United States EPA-600/R-93-181
Environmental Protection September 1993
Agency
Research and
Development
EMISSION TEST REPORT
FIELD TEST OF CARBON INJECTION
FOR MERCURY CONTROL
CAMDEN COUNTY MUNICIPAL
WASTE COMBUSTOR
Prepared for
Office of Air Quality Planning and Standards
Prepared by
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
-------
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
ABSTRACT
In 1992, the U.S. Environmental Protection Agency conducted a parametric
testing project to evaluate the use of powdered activated carbon for removing volatile
pollutants from municipal waste combustor (MWC) flue gas. This testing was conducted
at the spray dryer absorber/electrostatic precipitator (SD/ESP)-equipped MWC in
Camden County, New Jersey. The primary test objectives were to evaluate the effect of
carbon type, carbon feed rate, carbon feed method, and ESP operating temperature on
emissions of mercury (Hg) and chlorinated dioxins and furans (CDD/CDF), and to
assess the impact of carbon injection on the paniculate matter control performance of
the ESP. Secondary objectives were to examine the impact of carbon injection on
emissions of other metals and volatile organic compounds (VOC). The testing included
operation of three different carbon injection systems and examined 16 different SD/ESP
and carbon injection system operating conditions. The test was conducted as a follow-on
to an EPA-funded testing program at a SD/fabric filter-equipped MWC that focused on
the performance of carbon injection for controlling Hg emissions.
The test results indicate that carbon injection upstream of an SD/ESP could
achieve high levels (greater than 90%) of Hg and CDD/CDF reduction. Key system
operating parameters are carbon feed rate, carbon feed method, and ESP temperature.
No detrimental impacts on ESP performance were identified. The study also found that
carbon injection does not have a significant impact on emissions of the other metals
examined or of VOC.
n
-------
EPA-600/R-93-181
September 1993
EMISSION TEST REPORT
FIELD TEST OF CARBON INJECTION FOR MERCURY CONTROL
CAMDEN COUNTY MUNICIPAL WASTE COMBUSTOR
Prepared by:
D.M. White, W.E. Kelly, MJ. Stucky
J.L. Swift, M.A. Palazzolo
Radian Corporation
1300 E. Chapel Hill Road/Nelson Highway
P.O. Box 13000
Research Triangle Park, North Carolina 27709
EPA Contract Nos./Work Assignment Nos.
68-W9-0069/25
68-D9-0054/71
Project Officer:
James D. Kilgroe
U.S. Environmental Protection Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, p.C. 20460
-------
TABLE OF CONTENTS, continued
Section Page
4.0 CARBON INJECTION PARAMETRIC TESTING 4-1
4.1 Carbon Feed System Data 4-1
4.2 Combustor Operating Data 4-1
4.3 Spray Dryer Absorber/Electrostatic Precipitator Operating Data .... 4-6
4.4 Mercury '. 4-6
4.5 Cadmium and Lead 4-12
4.6 Other Metals 4-14
4.7 Paniculate Matter 4-21
4.8 CDD/CDF 4-21
4.9 Volatile Organic Compounds 4-27
4.10 Fly Ash Carbon Content 4-30
4.11 Volumetric Flow and Moisture by EPA Methods 1 and 4 4-30
4.12 Continuous Emissions Monitoring Data 4-34
5.0 ELECTROSTATIC PRECIPITATOR PERFORMANCE TESTING 5-1
5.1 Carbon Feed System Data 5-1
5.2 Combustor Operating Data 5-5
5.3 Spray Dryer Absorber/Electrostatic Precipitator Operating Data .... 5-5
5.4 Mercury 5-8
5.5 Cadmium and Lead 5-8
5.6 Paniculate Matter 5-11
5.7 Fly Ash - Percent Carbon 5-11
5.8 Particle Size Distribution 5-14
5.9 Volumetric Flow and Moisture by EPA Methods 1 and 4 5-16
5.10 Continuous Emission Monitoring Data 5-16
6.0 FLUE GAS SAMPLING AND ANALYTICAL PROCEDURES 6-1
6.1 Paniculate Matter and Multiple Metals 6-1
6.2 CDD/CDF 6-8
6.3 Volatile Organic Compounds 6-16
6.4 Fly Ash Carbon Content 6-20
6.5 Particle Size Distribution 6-20
6.6 Volumetric Flow Rate and Moisture Content 6*24
6.7 Continuous Emission Monitors 6-25
6.8 Process Data Collection 6-26
IV
-------
TABLE OF CONTENTS
Section
Abstract
Figures [[[ V1
Tables
Acknowledgements ............................................. ' '
Conversion Factors
1.0 SUMMARY
1.1 Introduction ........................................... 1*1
