£EPA
United States
Environmental Protection
Agency
EPA-600/R-04-024
February 2004
       The Use of
       Surrogate Compounds as
       Indicators of PCDD/F
       Concentrations in
       Combustor Stack Gases
                           «=*

-------
                                      EPA-600/R-04/024
                                      February 2004
 The Use of Surrogate Compounds as
Indicators of PCDD/F Concentrations in
         Combustor Stack Gases
            Annual Performance Measure 140
              Goal 5 Waste Management
                   Prepared by

                  Paul M. Lemieux
        United States Environmental Protection Agency
           Office of Research and Development
            Research Triangle Park, NC 27711
                   Prepared for

        United States Environmental Protection Agency
           Office of Research and Development
                  Washington, DC

-------
                                    Abstract


Emissions of polychlorinated dibenzo-p-dioxins and poly chlorinated dibenzofurans (PCDDs/Fs)
from stationary combustion sources are of concern due to their carcinogenicity and endocrine
effects. PCDDs/Fs are typically present only in minute concentrations in combustor stack gases,
which makes  sampling and analysis of these compounds extremely expensive and time con-
suming. Direct, real-time measurement of all PCDD/F isomers of concern is not possible using
current technology. It is possible, however, to estimate stack concentrations of PCDDs/Fs by
measuring other indicator (surrogate) compounds that are present in the stack gases at much
higher concentrations. Appropriately selected surrogate compounds would be easier to measure
than PCDDs/Fs and can be measured in real-time or near real-time by stack gas monitoring
technology that is currently available either commercially, or in various stages of development.
This report discusses the various surrogates that can be used to indicate PCDD/F concentrations,
how those surrogates can be measured, and  a state-of-the-art assessment of availability and
effectiveness of analyzers for measuring those compounds.

-------
                                      Foreword
    The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the
Nation's land, air, and water resources. Under a mandate of national environmental laws, the Agency
strives to formulate and implement actions leading to a compatible balance between human activities and
the ability of natural systems to support and nurture life. To meet this mandate, EPA's research program
is providing data and technical support for solving environmental problems today and building a science
knowledge  base necessary to manage our ecological resources wisely, understand how pollutants affect
our health,  and prevent or reduce environmental risks in the future.

    The National Risk Management Research Laboratory (NRMRL) is the Agency's center for
investigation of technological and management approaches for preventing and reducing risks from
pollution that threaten human health and  the environment. The focus of the Laboratory's research
program is  on methods and their cost-effectiveness for prevention and control of pollution to air, land,
water, and  subsurface resources; protection of water quality in public water systems; remediation of
contaminated sites, sediments and ground water; prevention and control of indoor air pollution; and
restoration  of ecosystems. NRMRL collaborates with both public and private sector partners to foster
technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's research
provides solutions to environmental problems by: developing and promoting technologies that protect
and improve the environment; advancing scientific and engineering information to support regulatory and
policy decisions; and providing the technical support and information transfer to ensure implementation
of environmental regulations and strategies at the national, state, and community levels.

    This publication has been produced as part of the Laboratory's strategic long-term research plan. It
is  published and made available by EPA's  Office of Research and Development to assist the user
community and to link researchers with their clients.
                                         Lee A. Mulkey, Acting Director
                                         National Risk Management Research Laboratory
                                             in

-------
                           EPA Review Notice
This report has been peer and administratively reviewed by the U.S. Environmental Protection
Agency and approved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Information Service,
Springfield, Virginia 22161.
                                       IV

-------
                                   Contents

Section                                                                      Page
Abstract  	  ii
List of Figures 	vi
List of Tables	  vii
Glossary	viii
Acknowledgments	x

1.0 Introduction 	1-1
   1.1 Emissions of PCDDs/Fs	1-1
   1.2 Formation of PCDDs/Fs  	1-1
   1.3 Current Regulatory Approach	1-2
   1.4 Sampling and Analytical Methods	1-3
   1.5 Surrogate Concept	1-4
2.0         Approaches 	2-1
   2.1 Measurement of Carbon Monoxide	2-1
   2.2 Measurement of Total Hydrocarbons  	2-2
   2.3 Measurement of Volatile Organic Compounds	2-3
   2.4 Measurement of Chlorobenzenes and Chlorophenols	2-6
   2.5 Measurement of Poly chlorinated Biphenyls 	2-9
   2.6 Measurement of Poly cyclic Aromatic Hydrocarbons  	2-9
   2.7 Measurement of Lower Chlorinated Dioxins and Furans  	2-10
3.0         Analytical Techniques	3-1
   3.1 Continuous Emission Monitors for Carbon Monoxide	3-1
   3.2 Continuous Emission Monitors for Total Hydrocarbons	3-1
   3.3 Continuous Emission Monitors for Volatile Organic Compounds  	3-1
   3.4 Continuous Emission Monitors for Chlorobenzenes and Chlorophenols	3-2
   3.5 Continuous Emission Monitors for Polycyclic Aromatic Hydrocarbons 	3-3
   3.6 Continuous Emission Monitors for Lower Chlorinated Dioxins and Furans 	3-4
4.0 Conclusions 	4-1
5.0           	5-1
                                        v

-------
                             List of Figures
Figure                                                                  Page
1-1   Formation Pathways of PCDDs/Fs and Other High MW Pollutants  	  1-2
1-2   Relative Concentrations of PCDDs/Fs and Their Surrogates	  1-4
2-1   CO vs. Total PCDD/F Stack Emissions in an RDF Combustor	  2-2
2-2   THC vs. Total PCDD/F Stack Emissions in an RDF Combustor	  2-3
                                     VI

-------
                              List of Tables
Table                                                                   Page
1-1   PCDD/F Emission Limits from Waste Combustors	  1-3
2-1   Regression Results of C2 Chloroalkenes vs. Total PCDDs/Fs	  2-5
2-2   List of CBz and CPh I-TEQ Surrogates	  2-8
2-3   Lower Chlorinated CDDs/Fs vs. I-TEQ	 2-11
3-1   Techniques for Rapid Measurement of VOC PICs	  3-2
3-2   Techniques for Rapid Measurement of CBz and Cph 	  3-3
                                     vn

-------
                                 Glossary
    Term
Definition
APCS           air pollution control system
BIFs            boilers and industrial furnaces
CBz            chlorobenzenes
CEMs           continuous emission monitors
Cl              chlorine
CO             carbon monoxide
CO2            carbon dioxide
CPh            chlorophenols
DOAS          differential optical absorption spectroscopy
ESP            electrostatic precipitator
FID            flame ionization detector
FTIR           Fourier transform infrared
GC             gas chromatography
HWCs          hazardous waste combustors
IMS            ion mobility spectroscopy
I-TEQ           international toxic equivalency
MCBz          monochlorobenzene
MS             mass spectrometry
MW            molecular weight
MWCs          municipal waste combustors
NDIR           non-dispersive infrared
NITEP          National Incinerator Testing and Evaluation Program
O2              oxygen
PAHs           polycyclic aromatic hydrocarbons
PCBs           polychlorinated biphenyls
PCDDs/Fs       polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans
PCNs           polychlorinated naphthalenes
                                      Vlll

-------
    Term
                        Glossary (continued)
Definition
PICs           products of incomplete combustion
RCRA         Resource Conservation and Recovery Act
RDF           refuse-derived fuel
REMPI-TOF    resonance-enhanced multiphoton ionization-time of flight
SVOCs         semivolatile organic compounds
TEFs           toxic equivalency factors
TEQ           toxic equivalency
THC           total hydrocarbons
VOCs          volatile organic compounds
                                    IX

-------
                          Acknowledgments
The author would like to acknowledge the assistance of Drs. Brian Gullett and Jeong Eun Oh
for helping locate some of the references that are used in the writing of this report. The author
would also like to thank Mr. Michael Galbraith of OSW for his help in explaining the intricacies
oftheMACTrule.
                                      x

-------
1.0   Introduction
1.1    Emissions of PCDDs/Fs
The  emissions of polychlorinated dibenzo-p-dioxins and  polychlorinated dibenzofurans
(PCDDs/Fs) from incinerators and other stationary combustion devices have been of concern
since they were first measured in the stack gases from municipal waste combustors (MWCs).1
PCDDs/Fs have been shown to be carcinogenic and bioaccumulative and have been found in
various concentrations in the exhaust gases from almost every combustion source.2 Of the 75
possible isomers of PCDD and the 135 possible isomers of PCDF, the 17 isomers with chlorine
(Cl) substituted at the 2,3,7, and 8 positions exhibit the carcinogenic behavior.

