such as PCBs: 1) reaction tem-
perature within the combustion and
post-combustion zones, 2) residence
time of reactants (air and fuel) and
products in the high temperature zone,
3) turbulence or mixing efficiency of fuel
and air, 4) air/fuel ratio including the
effects of operating cycles on combus-
tion air supply, and 5) fuel feed size.
Thermal destruction tests on PCBs
indicate that essentially complete de-
struction occurs in well designed in-
cineration systems. The most important
cause of incomplete combustion of fuel
is lack of turbulence or incomplete
mixing of fuel, air, and combustion
products.
The potential for formation of poly-
chlorinated dibenzofurans (PCDFs) and
dibenzo-p-dioxins (PCDDs) during
thermal destruction of PCBs can be
examined by thermochemical equilibrium
calculations. Because accurate thermo-
dynamic data are unavailable for these
compounds, the thermodynamic feasi-
bility of formation of PCDFs and PCDDs
was examined indirectly by investigating
combustion conditions and waste types
that would favor the formation of inter-
mediates, such as chlorobenzenes and
chlorophenols. The calculations predict
that, under oxidizing conditions, forma-
tion of PCDFs and PCDDs is not thermo-
dynamically favored. However, under
pyrolytic conditions (i.e., the absence or
near absence of oxygen), as may arise in
an inadequately designed or operated
incinerator, thermochemical equilibrium
calculations indicate that trace amounts
of possible precursors to PCDFs and
PCDDs can form.
Theory and experiment indicate that
essentially complete destruction of
PCBs can be achieved in both Annex I
incinerators and high efficiency boilers.
However, inadequate design or opera-
tion can lead to inadequate destruction
efficiency and/or formation of highly
toxic compounds such as PCDFs and
PCDDs.
Annex I Incinerators
One disposa I option for PCBs and PCB
Items is thermal destruction in an
approved incinerator. In general, incin-
erators must meet specific operating
requirements set forth in Annex I of the
PCB Regulations. An incinerator which
meets these requirements or which,
based on evidence, is capable of operat-
ing without presenting unreasonable
health or environmental risks from
PCBs when one or more of the Section
761.40 requirements is not met, is
known as an Annex I incinerator.
The operator of an incinerator is
required to obtain written approval from
the cognizant EPA Regional Administra-
tor prior to incineration of PCBs or PCB
Items. The requirements for approval of
a US GOVERNMENT PRINTING OFFICE 1961 -757-012/7229
-------
incinerators for PCB disposal are stated
in the PCB Regulations (40 CFR 761.40).
The EPA Regional Administrator also
may prescribe any additional require-
ments which are necessary to satisfy
the intent of the regulations. For ex-
ample, incinerator design and opera-
tional criteria are very broad, and more
definitive restrictions may need to be
imposed in an approval. The Regional
Administrator also may want to waive
existing requirements or specify addi-
tional requirements in the areas of
monitoring and recordkeeping, sampling
and analysis, restrictions on waste
composition, or compliance assurance.
High Efficiency Boilers
One disposal option for liquids con-
taining 50 ppm or greater of PCBs (but
less than 500 ppm) is thermal destruc-
tion in a high efficiency boiler. In
general, high efficiency boilers must
meet specific design and operational
requirements set forth in the PCB Regu-
lations (40 CFR 761.10).
The notification and approval process
for high efficiency boilers depends on
the type of liquid being burned. Inciner-
ation of mineral oil dielectric fluid from
PCB-contaminated transformers only
requires that the cognizant EPA Regional
Administrator be given written notice
30 days prior to the burn. Thermal
destruction of other liquid wastes con-
taining from 50 to 500ppm PCBs in high
efficiency boilers requires written notice
as well as authorization by the EPA
Regional Administrator.
References
American Society for Testing and Mate-
rials. 1978. Standard Method of Sam-
pling Coke for Analysis. Part 26. p. 213-
ff.
Beard, J.H., III, and J. Schaum. 1978.
Sampling Methods and Analytical Pro-
cedures Manual for PCB Disposal:
Interim Report. U.S. EPA, Office of Solid
Waste.
EPA. 1980. Test Methods for Evaluating
Solid Waste. Physical/Chemical Meth-
ods. EPA Report No. SW-846.
EPRI. 1980. Disposal of Polychlorinated
Biphenyls and PCB-Contaminated
Materials. Volume 4: Test Incineration
of Capacitors Containing PCBs. Electric
Power Research Institute. Report FP-
1207, Volume 4.
Grant, D.M. 1978. Open Chemical Flow
Measurement Handbook. Instrumenta-
tion Specialties Company.
Rollins. 1980. The PCB Incineration
Test Made by Rollins Environmental
Services (TX) at Deer Park, Texas, m
November 12-16, 1979. ^
Standard Methods. 1 975. American
Public Health Association - American
Water Works Association - Water Pol-
lution Control Federation. Standard
Methods for the Examination of Water
and Wastewater. 14th Edition. Wash-
ington, DC.
Tennessee Eastman. 1979. Destruction
of a Dilute PCB-Contaminated Waste in
a Coal-Fired High Efficiency Boiler at
Tennessee Eastman Company.
Zelenski, S.G. and S. Haupt. 1979.
Evaluation of PCB Destruction Efficiency
in Industrial Boilers. Draft Final Report
to EPA, Contract No. 68-02-2607, Task
33. (In final form, Applying for a Permit
to Destroy PCB Waste Oil, Volumes I
(Summary) and II (Documentation),
EPA-600/7-81-033a and -033b, March
1981.)
D. G. Acker man and R. Scofield are with TRW, Inc., Redondo Beach, CA 90278.
David C. Sanchez is the EPA Project Officer (see below).
The complete report, entitled "Guidelines for the Disposal of PCBs and PCB
Items by Thermal Destruction." (Order No. PB 81-182 339; Cost: $23.00,
subject to change) will be available only from.
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
I '.00 03; i,
5 r ?"i?rK i-POTt.C i
'
* $0 6 u
CHIC*'*,)
iv r
Sir?,,
-
------- |