1.2 Test Objectives ......................................... I'2
1.3 Test Design ............................................ 1-3
1.4 Conclusions ............................................ 1-3
1.5 Apparent Data Gaps ..................................... 1-5
1.6 Report Organization ..................................... 1-6
2.0 TEST DESIGN ............................................... 2-1
2.1 Description of the Camden County Municipal Waste Combustor .... 2-1
2.2 Test Matrix ............................................ 2-3
2.3 Carbon Feed Systems .................................... 2-9
2.4 Description of Tested Carbons .............................. 2-14
2.5 Sampling Locations .................................. .... 2-14
2.6 Sampling and Analytical Methods ........................... 2-19
3.0 INTERPRETATION OF RESULTS .............................. 3-1
3.1 Data Summary ......................................... 3-1
3.2 Mercury .............................................. 3-1
3.3 Other Metals ........................................... 3-19
3.4 CDD/CDF ............................................ 3-22
3.5 Volatile Organic Compounds ............................... 3-24
3.6 Acid Gases ............................................ 3-25
-------
FIGURES
Figure Page
2-1 Schematic of the Camden County Municipal Waste Combustor 2-2
2-2 Dry Carbon Injection System 2-11
2-3 Short Duration Slurried Carbon Injection System 2-13
2-4 Economizer Outlet Flue Gas Sample Location 2-20
2-5 Stack Flue Gas Sample Location 2-21
3-1 Effect of Carbon Type on Mercury Reduction 3-6
3-2 Effect of Injected Carbon Concentration on Mercury Reduction 3-6
3-3 Effect of Injected Carbon Concentration on Mercury Emissions 3-8
3-4 Effect of Total Carbon Concentration on Mercury Reduction 3-9
3-5 Effect of Total Carbon Concentration on Mercury Emission 3-10
3-6 Effect of Carbon Feed Method on Mercury Reduction 3-10
3-7 Effect of Carbon Feed Method on Mercury Emissions 3-12
3-8 Effect of Carbon Retention Time in Lime Slurry on Mercury Reduction .... 3-13
3-9 Effect of ESP Temperature on Mercury Reduction 3-13
3-10 Effect of Paniculate Matter Control on Mercury Reduction 3-15
3-11 Regression Analysis Results for Mercury Reduction 3-17
3-12 Predicted Mercury Reduction Variability with Dry Carbon
Injection at 270ฐF 3-17
3-13 Regression Analysis Results for Mercury Emissions 3-20
3-14 Effect of Carbon Injection on CDD/CDF Reduction 3-23
3-15 Effect of Carbon Injection on CDD/CDF Concentration 3-23
3-16 Effect of Carbon on VOC Reduction 3-24
vi
-------
TABLE OF CONTENTS, continued
Section
7.0 QUALITY ASSURANCE/QUALITY CONTROL 7'1
7.1 Overview of Data Quality 7~2
7.2 Sampling Quality Control 7'2
7.3 Sample Storage and Holding Time 7'12
7.4 Analytical Quality Control 7'13
7.5 Continuous Emission Monitors
8.0 REFERENCES
8-1
-------
TABLES
Table Page
2-1 Unit B Test Matrix 2-5
2-2 Unit A Test Matrix 2-8
2-3 Description of Activated Carbons Tested 2-15
2-4 Sampling Matrix 2-20
2-5 Sampling Times, Minimum Sampling Volumes, and Detection Limits 2-21
3-1 Summary of Test Conditions and Mercury Test Results 3-2
3-2 Average SD/ESP Removal Efficiency for Selected Test Conditions 3-21
4-1 Carbon Feed System Data for Unit B 4-2
4-2 Unit B Combustor Operating Data 4-4
4-3 Unit B Spray Dryer Absorber/ESP Operating Data 4-7
4-4 Unit B Mercury Results 4-10
4-5 Unit B Cadmium and Lead Results 4-13
4-6 Unit B Other Metal Results 4-15
4-7 Unit B Particulate Matter Results 4-22
4-8 CDD/CDF Results 4-24
4-9 Frequency of VOCs Detected in Tube Pairs 4-28
4-10 Volatile Organic Compound Results 4-29
4-11 Unit B Fly Ash Carbon Results 4-31
4-12 Unit B Volumetric Flow and Moisture Results 4-32
4-13 Unit B CEM Results 4-35
Vlll
-------
FIGURES, continued
Figure
3-17 Effect of Carbon Injection on SO2 Reduction 3'26
3-18 Effect of Carbon Injection on NOX Concentration 3-26
3-19 Changes in ESP Performance Over Time 3'27
6-1 Schematic of Multiple Metals Sampling Train 6-2
6-2 Metals Sample Recovery Scheme 6-5
6-3 Metals Sample Preparation and Analysis Scheme 6-7