To account for the varying levels of toxicity of the various PCDD/F isomers, stack gas concen-
trations of PCDDs/Fs are frequently expressed in units of toxic equivalency (TEQs). TEQs are
weighted concentration values based on a series of toxic equivalency factors (TEFs) that are
estimated using various toxicological models based on in vivo or in vitro studies. Each of the
17 toxic isomers has an associated TEF, normalized so that 2,3,7,8-tetrachloro dibenzo-p-dioxin
(TCDD) is defined as having a TEF of 1. The TEQ is calculated  using Equation (1).


                             TEQ = EqTEF;                                 (1)

where C; represents the concentration of the ith isomer (usually in ng/dscm) and TEF; represents
the TEF for the ith isomer. To account for dilution, concentrations are corrected to a common
oxygen (O2) or carbon dioxide (CO2) concentration, such as 7% O2. For the purposes of this
report, the  International TEQ2 (I-TEQ) will be used for all of the correlations between
PCDDs/Fs  and other pollutants  except for the contribution of other  chloroorganics  like
polychlorinated biphenyls (PCBs), which will use TEQs calculated from TEFs derived from the
World Health Organization.3

1.2    Formation of PCDDs/Fs
Laboratory and field studies of PCDD/F formation mechanisms have resulted in theories of the
chemical pathways that  lead to the production of PCDDs/Fs as  unwanted trace by-products
from combustion devices. It is generally accepted4"6 that one of the pathways involves organic
                                       1-1

-------
products of incomplete combustion (PICs) leaving the high temperature zones of a combustor
in the form of volatile or semivolatile organic compounds (VOCs or SVOCs).  These
compounds,  known as precursors, can undergo heterogeneous reactions with flyash-bound
metallic catalysts (such as copper) in the cooler regions of the combustor, including transition
ducts and the air pollution control system (APCS), such as an electrostatic precipitator (ESP),
that result in the formation of PCDDs/Fs. The heterogeneous reactions are strongly dependent
on the temperature  and residence time within the APCS. lino  et  al.7 evaluated isomer
distribution patterns from several waste incineration facilities and found that there was a great
deal of consistency among them and suggested a mechanism that accounts for variations in the
isomer patterns and subsequently the TEQs. A simplified diagram of the formation pathways
of PCDDs/Fs, PCBs, polychlorinated naphthalenes (PCNs) is shown in Figure 1-1.


CO
/
x-V N_
^
C °C

PCDDs
	 ^
A PCNs
                   CO2, H2O
                   C\2,
          Figure 1-1.  Formation Pathways of PCDDs/Fs and Other High MW Pollutants
1.3    Current Regulatory Approach
Emissions of PCDDs/Fs from MWCs are regulated under the Clean Air Act Amendments of
19908.  Emissions of PCDDs/Fs from hazardous waste combustors  (HWCs),  boilers  and
industrial furnaces (BIFs), including cement kilns, halogen acid furnaces, and lightweight
aggregate kilns, are regulated under the Resource Conservation and Recovery Act (RCRA)9.
The emission limits from  hazardous and municipal waste combustion facilities are listed in
Table 1-1.  It is apparent from these emission limits that average day-to-day concentrations of
PCDDs/Fs are exceedingly low; i.e., in the low parts-per-trillion (ppt) range.

                                       1-2

-------
              Table 1-1.  PCDD/F Emission Limits from Waste Combustors
Facility Type
Municipal Waste Combustors
Hazardous Waste Combustors
PCDD/F Emission Limit (corrected to 7% O2)
1 3 ng/dscm total mass (mandatory) or 7 ng/dscm total
mass (optional to qualify for less frequent testing)10
0.2 ng TEQ/dscm or 0.4 ng TEQ/dscm if paniculate matter
APCD inlet temperature <400 °F
1.4   Sampling and Analytical Methods
Due to the low stack gas concentrations of the target analytes,  sampling and analysis for
PCDDs/Fs is a complicated, labor-intensive, and expensive process. Using EPA Method 23 A11,
an isokinetic sample is drawn from the stack,  usually for several hours, using an extractive
sampling probe, a heated filter, an XAD-2 resin trap, and a series of impingers. This sampling
train is then broken down and recovered, yielding filters, XAD-2 resin, and various rinsates.

The recovered samples are then brought to an analytical laboratory where a series of extraction
and cleanup procedures are performed and the samples are eventually analyzed by gas chroma-
tography (GC) and mass spectrometry (MS). If a high resolution GC/MS system is used, then
the analysis is described in EPA Method 829012; if a low resolution GC/MS system is used, then
the analysis is described in EPA Method 8280A13. Analytical costs are typically in the range of
$1000 per sample.

Since triplicate samples are typically required for compliance testing,9 it usually requires a
sampling team to be in the field for a week or two for a single stack test. These stack tests are
done periodically depending on the requirements of the permitting authority. During the periods
when stack sampling is not occurring, facilities typically are required to maintain operations
within  a specified window  defined by other indirect parameters such as temperature, O2 or
carbon monoxide (CO) concentrations defined during compliance testing or trial burns.

Because of the time and expense involved in  sampling for PCDDs/Fs, it is not practical to
perform system optimizations to minimize emissions of PCDDs/Fs. In addition, no indication
is available as to the temporal variability of PCDD/F emissions due to operational fluctuations.

The standard analytical methods for PCDD/F only measure the tetra- through octa-chlorinated
isomers because, from a regulatory standpoint, only the isomers with chlorine substituted at the
2,3,7,8- positions are useful. Recently, however, expanded methods have been made available
                                        1-2

-------
that analyze for the mono-, di-, and tri-substituted CDDs and CDFs for purposes of aiding in
the understanding of the PCDD/F formation mechanisms.

1.5    Surrogate Concept
It would be ideal if all of the toxic PCDD/F isomers could be measured continuously in real-
time. However, current state-of-the-art instrumentation is not capable of achieving this goal.
Instead, other more easily-measurable parameters can be used to give an indication of the
concentrations of PCDDs/Fs in the stack gases. These more easily measurable compounds are
called surrogates. Figure 1-2 elaborates on the earlier PCDD/F formation pathway figure by
describing the concentrations of the intermediate species that can be found in the stack gases
and are important in the PCDD/F formation mechanism. The main problem from a practical
standpoint is that the higher concentration and subsequently more easily measurable surrogate
compounds are less directly involved in the mechanism that forms PCDDs/Fs.
         {3
         C
O
O
         o
         Q_
1000 _

 100,



   1 ,

 10'1.

 W2.
                        CO       Chlorinated
                                 c.,C,
                                 Compounds
                   HCI, C!;i, flyash. soot
               PIU5
 •, f., .
Chlorinated
Aromatira
                                                       PCDDs
                                                       PCDFs
                                                       RCNs
            Figure 1-2. Relative Concentrations of PCDDs/Fs and Their Surrogates

The purposes of this document are as follows:
   • D To discuss current approaches and their limitations for using surrogates,
   • D To discuss what other options exist for using surrogate indicators of PCDDs/Fs,
   • D To discuss methods for continuously or semi-continuously measuring those surrogate
       compounds, and
   • D To discuss the timeline of commercial availability of those various analytical techniques.
                                        1-4

-------
2.0   Surrogate Approaches
A surrogate parameter can be a single pollutant that can itself, or in combination with other
parameters, account for the variability of the PCDD/F data. A linear relationship between the
surrogate and the PCDDs/Fs, derived using linear least squares fitting, would be the simplest
use of a surrogate. However, given that minor changes in operational parameters can result in
orders of magnitude changes in the concentrations of PCDDs/Fs, it is likely that the most
effective surrogates will probably exhibit a non-linear relationship with PCDDs/Fs, such as a
log-linear relationship. Multiple parameter models can improve the overall fit, although given
the general lack of highly robust PCDD/F vs. operational conditions data sets, adding too many
parameters to the model can result in simply re-predicting the original data set as opposed to
elucidating the statistical relationship between the surrogate and the PCDD/F. In addition,
multiparameter models can become counter intuitive due to competing effects of individual
parameters, and they are difficultto visualize on paper. For the purposes of this report, statistical
significance is defined as when the "P value" of the parameter is less than 0.05. In addition, the
correlation coefficients used will be the R2 value.

For the purposes of this report, the following potential surrogate PCDD/F indicators will be
discussed:
    • D Carbon Monoxide,
    • D Total Hydrocarbons (THC)
    • D Low Molecular Weight (MW) VOCs,
    • D Chlorobenzenes (CBz) and Chlorophenols (CPh),
    • D PCBs,
    • D Poly cyclic Aromatic Hydrocarbons (PAHs), and
    • D Lower Chlorinated Dioxins and Furans.
2.1    Measurement of Carbon  Monoxide
CO is commonly used as an indicator of poor combustion and is relatively easy to measure using
non-dispersive  infrared   (NDIR)  continuous  emission  monitors  (CEMs).  Stack  CO
concentrations are typically limited to a maximum hourly rolling average in many combustion
facility permits and are typically found at levels between 1 and 100 ppmv in well operated
combustors. During the National Incinerator Testing and Evaluation Program (NITEP) that was
performed collaboratively between the U.S. EPA and Environment Canada during the late
                                       2-1

-------
1980s and early 1990s,14 analyses were performed at a variety of combustion conditions to see
if CO could be an effective surrogate for PCDDs/Fs in the stack gases of municipal waste
combustion facilities. It was found that CO was a good indicator of PCDD/F concentrations
when the facility was operating poorly; however, for a well-operated facility, CO was not a
good indicator of PCDDs/Fs. Figure 2-1 illustrates the CO emissions vs. total PCDDs/Fs at a
variety of combustion conditions from a refuse-derived fuel (RDF) combustor. Observe how
PCDD/F tracks CO for conditions where CO emissions are high, but there is no apparent trend
in the region of the plot where CO emissions are low. This suggests that low CO emissions are
a necessary condition to minimize PCDDs/Fs but are not sufficient to assure compliance. This
observation is also supported in later work by German researchers.15
           »Q§
    e
    o
    V.
    T3
    bfc
    U,
    Q
    U
    I
    a
    u
           1500
1000
 500

O
o
o° o
QQ
o

1
•(
»


»







*


O Gcxxf Combumon
Q foot Combustion


                                     400
                                       600
soo
1000
                                         CO (ppm)
                 B» 12*
             Figure 2-1. CO vs. Total PCDD/F Stack Emissions in an RDF Combustor14
2.2    Measurement of Total  Hydrocarbons
Similarly, THC is another parameter that is frequently measured at combustion facilities to
assure good combustion and is fairly easy to measure using flame ionization detector- (FID)
based CEMs, either heated or unheated. THC concentrations are typically in the 1 to 10 ppmv
range for a well operated combustion facility. THC results are reported in units of methane or
propane equivalents depending on how the CEM was calibrated. THC was also evaluated as a
potential surrogate for PCDDs/Fs during the NITEP testing program. Conclusions were drawn
                                       2-2

-------
similar those for using CO for a PCDD/F surrogate; low THC is a necessary condition to assure
low PCDD/F emissions, but low THC is not a sufficient condition to assure compliance for a
well operated facility. Figure 2-2 shows THC vs. total PCDD/F emissions from the stack of an
RDF combustor.


    M
    1
    c
    IL
    D
    O
    I
    8
    Cu

A
4
A
A
A A
A





A





a



A


€k»d Co-ifilxatkin
A f
-------
may eventually result in formation of PCDDs/Fs. Since these species are formed in the high
temperature regions of the combustor and the PCDDs/Fs are formed in the lower temperature
zones, it is likely that VOC PICs would require a temperature parameter such as an exhaust duct
temperature, stack temperature, or quench rate to correlate effectively with PCDDs/Fs.

Lemieux et al.  analyzed three data  sets from a pilot-scale incineration facility to evaluate
potential correlations between C2 VOC PICs and total PCDDs/Fs.16 Test conditions ranged over
a fairly wide set of variations, including different surrogate wastes containing varying amounts
of bromine and chlorine. In these datasets, total PCDDs/Fs were available but not I-TEQ; also
many of the  potential VOC PIC indicators were present at concentrations at or near the
instrument (an online GC) detection limits, which significantly reduced the available data with
which to develop correlations. Out of the 10 Cl and C2 target analytes of the online GC, only
5   (vinyl  chloride,   1,1-dichloroethene,   1,2-dichloroethene,  trichloroethylene,   and
tetrachloroethylene)  were consistently present at quantifiable levels in all the  data sets.
Statistical analyses were performed,  both on individual data sets and on the data set derived
from combining all three individual sets of data. Each data set contained one or more C2
chloroalkenes that were able to account for a statistically significant fraction of the variance in
PCDD/F emissions.

For each individual set of data, simple linear regressions were generated between individual C2
chloroalkenes and the total PCDDs/Fs. Variations in the vinyl chloride concentrations were able
to account for the variations in the PCDD/F concentrations strongly in two of the three data sets
and weakly in the other. A regression on the combined data set showed a significant (R2 =
0.582) relationship between vinyl chloride (log) concentrations and PCDD/F  concentrations.
Performing a  two-parameter regression by combining one temperature-related parameter with
a C2 chloroalkene concentration-related parameter yielded statistical models that were able to
account for more of the variance in PCDD/F concentrations. In the referenced paper,16 several
temperature parameters were evaluated, and  the quench rate (the rate at which the duct
temperature dropped from the furnace exit to the sampling location) was found to yield the best
two-parameter  fits. Again,  vinyl chloride and  other C2 chloroalkene-related parameters
exhibited  statistical  significance as  well, including  the  total concentration of all  C2
chloroalkenes. The advantage of choosing a single C2 chloroalkene to serve as an indicator is
that available  measurement techniques are much less expensive when directed at a more limited
target analyte set. A limitation of using this technique is that the C2 chloroalkene must be
measured  prior to  any VOC removal  device.   Some  combustion devices  utilize carbon
adsorption systems to remove organic compounds from stack gases in the APCS, which would
require that the chloroalkenes be measured upstream of the carbon bed as opposed to inside the

                                         2-4

-------
stack. In other words, the pre-APCS VOC concentrations  would need to correlate with
PCDDs/Fs in the stack. Overall, it appears that concentrations of C2 chloroalkenes, when
coupled with a temperature related parameter, may have good potential as a surrogate indicator
of PCDDs/Fs, provided sufficient detection limit goals are met. Table 2-1 lists the results of
using lowMW VOC PICs as a surrogate for PCDDs/Fs; the results with statistical significance
are shown.
        Table 2-1. Regression Results of C2 Chloroalkenes vs. Total PCDDs/Fs
                                                                        16
No. of Model Parameters
1
2
Data Set
1
2
3
Combined
Combined
Parameter(s)
1 ,2-Dichloroethene
Trichloroethylene
Vinyl Chloride
Vinyl Chloride
1,1-Dichloroethene
1 ,2-Dichloroethene
Tetrachloroethylene
Vinyl Chloride
1,1-Dichloroethene
1 ,2-Dichloroethene
Tetrachloroethylene
E C2 chloroalkenes
Vinyl Chloride + Quench Rate
1 ,1-Dichloroethene + Quench Rate
1 ,2-Dichloroethene + Quench Rate
Trichloroethylene + Quench Rate
Tetrachloroethylene + Quench Rate
E C2 chloroalkenes + Quench Rate
R2
0.633
0.615
0.287
0.681
0.436
0.452
0.773
0.582
0.162
0.476
0.236
0.282
0.677
0.399
0.589
0.302
0.645
0.673
Based on this information, the following observations can be made about using low MW VOC
PICs as a surrogate for PCDDs/Fs:
    • D Low MW VOC PICs may be useful as PCDD/F indicators, but a temperature-related
       parameter such as a flue gas temperature, stack temperature, or quench rate should be
       included in the correlation to improve the fit.
    • D Instrument detection limits of 1 ppbv may not be sufficient to allow development of
       correlations between low MW VOCs and PCDDs/Fs.
                                       2-5

-------
   • D The data used for comparison were all taken on a single facility using different fuels and
       conditions; this suggests that low MW VOCs may have potential across different fuel
       types, but it is unknown how the correlations hold from facility to facility.