6-4 CDD/CDF Sampling Train Configuration 6-10
6-5 CDD/CDF Field Recovery Scheme 6-12
6-6 Extraction and Analysis Schematic for CDD/CDF Samples 6-15
6-7 Schematic of VOST Sampling Train 6-18
6-8 VOST Analysis Protocol 6-21
6-9 Sampling Train for Particle Size Distribution Tests 6-23
vn
-------
TABLES, continued
Table Page
7-8 Audit Sample Results for Metal Analysis, Phase II 7-23
7-9 Surrogate Recovery Results for CDD/CDF Phase II 7-26
7-10 CDD/CDF Method Spike Results, Phase II 7-27
7-11 CDD/CDF Flue Gas Blank Results, Phase II 7-29
7-12 CDD/CDF Audit Results 7-30
7-13 VOST Field Blank Results, Condition BIO, Phase II 7-32
7-14 VOST Field Blank Results, Condition fill, Phase II 7-33
7-15 Surrogate Recovery Results and Hold Times for Inlet VOST 7-35
7-16 Surrogate Recovery Results and Hold Times for Outlet VOST 7-38
7-17 VOST Method Spike Recovery Results, Precision and Accuracy
VOST Analyses 7-41
7-18 VOST Audit Results, Phase U 7-42
7-19 CEM Daily Calibration Checks, Unit A 7-44
7-20 CEM Daily Calibration Checks, Unit B 7-45
-------
TABLES, continued
Table
5-1 Unit A Carbon Feed System Data 5'2
5-2 Unit A Long Term Carbon Feed Data 5'3
5-3 Unit A Combustor Operating Data 5-6
5-4 Unit A Spray Dryer Absorber/ESP Operating Data 5-7
5-5 Unit A Mercury Results 5-9
5-6 Unit A Cadmium and Lead Results 5-10
5-7 Unit A Particulate Matter Results 5-12
5-8 Unit A Fly Ash Carbon Results 5-13
5-9 Unit A Particle Size Distribution Data 5-15
5-10 Unit A Volumetric Flow and Moisture Results 5-17
5-11 Unit A CEM Results 5-18
6-1 CDD/CDF Sample Fractions Shipped to Analytical Laboratory 6-13
6-2 Target CDD/CDF Congeners 6-14
6-3 Volatile Compounds Quantified by SW-846 Method 8240 (VOST) 6-17
7-1 Comparison to Quality Control Objectives 7-3
7-2 Metals Stack Sampling Quality Control Data 7-5
7-3 CDD/CDF Stack Sampling Quality Control Data 7-10
7-4 VOST Stack Sampling Quality Control Data 7-11
7-5 Matrix Spike Results for Mercury in Flue Gas 7-15
7-6 Matrix Spike Results for Metals in Flue Gas, Phase I 7-17
7-7 Matrix Spike Results for Metals in Flue Gas, Phase II 7-19
ix
-------
CONVERSION FACTORS
Area
Density
Energy
Force
Length
Mass
Mass Concentration
Power
Pressure
Temperature
Volume
Volumetric Flow
Weight
To Convert From
ft2
lbm/ft3
Btu
Ibf
ft
in.
in.
Ibm
Ibm
gr
gr/ft3
ft - Ibf/s
lbf/in.2
oF
ft3
ftVs
ton
To
m2
kg/m3
J
N
m
m
mm
kg
g
g
g/m3
W
Pa
ฐC
m3
m3/s
Mg
Multiply By
9.2903E-2
1.6019E+1
1.055 IE +3
4.4482
3.048E-1
2.5400E-2
2.540E+1
4.535E-1
4.535E+2
6.486E-2
2.29
1.3558
6.895E+3
5/9(TF - 32ฐ)
2.8317E-2
2.8317E-2
1.10
Xll
-------
ACKNOWLEDGEMENTS
The U.S. Environmental Protection Agency's Air and Energy Engineering
Research Laboratory (AEERL) would like to acknowledge the technical assistance and
cooperation given by the following groups and individuals. Without their help, this
project would not have been possible.
Foster-Wheeler Power Systems, Inc. and the Camden County Resource
Recovery Authority who supported the project through involvement in test
planning, allowing use of their facility, and providing the carbon injection
equipment and carbon used during the testing. Specific appreciation is
extended to Bruce Studley, Newt Wattis, Steve Warlick, Mike Cooper, and
Billy Pfoutz, all of Foster-Wheeler.
Ed Weaver and Tony Santacana of Belco Technologies Corporation for
assistance in planning and execution of the project.
Joy Environmental Technologies, Inc. for their funding of additional metals
analyses.
Ted Brna of AEERL for his assistance in the design and execution of the
field testing effort.
XI
-------
systems already in existence, as well as the potential for some existing MWCs currently
equipped with an ESP only to retrofit a SD or other control technology upstream of the
existing ESP to reduce acid gas and organic emissions.
In addition, very little data are available from either SD/FF or SD/ESP-equipped
MWCs on the effectiveness of carbon injection for reducing emissions of polychlorinated
dibenzo-p-dioxins and polychlorinated dibenzofurans '(CDD/CDF) and various volatile
organic compounds (VOC).