2.4    Measurement of Chlorobenzenes and Chlorophenols
Current knowledge of the formation mechanism of PCDDs/Fs in incinerators proposes that
condensation reactions of chlorinated aromatic compounds such as CBz and CPh  may be
responsible for a significant amount of the PCDDs, and possibly the PCDFs, formed in the
lower temperature regions of waste combustors.2  As  such, it  would be logical that the
concentrations of PCDDs/Fs in the stack would be a function of the concentrations of CBz and
CPh  at some point in the combustor.  A potential drawback of the use of CBz and CPh as a
surrogate for PCDDs/Fs is that the concentrations of CBz and CPh in a well operated waste
combustor are fairly low, typically in the range of 1 to 10 jig/m3—about an order of magnitude
lower than the low molecular weight VOC PICs.15 Another potential drawback is that many of
the CBz isomers and all of the CPh isomers  are semi-volatile compounds, have fairly high
boiling points, and are typically associated with particulate matter at stack temperatures as
opposed to being in the gas-phase.  In addition, CPh are fairly reactive and can chemisorb onto
surfaces in the APCS. Only monochlorobenzene and dichlorobenzene can be measured using
the standard  EPA method for VOCs;17"19 the standard method for semivolatile compounds20
requires extensive laboratory preparation (extraction, concentration) prior to analysis.

There has been extensive work performed in  Germany examining the relationships between
PCDD/F  in the flue and  stack gases and concentrations of various chlorinated aromatic
compounds. Kaune et al.15 initially examined field data from a rotary kiln hazardous waste
combustion facility equipped with  a heat recovery boiler, dry ESP, condenser, then a wet ESP.
They found relatively poor correlations between CPh and PCDDs/Fs because the extremely low
concentrations of the CPh made it difficult to develop good fits. However, correlations were
successfully developed between pentachlorobenzene (Cl5Bz) and various PCDD/F homologue
groups in the flue gases, individual toxic isomers, total PCDDs/Fs, and I-TEQ. In particular, a
correlation with R2=0.94 was found between Cl5Bz and I-TEQ at an intermediate location
within the flue gas cleaning system. It must be noted that for the test series in question, the
concentrations of the PCDDs/Fs varied over 2 orders of magnitude, as did the concentrations
of CBz.

A follow-on study at the  same incinerator21 showed that varying operation of the flue gas
cleaning system by altering injection rates of activated carbon did not result in failure of the
ability of Cl5Bz  to still yield good  correlations with  I-TEQ. Further  work by the  same

                                       2-6

-------
investigator22 showed that the sum of the tetrachlorobenzene (Cl4Bz) isomers also yielded good
correlations (R2=0.894) with I-TEQ. They were not able to find any correlation in the stack
gases.

Further work by the same investigators23 examined data from three different incinerators of
slightly different designs and several different sampling points within the incinerators. They
found that the data points from the different facilities either fell on the same regression line or
on parallel regression lines that differed only by their intercept. This suggested that the slope
of the regression  line for a facility that has  not been investigated could be taken from
correlations developed at a similar sampling point of an already-investigated incinerator. This
would significantly reduce the workload and cost necessary to develop detailed  regression
equations.

Some researchers in Finland24 examined correlations between gas-phase and particulate-bound
chlorinated aromatic compounds and PCDDs/Fs and found that gas-phase indicator compounds
gave good correlations whereas particulate-phase compounds did not. This observation is an
important implication when on-line measurement of these compounds is concerned, since the
most  promising on-line  measurement methods all  involve measurement  of  gas-phase
compounds only.

All of the above  CBz and CPh measurements were made using conventional  extractive
sampling techniques. Ideally, the surrogates would be measured in real time or near-real time
in order to be useful for system optimization and to minimize the cost of their application.
Further work has been done on one of the same three German incinerators using  an on-line
technique—resonance enhanced multiphoton ionization-time of flight (REMPI-TOF) mass
spectrometry—in which chlorinated aromatics  were measured  in real time and compared to
extractively collected PCDD/F measurements.25 In these tests, monochlorobenzene (MCBz) was
measured in real time at two locations in the incinerator: the boiler exit and the  stack. The
investigators found a relationship between MCBz measured in  the flue gas and I-TEQ at the
boiler exit with an R2=0.82 and at the stack exit with an R2=0.76 using the same  regression
methodology. This suggests that measurements of some compounds such as MCBz in higher
temperature regions of the combustor, where they are mainly in the gas-phase, can be useful
predictors of I-TEQ in the stack, where those compounds may be partially in the solid-phase
(and thus inaccessible to gas-phase measurement techniques like REMPI-TOF).

In a separate paper,26 principal component analysis was performed on various CBz  isomers to
see which ones tracked I-TEQ best. They found that MCBz,  1,3-dichlorobenzene (l,3-C!2Bz),

                                        2-7

-------
1,2-dichlorobenzene (l,2-C!2Bz), and Cl5Bz (as reported earlier15) all gave good correlations
with I-TEQ. More importantly, it was found that the isomer pattern of CBz and PCDD/F did
not significantly change as the flue gases passed through the APCS, which strengthens the
argument that measurement of chlorinated aromatics are a robust indicator  of PCDD/F
concentrations. Additional work on a pilot-scale waste combustor27 further strengthened the
usefulness of lower chlorinated CBz as indicators of PCDD/F.

Table 2-2 lists the CBz  and CPh compounds that have yielded  statistically significant
correlations with I-TEQ.

                   Table 2-2.  List of CBz and CPh I-TEQ Surrogates
Compound
MCBz
1 ,2-CI2Bz
1 ,4-CI2Bz
1 ,2,3-CI3Bz
1 ,2,3,4-CI4Bz
S CI4Bz
CI5Bz
CI6Bz
2,4-CI2Ph
2,4,6-CI3Ph
2,3,4-CI3Ph
2,3,5,6-CI4Ph
2,3,4,6-CI4Ph
CI5Ph
Approximate
Concentration28
(pptv)
300
50
30
20
30
NAa
10
5
50
30
10
5
5
10
NA
Correlation
Coefficient, R2
0.72
0.61
0.42
0.58
0.83
0.89
0.62
0.55
0.64
0.67
0.41
0.48
0.56
0.62
0.94
Source
[28]
[28]
[28]
[28]
[28]
[22]
[28]
[28]
[28]
[28]
[28]
[28]
[28]
[28]
[151
       1 NA=not available
Based on these sources of information, the following observations can be made concerning the
use of chlorinated aromatic compounds as surrogates for PCDDs/Fs:
    • D Sampling of the indicator compounds and PCDDs/Fs does not necessarily need to be
       performed at the same location to develop the correlations. This is an important point,
       especially as the discussion of potential measurement methodologies comes into play.
                                        2-8

-------
   • D The development  of the  correlations is likely to be  facility-specific, although
       information acquired during the development of detailed regression correlations at one
       facility may be used successfully at other similar facilities to reduce the cost and
       workload of developing correlations at those facilities.
   • D The ability of the surrogates to predict PCDD/F concentrations improves when a wider
       range of concentrations is used to develop the correlations.
   • D In developing the facility-specific correlations, individual isomers should be considered
       as surrogates as well as homologue groups and entire classes of compounds.
   • D Developing correlations using simultaneous measurements of CBz, CPh, and PCDDs/Fs
       may not be useful to assure compliance; however, it may be extremely useful to allow
       facilities to optimize operation to minimize PCDD/F emissions.

2.5    Measurement of Polychlorinated Biphenyls
PCBs are believed to be formed through a set of reactions similar to those that form PCDDs/Fs.
The TEQs from PCBs (calculated using TEFs published by the World Health Organization3)
in waste combustors typically only represent a small fraction (less than 5%) of the PCDD/F -
derived I-TEQ.29 It may be possible to infer concentrations of PCDDs/Fs from total PCB
concentrations or from concentrations of individual congeners.15 In Kaune et al.,15 it was found
that heptachlorobiphenyls (C17B) gave an R2=0.87 correlation with I-TEQ, and individual PCB
congeners may give good correlations with I-TEQ. However, PCBs  are typically found in
concentrations similar to PCDDs/Fs, and measurement of PCBs is typically done using the same
stack gas sampling and analysis methods as is done with measurement of PCDDs/Fs,  so the
usefulness of PCBs as a surrogate indicator of PCDDs/Fs is highly questionable.

2.6    Measurement of Polycyclic Aromatic Hydrocarbons
Since PAHs are formed as a result of incomplete combustion in the furnace and are present
typically in concentrations two orders of magnitude higher than PCDDs/Fs, they may be worth
investigating as potential PCDD/F indicator compounds. Kaune et al.15 found that fluoranthene,
pyrene, benzo[ghi]fluoranthene/benzo[c]phenanthrene, chrysene, and  total PAH  gave
correlation coefficients of 0.76,0.76,0.73,0.71, and 0.84 vs. I-TEQ, respectively, at one sample
point. However, no correlation was found at any of the other sample points, which suggests that
the correlations of PAHs vs. PCDDs/Fs are not as robust as the  correlations with the  single
ringed chlorinated aromatic compounds. Blumenstock et al.27 used principal component analysis
and found that PAHs were more closely correlated with CO emissions than  with PCDD/F
emissions. This suggests that PAHs, like CO, are useful indicators of whether a combustor is
operating well or poorly, but may not be useful as an indicator of PCDDs/Fs.
                                       2-9

-------
2.7    Measurement of Lower Chlorinated Dioxins and Furans
Regulatory compliance sampling for PCDDs/Fs has typically only evaluated emissions of the
tetra- through octa-chlorinated PCDD/F isomers, since the toxic isomers all have a minimum
of four chlorines substituted at the 2, 3, 7, and 8 positions on the CDD/F molecule. Recently,
though, the analytical techniques have expanded to include the mono-, di-, and tri-chlorinated
CDD/F molecules due to their promise as useful indicators of PCDD/F concentrations and I-
TEQ 3o,3i jjjg acjvantages afforded by the lower chlorinated CDDs/Fs include:
   • D They  are present  typically at higher concentrations  than the  higher chlorinated
       congeners, although they are still present in very low concentrations—on the order of
       1-10 ng/m3 in a well operated combustion facility;
   • D Because of their higher  volatility than the higher chlorinated PCDD/F congeners, a
       higher fraction of the lower chlorinated congeners are present in the gas-phase at flue-
       and stack-gas conditions; and
   • D Because of their lower numbers of substituted chlorines, they have better detection
       limits when measured by on-line methods such as REMPI-TOF.

Work by Oser et al.31 has developed the concept of using lower chlorinated CDD/F isomers as
indicators of I-TEQ.  Jet-REMPI, a variant on REMPI-TOF where a supersonic jet is used to
cool the gas sample to temperatures approaching absolute zero, has been used to dramatically
increase the sensitivity of the REMPI-TOF method to successfully measure these compounds.

Blumenstock et al.28  used REMPI-TOF on a German hazardous waste combustor to measure
various MCDD/F, DCDD/F, and TriCDD/F isomers and the sum of their homologue groups and
examine the  statistical significance of using them to account for I-TEQ. They found that, for
the plant on which they made the measurements, several of the lower chlorinated CDD isomers,
and one of the lower chlorinated CDF isomers gave statistically significant correlations. Gullett
and Wikstrom compared the lower chlorinated CDD/F isomers and homologue groups to the
total  PCDD/F and I-TEQ for three different datasets: one from a full-scale RDF combustor
firing various mixtures  of municipal waste and coal, and two from a pilot-scale research
combustor burning solid fuel pellets. They developed single-, two-, and three- parameter
statistical models to develop lower chlorinated CDD/F indicator relationships with both total
PCDDs/Fs and I-TEQ. A summary of the results from those two studies (limited to correlations
with  I-TEQ) is shown in Table 2-3.
                                       2-10

-------
Table 2-3.  Lower Chlorinated CDDs/Fs vs. I-TEQ
Number of Model
Parameters
1
2
3
Compound(s)
1,3-DiCDD
1 ,4,7-TriCDD
1 ,2,3-TriCDD
1,7,8-TriCDD
1 ,4,6-TriCDD
1,2,6-TriCDD
1,2,9-TriCDD
1,2,3-TriCDF
2,4,6-TriCDF
E TriCDD
E TriCDF
1,2,3-TriCDF, 1,6-DiCDD
1,2,3-TriCDF, 2,4,6-TriCDF
E DiCDF, E TriCDF
E TriCDD, EDiCDF
E DiCDD, E DiCDF, E TriCDF
Correlation
Coefficient, R2
0.38
0.53
0.55
0.56
0.56
0.58
0.52
0.68
0.44
0.44
0.52
0.95
0.77
0.84
0.83
0.86
0.81
0.97
0.87
0.99
Source
[28]
[28]
[28]
[28]
[28]
[28]
[28]
[30] (Norfolk data set)
[30] (Umea-1 data set)
[30] (Umea-2 data set)
[28]
[30] (Umea-2 data set)
[30] (Norfolk data set)
[30] (Norfolk data set)
[30] (Umea-1 data set)
[30] (Umea-2 data set)
[30] (Norfolk data set)
[30] (Umea-2 data set)
[30] (Norfolk data set)
[30] (Umea-2 data set)
                   2-11

-------
3.0  Surrogate Analytical Techniques

3.1   Continuous Emission  Monitors for Carbon Monoxide
The measurement of CO in combustor stacks using NDIR CEMs is a reliable, proven method
that has been around for years.32 These instruments are sensitive down to concentrations in the
low parts per million range, and there are many different models that are commercially
available. Most combustion facilities are required to monitor CO emissions in their stack as a
requirement of their respective state's air quality permit.

3.2   Continuous Emission  Monitors for Total Hydrocarbons
There are two currently available technologies for the  measurement of THCs in combustor
stacks: NDIR33 and FID34. Both are reliable, proven methods that have been commercially
available for years and can measure  THC  concentrations  (reported as equivalent parts per
million methane or propane) down to the low parts per million levels. A variant on the FID
method utilizes a heated (=150 °C) sample line, filter, and analyzer, which provides additional
measurement capabilities of THCs that condense out between 150 °C and ambient temperatures.
Although not required in the permit of facilities as a general rule, many  facilities are equipped
with THC analyzers.

3.3   Continuous Emission Monitors for Volatile Organic Compounds
The real-time or near-real-time measurement of VOCs in flue and stack gases is not as well-
developed a technology as is the measurement of CO and THC. The U.S. EPA has proposed
draft performance specifications for the operational requirements for VOC CEMs.35'36 However,
several instruments at various stages of development exist and are adaptable from commercially
available equipment.  In  1995, the U.S.  EPA held a demonstration  at their pilot-scale
incineration research facility in Jefferson, AK and invited instrument manufacturers to bring
their instruments to the site, where the EPA paid for the  operation of the facility and all
sampling and analytical  activities. Several manufacturers brought their instruments and
attempted, with mixed results, to measure the concentrations of several target analytes in the
incinerator stack gas and compare the results  to standard EPA methods. Table 3-1 lists
information about the various techniques for rapid  measurement of VOC PICs and their
approximate instrument detection limits; note that this table only includes instruments that are
commercially available or are at a position in their development that would enable rapid
commercial development if the market forces allowed.