1.2 Test Objectives
To help develop a better understanding of the effectiveness of carbon injection in
reducing emissions of Hg, CDD/CDF, and VOC from MWCs, the EPA's Air and Energy
Engineering Research Laboratory (AEERL) conducted a series of tests at the Camden
County MWC in Camden, New Jersey. The objectives of these tests were to evaluate:
The level of Hg reduction achievable by carbon injection at
SD/ESP-equipped MWCs;
The extent to which emissions of other metals, CDD/CDF, and VOC can
also be reduced by carbon injection;
Whether carbon characteristics (particle size, pore size) or injection
method (dry powder, lime slurry) are important in SD/ESP systems;
Whether carbon residence time in lime slurry affects carbon performance;
Whether PM collection efficiency and operating temperature of the ESP
have a significant impact on Hg collection; and
Whether carbon injection has any detrimental impacts on the paniculate
matter (PM) collection performance of an ESP.
1-2
-------
1.0 SUMMARY
1.1 Introduction
The 1990 Clean Air Act Amendments require the U.S. Environmental Protection
Agency (EPA) to promulgate mercury (Hg) emission limits for municipal waste
combustor (MWC) facilities.1 To comply with this requirement, the EPA has gathered
data from MWCs to provide background information on various Hg control devices and
technologies.2 Most of the data on Hg control methods, including testing funded by EPA
in 1991 at the Ogden Martin Systems of Stanislaus, Inc. (OMSS) MWC, are from units
equipped with spray dryer absorber/fabric filter (SD/FF) systems.2"7
Data from SD/FF-equipped systems indicate that over 90 percent reduction in Hg
concentrations is achievable by adsorption of Hg onto carbon particles in the flue
gas.2J>'6'7 Based on available data, it appears the source of carbon can be residual carbon
present in fly ash emitted from the combustion system or commercially manufactured
activated carbon injected into the flue gas.2-4'6"8 The testing of activated carbon injection
at the OMSS MWC found that for SD/FF-equipped MWCs the carbon feed rate was the
primary factor affecting Hg control.4'6'7 The OMSS testing also indicated there are no
significant differences in Hg control performance as a function of the physical
characteristics of the carbon (original material, particle size, pore size, and density), the
method of injection (as a dry powder or mixed with SD slurry), or the location of
injection (economizer exit, SD inlet, and into the SD).4'6'7 It is not possible to determine
from these tests, however, how much of the Hg removal from the flue gas is achieved in
the SD versus the FF.
There are very little data available on Hg control at MWCs equipped with spray
dryer absorber/electrostatic precipitator (SD/ESP) systems.2'6 Because of uncertainties
regarding the mechanisms of Hg capture by carbon, it is not possible to directly translate
the data collected at SD/FF-equipped MWCs to units equipped with SD/ESP systems.
Data on the collection of Hg by SD/ESPs are of interest because of the number of such
1-1
-------
Hg reductions exceeding 90% are achievable by injection of dry carbon at
both of the ESP operating temperatures examined (270T and 350ฐF)."
The most important process variables affecting Hg emissions are carbon
feed rate, injection method, and ESP operating temperature.
The amount of unburned carbon present in fly ash plays a significant role
in determining baseline Hg emissions.
Carbon characteristics are not significant in determining Hg control
efficiency when carbon is injected as a dry powder. Carbon characteristics
may be important, however, if carbon is injected as a slurry.
Slurry injection of carbon is less effective in reducing Hg emissions than
dry injection. This conclusion is in contract with the results of the OMSS
testing and may be due to the performance characteristics of an ESP versus
a FF, to differences in carbon properties, or some other unknown
phenomena.
Assuming a baseline Hg removal efficiency of 30% by a SD/ESP without
carbon injection, the average reduction can be increased to 90% by
injecting approximately 200 mg of carbon per dry standard cubic meter
(mg/dscm). This injection rate is approximately triple the rate needed to
achieve 90% Hg reduction by a SD/FF-equipped MWC with similar
baseline Hg levels.
Injection of carbon can reduce stack emissions of CDD/CDF by over 75%.
However, there is no apparent effect of carbon injection on emissions of
VOC.
Emissions of other metals other than Hg are primarily associated with PM
and their control is determined primarily by the efficiency of the PM
control device. Possible exceptions to this relationship are molybdenum
and selenium. There is no apparent in reduction emissions of these metals
from carbon injection.
There is no apparent impact of carbon injection on the ESP's PM control
efficiency.
"English Engineering units were used in measurements during testing and are
customarily used at MWC facilities in the U.S. Conversion factors from English Engineering
to SI units are given at the end of the front matter.
1-4
-------
1.3 Test Design
To achieve the objectives stated above, the test program was divided into three
distinct testing efforts that were conducted during two testing phases. Phase I was
designed to provide baseline information on Hg control levels as a function of carbon
type and feed rates. To accomplish this objective, five days of testing were conducted at
baseline conditions and with two different carbon types and feed rates. This information
was used to select the carbon type and feed rates for Phase II.
Phase II was separated into two sections, parametric testing and ESP performance
testing. The parametric tests evaluated the impact of key carbon injection system
operating variables on emissions of Hg, other metals, and organic compounds. Specific
parameters of interest to the test design were:
Carbon feed rate;
Carbon injection method (as a dry powder and mixed with lime slurry);
ESP operating temperature; and
Number of ESP fields.
To accomplish this test program, eight test conditions were conducted.
The ESP performance testing was designed to evaluate whether there are any
detrimental impacts on ESP performance due to carbon injection over an extended time
period and to assess the relationship between PM collection efficiency and Hg control.
To satisfy these objectives, 5 days of sampling were conducted over a 13-day period.
1.4 Conclusions
Based on the data collected during the Camden County tests, the following
conclusions were reached:
1-3
-------
1.5.2 Evaluation of Carbon Injection on Other Combustion Sources
Previous testing of activated carbon injection has focused on combustion of
municipal waste and, to a lesser extent, medical waste. Based on the magnitude of
potential air toxic emissions from other combustion sources, testing of activated carbon
injection on other major stationary sources of Hg or VOC may be desirable. For
example, it may be beneficial to examine injection of activated carbon into the flue gas
from a coal-fired boiler of sewage sludge incinerator during future testing.
1.5.3 Development of an Activated Carbon Injection Process Model
There are sufficient data available to define key process parameters affecting the
performance on activated carbon injection. Development of a computerized process
model could be useful in better defining the relationship between and importance of key
parameters. For example, the data from the Camden County and OMSS tests provide
contradictory information on the performance of activated carbon when injected into
MWC flue gas as a slurry. Key uncertainties in defining the cause of these differences
relate to understanding wetted carbon behavior (e.g., surface wetting and pore pluggage);
the impact of SD design on slurry particle agglomeration, mixing, and reactor vessel
residence time; and the amount of Hg collection possible in a SD reactor and ESP
versus that occurring within the bag cake of a FF. Availability of a computerized process
model addressing these issues may enhance knowledge of how to most effectively apply
carbon injection technology.
1.6 Report Organization
The remainder of the report is divided into six section. Section 2 describes the
Camden County facility, the test matrix, carbon feed system, characteristics of the tested
carbons, and the sampling locations. Section 3 summarizes the collected process and
flue gas data, and. interprets the test data in light of the project objectives. Sections 4
and 5 provide details on the collected process and flue gas data for Units B and A,
1-6
-------
1.5 Apparent Data Gaps
The data collected during carbon injection testing at the Camden County MWC
and during the earlier testing at the OMSS MWC indicate that carbon injection
upstream of a SD/ESP or SD/FF is an effective control technique for reducing Hg
concentrations in MWC flue gas. There remain, however, a number of unanswered
questions regarding the potential performance of carbon injection when applied to other
types of combustors and air pollution control systems. For example, the Camden County
data indicate that carbon injection may be a viable technique for reducing emissions of
Hg, and potentially CDD/CDF, from some MWCs equipped with an ESP only. It is
unclear, however, whether carbon injection can be used to control emissions of volatile
metals and organics from other combustion sources, such as coal-fired boilers, that have
significantly different flue gas characteristics. These questions suggest apparent data
gaps in three primary areas.
1.5.1 Fundamental Studies on Carbon Adsorption of Speciated Mercury Compounds
Most of the Hg in MWC flue gas is present as a mercuric (+2) ion. Ionic Hg
may be more readily adsorbed onto untreated carbon particles, such as those used in the
Camden County and OMSS tests, than is elemental Hg. In combustion sources having
lower chlorine and/or higher sulfur contents, a greater portion of the total Hg in flue gas
is expected to be in the elemental form. In these situations, use of carbons that have
been impregnated with iodine, sulfur, or chlorine compounds to improve adsorption of
elemental Hg may be of value. The effects of Hg speciation and carbon properties on
Hg capture have not been defined by laboratory and field testing.
1-5
-------
respectively. Section 6 describes the sampling and analytical procedures used during the
study, and Section 7 provides summary statistics and discussion regarding measures taken
to control and assess data quality. Backup material from the field testing, laboratory
analysis, and statistical analyses of the data used in preparing this report are in separate
appendices. Backup materials from the field testing program and from subsequent
analytical and statistical analyses used to prepare this report are not included here
because of the large amount of material involved. This material has been placed in the
EPA's public docket on MWC standards development (A-90-45) by EPA's Office of Air
Quality Planning and Standards.
1-7
-------
NJ
Stack
I.D. Fan
Electrostatic Spray Ash Conveyors
Precipltator Dryer
Absorber
LEGEND
A : inlet Sampling Location
B : Dry Carbon Injection Location
C: Slurry Carbon Injection Location
D: Outlet Sampling Location
Figure 2-1. Schematic of the Camden County
Municipal Waste Combustor
Tipping Floor
S
8
K
(O
-------
2.0 TEST DESIGN
2.1 Description of the Camden County Municipal Waste Cnmbustor
The Camden County MWC is owned and operated by Camden County Energy
Recovery Associates, a subsidiary of Foster Wheeler Power Systems, Inc. It is located in
Camden, New Jersey, and began operating in 1991. The facility contains three identical
mass-burn waterwall combustion units, designated as Units A, B, and C. Each unit is
capable of burning 350 tons per day of municipal solid waste (MSW). The grate firing
system used in each of the three units was supplied by Detroit Stoker Company. The
MSW burned at the facility is supplied by the city of Camden and surrounding towns.