                                      3-1

-------
             Table 3-1. Techniques for Rapid Measurement of VOC PICs
Method
Online GC38
on mobility
;pectroscopy
(IMS)
Differential
optical
absorption
;pectroscopy
(DOAS)
rourier
ransform
nfrared
(FTIR)
Online Direct-
Sample
MS37'39
3EMPI25'31
Target
Analytes
VOCs (up
through
MCBz and
CI2Bz)
VOCs
UV absorbing
organic
compounds
Some VOCs
VOCs
Compounds
with aromatic
ring
Sampling/
Analytical
Time Frame
<1 hour
<1 min.
Real-time if no
concentration
is used; but
concentration
is necessary
to improve
detection
limits
Minutes if no
concentration
is used
1 min.
Real-time if no
concentration
is used
Approximate
Detection
Limit
Low ppbv
Low ppbv
Low ppmv
Low ppmv
Low ppbv
Low pptv
Issues
Potential for co-
eluting peaks
Low specificity
Detection limits
are too high
Detection limits
are too high
May require GC
to improve
specificity
Need spectra for
each congener
to be measured;
may require
concentration for
highly
halogenated
compounds
Availability
Can be assembled
from commercially
available
components
IMS for some
organic compounds
is available now
(ETG)
Available for some
organic compounds
now (ABB/OPSIS)
Commercially
available
Can be assembled
from commercially
available
components
Can be assembled
from commercially
available
components
3.4   Continuous  Emission   Monitors   for  Chlorobenzenes  and
      Chlorophenols
Options for rapid measurement techniques for CBz and CPh are much more limited. Other than
MCBz and Cl2Bz, these compounds are in the semivolatile range of boiling points. Getting a
valid stack gas sample into the instrument is not well established for all instrument types. CPh
are also fairly reactive compounds and can chemisorb on surfaces.40 In addition, the detection
                                    3-2

-------
limits for most of the instruments listed in Table 3-2 are not low enough for direct measurement
of CBz and CPh—some concentration step must be performed, and there are no data in the
literature for many of these instruments successfully measuring flue and stack gas CBz and
CPh. Table 3-2 lists the techniques for rapidly measuring CBz and CPh. Thus far, only the
variants on REMPI techniques have been successfully used for online measurements of nearly
the complete set of CBz and CPh isomers.

            Table 3-2. Techniques for Rapid Measurement of CBz and CPh
Method
Online GC38
=!EMPI25<31
Target
Analytes
MCBz and
CI2Bz only
All isomers of
Cbz and CPh
Sampling/
Analytical
Time Frame
<1 hour
Real-time if no
concentration
is used
Approximate
Detection
Limit
Low ppbv
Low pptv
Issues
Potential for co-
eluting peaks
Need spectra for
each congener
to be measured;
may require
concentration for
highly
halogenated
compounds
Availability
Can be assembled
from commercially
available
components
Can be assembled
from commercially
available
components
3.5   Continuous  Emission  Monitors   for  Polycyclic  Aromatic
      Hydrocarbons
A monitor for particle-bound PAH using photoelectric detection is commercially available from
EchoChem. The measurement is performed on a stack slip stream that is diluted with ambient
air. This instrument gives a single semi-quantitative  reading representing an equivalent
concentration of 2- and 3-ringed PAHs and is sensitive down to the 10 ng/dscm range. It has
a very fast response time and has been used in the past to measure  transient  changes in
emissions from a pilot-scale rotary kiln burning tire-derived fuel.

3.6   Continuous Emission Monitors for Lower Chlorinated Dioxins
      and Furans
The only instruments that have successfully demonstrated the capability of measuring lower
chlorinated CDDs/Fs from combustion systems has been the REMPI-TOF and the Jet-REMPI.
Due to phenomena associated with the REMPI technique, sensitivity falls off with increasing

-------
chlorine substitution on the target molecules. As such, a concentration step may be necessary
to achieve appropriate detection limits for use of this technique to the measurement of lower
chlorinated CDDs/Fs.
                                         3-4

-------
4.0 Conclusions
Based  on the current state-of-the-art  in using  surrogate indicators for  measurement  of
PCDDs/Fs, it appears that:
    1.  The development of correlations between indicator compounds and PCDDs/Fs is likely
       to be facility-specific, although information acquired during the development of detailed
       regression correlations at one facility may be used successfully at other similar facilities
       to reduce the cost and workload of developing correlations at those facilities.
    2.  Sampling of the indicator compounds and PCDDs/Fs does not necessarily need to be
       performed at the same location to develop the correlations. Targeting sampling locations
       so that desired indicators are  predominantly in  the gas-phase can improve detection
       limits for trace species.
    3.  The ability of the surrogates to predict PCDD/F concentrations improves when a wider
       range of concentrations is used to develop the correlations.
    4.  Low MW VOC  PICs  may  be useful  as PCDD/F  indicators, but some sort  of
       temperature-related parameter should be included in the correlation to improve the fit.
    5.  Instrument detection limits of  1 ppbv may not be sufficient to allow for development of
       correlations between low MW VOCs and PCDDs/Fs.
    6.  In developing  the facility-specific correlations between CBz/CPh and PCDDs/Fs,
       individual isomers as well as homologue groups and entire classes of compounds should
       be considered as surrogates.
    7.  The CBz, CPh, and low chlorinated CDDs/Fs are more effective surrogates than the low
       MW VOC PICs; however, their  rapid measurement is more difficult.
    8.  Based purely on the limited available datasets, it appears that using CBz, CPh, and low
       chlorinated CDDs/Fs can give roughly the same  ability to account for the variability in
       PCDD/F  emissions.  With the  current level of information, it is not possible  to
       definitively pick any one of these compounds as the ideal surrogate.  Rather, since the
       concentrations of many of these  compounds are likely to be cross-correlated with each
       other, linear combinations of all of these semivolatile compounds, derived through
       principal component analysis, may yield the most effective correlations.
    9.  Developing correlations using  simultaneous measurements of CBz, CPh, and PCDDs/Fs
       may not meet current compliance requirements; however, it will be extremely useful to
       allow facilities to optimize operation to minimize PCDD/F emissions and understand
       the role that transient emissions play in the overall emissions of PCDDs/Fs. In addition,
       development of facility-specific correlations would  enable a  facility to estimate
                                        4-1

-------
   bounding conditions and appropriate confidence intervals for dioxin surrogates that
   would assure compliance.
10. Before this technique can be used with adequate confidence at waste combustion
   facilities on a routine basis, additional datasets need to be generated. It would be useful
   to make detailed, isomer-specific measurements of CBz and CPh compounds, as well
   as low chlorinated CDDs/Fs, during every test where PCDDs/Fs are measured.
11. The REMPI-based instruments have shown their potential for being able to successfully
   measure the compounds that yield the best correlations with PCDDs/Fs. However, it is
   unlikely that they will be developed to be fully commercial without some sort of outside
   force generating a market for instrument manufacturers to enter. This force could be
   regulatory in nature (e.g., the EPA mandating their use) although such a requirement is
   not in place at this  time. Rather, it may  be that a forward-thinking facility or a
   government-operated facility may want to install one on their stack to perform system
   optimizations.
                                     4-2