Steam produced by the facility powers two 17-MW turbogenerators and electricity (or
steam) can also be sold directly to area industries. The facility is designed to process
1,050 tons of MSW per day, 365 days per year. A general schematic of each unit is
shown in Figure 2-1.
The air pollution control system on each combustor consists of a Belco SD
(licensee of Deutsche Babcock) and a Belco five-field ESP. Flue gas from the
combustor leaves the economizer and enters a vertical 76-inch inner diameter (ID)
circular duct. The flue gas travels down the duct, through a 90-degree elbow, and into a
cyclone located at the base of the SD. The cyclone separates coarse PM from the flue
gas and distributes flue gas to six vertical flow tubes that connect to the base of the SD
vessel. A two-fluid nozzle located at the top of each flow tube is used to inject lime
slurry upward into concurrently flowing flue gas for removal of acid gases. The lime
slurry flow rate is controlled by the stack SO2 concentration. Dilution water flow rate is
controlled by the reaction chamber exit temperature. The flue gas then proceeds upward
through the vertical SD reaction chamber and exits through a 64-inch outer diameter
(OD) circular duct. This duct makes a ISO-degree turn and the flue gases are directed
downward into the five-field ESP. During normal operation, only four of the ESP fields
are in operation, with the fifth field providing spare capacity in case of operating
problems or maintenance on one of the other fields.
2-1
-------
information on Hg control performance as a function of carbon type and carbon feed
rate. This information was used to define operating conditions during the Phase II tests.
The Phase II tests included two distinct efforts. One of these efforts focused on
parametric testing designed to provide data on the impact of key carbon injection system
operating variables on Hg control efficiency. The other Phase II testing effort examined
the impact of extended carbon injection on ESP performance and of PM collection
efficiency on Hg control.
The design of these three testing efforts is discussed below. With one exception
(Condition 4A), triplicate sampling runs were conducted at each test condition. One test
condition was completed per day. During all of the tests, the plant's process and
continuous emissions monitoring equipment was used to monitor combustor and SD/ESP
operating conditions.
2.2.1 Phase I - Characterization Testing
Phase I included the five test conditions listed in the upper portion of Table 2-1.
All testing in this phase was conducted on Unit B. During these tests, carbon was
injected as a dry powder into the flue gas duct just prior to the cyclone located at the
base of the SD. This location was selected because it was expected to provide sufficient
time and turbulence for good mixing of the carbon into the flue gas. Based on the fine
particle size of the injected carbon and preliminary experiments conducted prior to
Phase Bl tests, it is expected that the cyclone removed little if any of the injected
carbon.
The objective of these tests was to assess Hg control levels for two different
carbon types and feed rates. Both of the carbons, Darco FGD and Darco PC-100, were
produced by American Norit Company and were used during the OMSS MWC testing.4
A description of each carbon is provided in Section 2.4. The lower target carbon feed
rate was 4.5 kilograms per hour (kg/hr) (10 pounds per hour [lb/hr]), which equates to a
flue gas concentration of 60 mg of carbon per dry standard cubic meter of flue gas
2-4
-------
Typical operating conditions for the air pollution control system are:
Economizer exit temperature, 450 to 480ฐF;
SD exit temperature, 280ฐF;
Lime slurry flow, 6 to 7 gpm;
Lime slurry specific gravity, 1.08;
Dilution (cooling) water flow, 6 to 11 gpm;
Economizer exit SO2 concentration, 125 to 200 ppmv; and
ESP exit SO2 concentration, 20 to 40 ppmv.
The lime slurry feed rate can be automatically controlled to obtain a specified SO2 outlet
concentration or the lime slurry feed rate controller can be manually set to provide a
constant feed rate.
The gases from each ESP are ducted into a separate flue in the stack. The stack
contains four 72-inch ID, circular flues: one for each of the three operating units and
one reserved for a future unit. The stack exit is approximately 366 feet above ground
level.
The process control systems include a Bailey Net 90 (INFI90) for the boiler, a
separate control and data display system for the SD/ESP, and two separate data system
for the plant's continuous emissions monitoring system (CEMS). The CEMS equipment
includes extractive monitors for SO2 and O2 at the economizer exit and for O2, CO2,
H2O, CO, THC, CH4, SO2, HC1, and NOX in the stack, and a stack opacity monitor.
2.2 Test Matrix
The Camden County MWC test program encompassed three distinct testing
efforts and was conducted in two phases. Phase I was designed to provide baseline
2-3
-------
(mg/dscm) corrected to 7% O2." This is approximately equal to the high carbon feed
rate tests conducted at OMSS. The higher target carbon feed rate of 27 kg/hr (60 Ib/hr)
equates to approximately 360 mg/dscm of flue gas and was believed to be sufficiently
high to ensure Hg removal efficiencies in excess of 90% and emission levels less than
100 //g/dscm.
During each run, simultaneous sampling was conducted at the economizer exit
and in the stack for total PM and Hg using the multiple metals sampling train. Each
sampling run was one hour in duration (excluding port changes and any equipment
problems). In addition, a Method 5 type sampling train was operated at the economizer
exit to collect a daily composite sample of PM. The composite sample was then used for
determination of percent carbon in the fly ash resulting from incomplete combustion.
Both carbon types indicated similar levels of Hg control during the Phase I
testing. Based upon these results, the less expensive Darco FGD was selected as the
carbon for the Phase II testing.
2.2.2 Phase II - Parametric Testing
The Phase II parametric testing included eight test conditions and was designed to
evaluate the impact of carbon feed rate, carbon feed method, and flue gas temperature
on Hg control. The test conditions are described in the lower portion of Table 2-1. All
of these tests were conducted on Unit B.
During each run, simultaneous sampling was conducted at the economizer exit
and in the stack for total PM and Hg using the multiple metals sampling train. During
six of the test conditions (five from Phase H and one from Phase I), the sampling
fractions collected by the multiple metals train were analyzed for 16 other metals. These
" Based on a flue gas flow rate of 75,000 dscm per hour. Unless otherwise noted, all
flue gas flow rates used in this document are based on correction to standard conditions
[20ฐC (68ฐF) and 101.3 kPa (14.7 psia)] and 7% O2.
2-6
-------
TABLE 2-1 UNIT B TEST MATRIX
CAMDEN COUNTY MWC (1992)
Condition
Bl
B2
B3
B4
B5
B6
B7
B8
B9
BIO
Bll
B12
B13
Phase
I
I
I
I
I
II-PT
II-PT
II-PT
II-PT
II-PT
II-PT
II-PT
II-PT
ESP
Temperature
(ฐF)
270
270
270
270
270
350
350
270
270
270
270
270
270
Number of
ESP Fields
4
4
4
4
4
4
4
4
4
4
4
4
4
Carbon
Type
None
FGD
PC- 100
PC- 100
FGD
None
FGD
FGD
FGD
None
FGD
FGD
FGD
Carbon
Feed
Method
Dry
Dry
Dry
Dry
~
Dry
Dry
Dry
Dry
Slurry
Slurry
Carbon
Feed Rate
(Ib/hr)
~
10
10
60
60
50
25
5
50
50
25
Sample Analytes
Hg, PM, %C
Metals, PM, %C
Hg, PM, %C
Hg, PM, %C
Hg, PM, %C
Hg, PM, %C
Metals, PM, %C
Hg, PM, %C
Hg, PM, %C
Metals, PM, %C,
CDD/CDF, VOC
Metals, PM, %C,
CDD/CDF, VOC
Metals, PM, %C,
CDD/CDF
Metals, PM, %C
-------
TABLE 2-2 UNIT A TEST MATRIX
CAMDEN COUNTY MWC
Condition
Al
A2
A3
A4
A5
Phase
II-ESP
II-ESP
II-ESP
II-ESP
II-ESP
ESP
Temperature
(ฐF)
270
270
270
270
270
Number of
ESP Fields
4
4
4
4
3
Carbon
Type
None
FGD
FGD
FGD
FGD
Carbon
Feed
Method
--
Slurry
Slurry
Slurry
Slurry
Carbon Feed
Rate
(Ib/hr)
~
50
50
50
50
Sample Analytes
Hg, Cd, Pb, PM,
%C, PSD
Hg, Cd, Pb, PM,
%C, PSD
Hg, Cd, Pb, PM,
%C, PSD
Hg,Cd,Pb,PM,
%C, PSD
Hg, Cd, Pb, PM,
%C, PSD
oo
-------
metals included cadmium (Cd), lead (Pb), antimony (Sb), arsenic (As), barium (Ba),
beryllium (Be), chromium (Cr), cobalt (Co), copper (Cu) manganese (Mn),
molybedum (Mo), nickel (Ni), selenium (Se), silver (Ag), thallium (Ti) and
vanadium (V). In addition, a Method 5 type sampling train was operated at the
economizer exit to collect a daily composite sample of PM for determination of percent
carbon in the fly ash. Except for the three test conditions discussed below, the planned
sampling durations for each sampling run was one hour long.
The testing also included sampling for CDD/CDF during Conditions BIO, Bll,
and B12, and for VOC during Conditions BIO and Bll. Each sampling run during these
three test conditions was two hours in duration.