-------
5.0   References

1.  Olie, K.; Vermeulen, P.; Hutzinger, O. (1977), "Chlorodibenzo-p-dioxins and
   Chlorodibenzofurans are Trace Components of Fly Ash and Flue Gas of Some
   Municipal Incinerators in the Netherlands," Chemosphere, Vol. 8, pp. 455-459.
2.  U.S. EPA (2000), "Public Review Draft: Exposure and Health Reassessment of 2,3,7,8-
   Tetrachloro Dibenzo-p-dioxin (TCDD) and Related Compounds," EPA-600/P-00-
   OOlBb, National Center for Environmental Assessment, Washington, DC, September.
3.  Van den Berg, M.; Birnbaum, L.; Bosveld, A.; Brunstrom, B.; Cook, P.; Feeley, M.;
   Giesy, J.; Hanberg, A.; Hasegawa, R.; Kennedy, S.; Kubiak, T.; Larsen, J.; van
   Leeuwen, F.; Liem, A.; Nolt, C.; Peterson, R.; Poellinger, L.; Safe, S.; Schrenk, D.;
   Tillitt, D.; Tysklind, M.; Younes, M.; Waern, F.; Zacharewski, T. (1998), "Toxic
   Equivalency Factors (TEFs) for PCBs, PCDDs, PCDFs for Humans and Wildlife,"
   Environmental Health Perspectives, Vol. 106, pp. 775-792.
4.  Addink, R.; Olie, K. (1995), "Mechanisms of Formation and Destruction of
   Polychlorinated Dibenzo-p-dioxins and Dibenzofurans in Heterogeneous Systems,"
   Environmental Science and Technology, Vol. 29, pp. 1425-1435.
5.  Shaub, W.M.; Tsang, W. (1983), "Dioxin Formation in Incinerators," Environmental
   Science  and Technology, Vol. 17, pp. 721-730.
6.  Tsang, W. (1990), "Mechanisms for the Formation and Destruction of Chlorinated
   Organic Products of Incomplete Combustion," Combustion Science and Technology,
   Vol. 74, pp. 99-116.
7.  lino, F.; Imagawa, T.; Gullett, B. (2000), "Isomer Prediction Model of Polychlorinated
   Dibenzofurans from Municipal Waste Incinerators," Organohalogen Compounds, Vol.
   46, pp. 114-117.
8.  Clean Air Act Amendments, (1990) Public Law  101-549.
9.  U.S. EPA (2003), "Hazardous Waste Combustor MACT Rule,"
   http://www.epa.gov/hwcmact/ (Accessed February 2004)
10. U.S. EPA (1995), "Standards of Performance for New Stationary Sources and Emission
   Guidelines for Existing Sources," Code of Federal Regulations, Title 40, Part 60,
   Appendix A, U.S. Government Printing Office, Washington, DC, Vol. 60, pp. 65387-
   65436.
                                       5-1

-------
11. U.S. EPA (1991), "EPA Test Method 23 Determination of Poly chlorinated Dibenzo-p-
   dioxins and Poly chlorinated Dibenzofurans from Stationary Sources," Code of Federal
   Regulations, Title 40, Part 60, U.S. Government Printing Office, Washington, DC, July.
12. U.S. EPA (1997), "Method 8290, Polychlorinated Dibenzodioxins (PCDDs) and
   Polychlorinated Dibenzofurans (PCDFs) by High Resolution Gas Chromatography/High
   Resolution Mass Spectrometry (HRGC/HRMS)," Test Methods for Evaluating Solid
   Waste. Integrated Manual (SW-846), Volume IB: Laboratory Manual,
   Physical/Chemical Methods. Final Update III, SW-846 [NTIS PB97-156137], Office of
   Solid Waste, Washington, DC, June. Also located at
   http://www.epa.gov/epaoswer/hazwaste/test/pdfs/8290.pdf (Accessed February 2004)
13. U.S. EPA (1997), "Method 8280, Polychlorinated Dibenzodioxins (PCDDs) and
   Polychlorinated Dibenzofurans (PCDFs) by High Resolution Gas Chromatography/Low
   Resolution Mass Spectrometry (HRGC/LRMS)," Test Methods for Evaluating Solid
   Waste. Integrated Manual (SW-846), Volume IB: Laboratory Manual,
   Physical/Chemical Methods. Revised Update III, SW-846 [NTIS PB97-156111], Office
   of Solid Waste, Washington, DC, June.
14. U.S. EPA (1994), "National Incinerator Testing and Evaluation Program: The
   Environmental Characterization of Refuse-derived Fuel (RDF) Combustion
   Technology," EPA-600/R-94-140 [NTIS PB96-153432], Washington, D.C., December.
15. Kaune, A.; Lenoir, D.; Nikolai, U.; Kettrup, A. (1994), "Estimating Concentrations of
   Polychlorinated Dibenzo-p-dioxins and Dibenzofurans in the Stack Gas of a Hazardous
   Waste Incinerator from Concentrations of Chlorinated Benzenes and  Biphenyls,"
   Chemosphere, Vol. 29, pp. 2083-2096.
16. Lemieux, P.M.; Lee, C.W.; Ryan, J.V. (2000), "Prediction of Dioxin/Furan Incinerator
   Emissions Using Low Molecular  Weight Volatile Products of Incomplete Combustion,"
   J. AWMA, Vol. 50, pp. 2129-2137.
17. U.S. EPA (1986), "Test Method 0030, Volatile Organic Sampling Train," Test Methods
   for Evaluating Solid Waste, Volume II, SW-846 [NTIS PB88-239223], Office of Solid
   Waste, Washington, DC, September. Also located at
   http://www.epa.gov/epaoswer/hazwaste/test/pdfs/0030.pdf'(Accessed February 2004)
18. U.S. EPA (1986), "Test Method 5040, Protocol for Analysis of Sorbent Cartridges from
   Volatile Organic Sampling Train," Test Methods for Evaluating SolidWaste, Volume II,
   SW-846 [NTIS PB88-239223], Office of Solid Waste, Washington, DC, September.
19. U.S. EPA (1986), "Test Method 8240, Gas Chromatography/Mass Spectrometry for
   Volatile Organics," Test Methods for Evaluating Solid Waste, Volume II, SW-846 [NTIS
   PB-239223], Office of Solid Waste, Washington, DC, September.
                                       5-2

-------
20. U.S. EPA (1986), "Test Method 0010 Modified Method 5 Sampling Train," Test
   Methods for Evaluating Solid Waste, SW-846 [NTIS PB88-239223], Office of Solid
   Waste, Washington, DC, September. Also located at
   http ://www.   gov/epaoswer/hazwaste/test/pdfs/OO 10.pdf (Accessed February2004)
21. Kaune, A.;  Schramm, K.-W.; Henkelmann, B.; Kettrup, A.; Nikolai, U.; Zimmermann,
   R.; Boesl, U. (1996), "Pentachlorobenzene as Indicator for PCDD/F Emissions from a
   Hazardous Waste Incinerator: Effect of Using Active Carbon in the Flue Gas Cleaning,"
   Organohalogen Compounds, Vol. 27, pp. 159-162.
22. Kaune, A.;  Schramm, K.-W.; Kettrup, A.; Jaeger, K. (1996), "Indicator Parameters for
   PCDD/F in the Flue Gas of the Hazardous Waste Incinerator at Leverkusen, Germany,"
   Organohalogen Compounds, Vol. 27, pp. 163-166.
23. Kaune, A.;  Lenoir, D.; Schramm, K.-W.; Zimmermann, R.; Kettrup, A.; Jaeger, K.;
   Riickel, H.G.; Frank, F. (1998), "Chlorobenzenes and Chlorophenols as Indicator
   Parameters for Chlorinated Dibenzodioxins and Dibenzofurans in Incineration
   Processes: Influences of Various Facilities and Sampling Points," Environmental
   Engineering Science,  Vol. 15, pp. 85-95.
24. Tuppurainen, K.A.; Ruokojarvi, P.H.; Asikainen, A.H.; Aatamila, M.; Ruuskanen, J.
   (2000), "Chlorophenols as Precursors of PCDD/Fs in Incineration Processes:
   Correlations, PLS Modeling, and Reaction Mechanisms," Environmental Science and
   Technology, Vol. 34, pp. 4958-4962.
25. Zimmermann, R.; Heger, H.J.; Blumenstock, M.; Dorfner, R.; Schramm, K.-W.; Boesl,
   U.; Kettrup, A. (1999), "On-line Measurement of Chlorobenzene in Waste Incineration
   Flue Gas as a Surrogate for the Emission of Poly chlorinated Dibenzo-p-dioxins/Furans
   (I-TEQ) Using Mobile Resonance Laser lonization Time-of-Flight Mass Spectrometry,"
   Rapid Communications in Mass Spectrometry, Vol. 13, pp.  307-314.
26. Blumenstock, M.; Zimmermann, R.; Schramm, K.-W.; Kaune, A.; Nikolai, U.; Lenoir,
   D.; Kettrup, A. (1999), "Estimation of the Dioxin Emission (PCDD/F I-TEQ) from the
   Concentration of Low Chlorinated Aromatic Compounds in the Flue  and Stack Gas  of a
   Hazardous Waste Incinerator," Journal of Analytical and Applied Pyrolysis, Vol. 49, pp.
   179-190.
27. Blumenstock, M.; Zimmermann, R.; Schramm, K.-W.; Kettrup, A. (2000), "Influence  of
   Combustion Conditions on the PCDD/F-, PCB-, PCBz-, and PAH-Concentrations in the
   Post-Combustion Chamber of a Waste Incineration Pilot Plant," Chemosphere, Vol. 40,
   pp.987-993.
28. Blumenstock, M.; Zimmermann, R.; Schramm, K.-W.; Kettrup, A. (2001),
   "Identification of Surrogate Compounds for the Emission of PCDD/F (I-TEQ Value) and
   Evaluation  of Their On-Line Real-Time Detectability in Flue Gases of Waste
                                       5O

-------
   Incineration Plants by REMPI-TOFMS Mass Spectrometry," Chemosphere, Vol. 42, pp.
   507-518.
29. Sakurai, T.; Weber, R.; Ueno, S.; Nishino, J.; Tanaka, M. (2003), "Relevance of
   Coplanar PCBs for TEQ Emission of Fluidized Bed Incineration and Impact of Emission
   Control Devices," Chemosphere, Vol. 53, pp. 619-625.
30. Gullett, B.; Wikstrom, E. (2000), "Mono- to Tri-chlorinated Dibenzodioxin (CDD) and
   Dibenzofuran (CDF) Congeners/Homologues as Indicators of CDD and CDF Emissions
   from Municipal Waste and Waste/Coal Combustion," Chemosphere, Vol. 40, pp. 1015-
   1019.
31. Oser, H.; Thanner, R.; Grotheer, H.-H.; Gullett, B.K.; Natschke, D.; Raghunathan, K.
   (1998), "Lowly Chlorinated Dibenzodioxins as TEQ Indicators. A Combined Approach
   Using Spectroscopic Measurements with DLR Jet-REMPI and  Statistical Correlations
   with Waste Combustor Emissions," Combustion Science and Technology., Vol. 134, pp.
   201-220.
32. U.S. EPA, "EPA - TIN EMC Method 10 - Carbon Monoxide - NDIR,"
   http://www.epa.gov/ttn/emc/methods/method 10.html. (Accessed February 2004)
33. U.S. EPA, "Method 25b - Determination of Total Gaseous Organic Concentration Using
   a Nondispersive Infrared Analyzer,"
   http://www.epa.gov/ttn/enic/methods/niethod25b.htnil. (Accessed February 2004)
34. U.S. EPA, "Method 25A - Determination of Total Gaseous Organic Concentration Using
   a Flame lonization Detector," http://www.epa.gov/ttn/emc/methods/method25a.html.
   (Accessed February 2004)
35. U.S. EPA (1997), "EPA Performance Specification 9 - Specifications and Test
   Procedures for Gas Chromatographic Continuous Emission Monitoring Systems in
   Stationary Sources," Code of Federal Regulations, Title 40, Part 60, U.S. Government
   Printing Office, Washington, DC, July.
36. U.S. EPA (1997), "EPA Performance Specification 8 - Performance Specifications for
   Volatile Organic Compound Continuous Emission Monitors," Code of Federal
   Regulations, Title 40, Part 60, Appendix B, U.S. Government Printing Office,
   Washington, DC, July.
37. Ghorishi, S.B.; Whitworth, W.E.; Goldman, C.G; Waterland, L.R. (1997), "Testing the
   Performance of Real-Time Incinerator Emission Monitors," EPA-600/R-97-024 [NTIS
   PB97-142871], National Risk Management Research Laboratory, Cincinnati, OH,
   March.
38. Ryan, J.V.; Lemieux, P.M.; Preston, W.T. (1998), "Near-Real-Time Measurement of
   Trace Volatile Organic Compounds from Combustion Processes Using an On-Line Gas
   Chromatograph," Waste Management., Vol. 18, pp. 403-410.

                                       5-4

-------
39. Wada, E.T.; Sterling, A.M. (2000), "Development and Evaluation of a Mass
   Spectrometer-Based Continuous Emission Monitor for Volatile Organic Compounds,"
   International Conference on Incineration and Thermal Treatment Technologies,
   Portland, OR, May 8-12.
40. Lenoir, D.; Wehrmeir, A.; Sidhu, S.S.; Taylor, P.H. (2001), "Formation and inhibition of
   chloroaromatic micropollutants formed in incineration processes," Chemosphere, Vol.
   43, pp. 107-114.
41. Lemieux, P.M. (1994), "Pilot-Scale Evaluation of the Potential for Emissions of
   Hazardous Air Pollutants from Combustion of Tire-Derived Fuel," EPA-600-R-94-070
   [NTIS PB94-169463], Air and Energy Engineering Research Laboratory, Research
   Triangle Park, NC, April.
                                        5-5

-------
                                      TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
1. RE PORT NO.
      EPA-600/R-04/024
                                2.
                                                                3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
The Use of Surrogate Compounds as Indicators of PCDD/F
Concentrations in Combustor Stack Gases
             5. REPORT DATE
              February 2004
             6. PERFORMING ORGANIZATION CODE
7. AUTHORS
Paul M. Lemieux (EPA)
                                                                8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
See Block 12
                                                                10. PROGRAM ELEMENT NO.
                                                                11. CONTRACT/GRANT NO.

                                                                 In-house
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, North Carolina 27711
              13. TYPE OF REPORT AND PERIOD COVERED
              Final; APM 140
              14. SPONSORING AGENCY CODE

              EPA/600/13
15. SUPPLEMENTARY NOTES
The EPA Project Officer is Paul M. Lemieux, MD E305-01, phone (919) 541-0962.
16. ABSTRACT
Emissions of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDDs/Fs) from
stationary combustion sources are of concern because they are carcinogenic and may result in disruption of
endocrine systems in human and wildlife populations. PCDDs/Fs are typically present in only minute
concentrations in combustor stack gases, which makes sampling and analysis of these compounds
extremely expensive and time consuming. Direct,  real-time measurement of all PCDD/F isomers of concern
is not possible using current technology.  However, it is possible to estimate stack PCDD/F concentrations
by measuring other indicator (surrogate)  compounds that are present in the stack gases at much higher
concentrations. These appropriately selected surrogate compounds would be easier to  measure than
PCDDs/Fs, and could  be measured in real-time or near real-time by stack gas monitoring technology that is
currently available either commercially or that is in various stages of development. The  report discusses the
various surrogates that can be used to indicate PCDD/F concentrations, how those surrogates can be
measured, and a state-of-the-art assessment of availability and effectiveness of analyzers for measuring
those compounds.
17.
                                     KEYWORDS AND DOCUMENT ANALYSIS
             DESCRIPTORS
                                                   b. IDENTIFIERS/OPEN ENDED TERMS
                                                                               c. COSATI Field/Group
Air Pollution
Combustion Products
Dioxins
Furans
Substitutes
Measurement
Pollution Control
Stationary Sources
13B
21B
07C

14G
18. DISTRIBUTION STATEMENT
                                                   19. SECURITY CLASS (This Report)
                                                   Unclassified
                                                                               21. NO. OF PAGES
       Release to Public
20. SECURITY CLASS (This Page)
 Unclassified
                                                                               22. PRICE
EPA Form 2220-1 (Rev. 4-77 ) PREVIOUS EDITION IS OBSOLETE
                                                5-6 D
                          forms/admin/techrpt.frm  7/8/99 pad

-------