2.2.3 Phase II - Electrostatic Precipitator Performance Testing
The other objectives of the Phase II testing were to evaluate whether there are
any detrimental impacts on ESP performance due to carbon injection over an extended
time period, and to assess the relationship between PM collection efficiency and Hg
control. To satisfy these objectives, five days of sampling were conducted over a 12-day
period on Unit A. Following an initial day of testing without carbon injection that was
used to establish baseline performance, FGD carbon was added to the lime slaking tank
and continuously fed as a slurry into the spray dryer.
As shown in Table 2-2, the first four days of sampling were conducted with four
ESP fields in service. These tests were run on the day prior to the start of carbon
injection and on the first, third, and eighth days after the start of carbon injection. After
completion of testing on the eighth day, the last ESP field was turned off, thus resulting
in operation with only three fields. On the fourth day after the unit had been operating
with three fields, the fifth day of sampling was conducted. The delay in sampling until
the fourth day after reducing the ESP to three-field operation was designed to allow the
" Unless otherwise noted, all run durations mentioned in the report are actual
sampling times and exclude port changes and equipment problems.
2-7
-------
transport system. Carbon feed rates were controlled by adjusting the screw feeder speed.
A schematic of this system is shown in Figure 2-2. The metered carbon passed out the
end of the screw feeder tube and dropped into a funnel connected to the pneumatic
transport system. The transport system consisted of a Fox Air Eductor to provide
air/carbon mixing and a flexible transport hose connecting the eductor to the injection
probe. Transport air was supplied by the plant compressed air system.
The carbon injection probe consisted of a 1-inch pipe inserted into the side of the
90ฐ elbow located just prior to the cyclone. The end of the probe was cut at a 45ฐ angle,
which faced downstream. The end of the probe was located five inches below the duct
centerline. The off-center location was chosen to avoid a downstream obstruction and to
take advantage of the turbulence created by the 90ฐ turn in the flue gas flow. The
cyclone also provided additional turbulence for mixing and equalized the distribution of
flue gas flow to the SD. The cyclone was sized to remove only large particles in the flue
gas, and was expected to have negligible removal of the injected carbon.
Prior to the start of testing, the feeder was calibrated by recording the voltage
applied to the screw feeder DC motor over a range of voltages and the corresponding
mass, feed rate of carbon. Based on these data, a calibration curve was developed.
During each test condition, the desired motor voltage was set. In addition, the carbon
level in the feeder hopper was regularly monitored. When the carbon level fell to a
preset point, the hopper was refilled. By recording the amount of carbon added and the
time between refilling, the carbon feed rate was confirmed.
At the end of each testing day, the carbon feed rate was adjusted to the target
level for the next day of testing. The feeder then operated overnight at this rate to
condition the SD/ESP prior to the start of the next day of testing.
2-10
-------
ESP to reach equilibrium with regard to PM collection efficiency. The three-field tests
were conducted to evaluate probable carbon injection effects on MWCs with smaller
ESPs than at Camden.
During each run, simultaneous sampling was conducted at the economizer exit
and in the stack for total PM, Hg, Cd, and Pb using the multiple metals sampling train.
At the stack sampling location during each run, two eight-stage Andersen impactors were
operated to evaluate the particle size distribution (PSD) of emitted PM. In addition, a
Method 5 type sampling train was operated at the economizer exit to collect a daily
composite sample of PM for determination of percent carbon in the fly ash. The PM
and PSD data provided a direct indication of whether degradation in ESP performance is
associated with carbon injection. Because of enrichment of Cd and Pb onto fine
participate, these two elements are expected to be sensitive indicators of degraded
performance. Each metals train sampling run was one hour long. The two PSD trains
were run throughout each test day to collect sufficient particulate for quantitative
measurement of the weight gain by each impactor stage.
2.3 Carbon Feed Systems
Carbon was fed to Units A and B by two different methods using three different
injection systems. The testing on Unit B included injection of dry carbon and addition of
carbon into a slurry mix tank installed just prior to the SD. Carbon was injected into
Unit A by addition of carbon to lime slurry in the plant's existing lime slurry feed tank.
2.3.1 Dry Carbon Feed System
The primary carbon feed method used during the Unit B testing was injection of
dry carbon into the flue gas duct just upstream of the SD inlet cyclone. The dry
injection system consisted of a K-tron Model S-200 screw feeder and a pneumatic
2-9
-------
2.3.2 Short Retention Time Carbon Slurry Feed System
The second carbon feed system used during testing on Unit B involved addition of
carbon to the lime slurry in a mixing tank installed just prior to the spray dryer. This
mixing system was designed to provide a relatively short contact time between the carbon
and lime slurry prior to injection of slurry into the SD through the existing slurry
atomization nozzles. Estimated carbon retention time in the slurry with this system was
about 10 minutes.
The system consisted of a 200 gallon polyethylene holding tank equipped with a
mixer and pump, and a K-tron Model S-200 volumetric screw feeder. A schematic of
this system is shown in Figure 2-3. Lime slurry was supplied to the small tank from the
existing slurry system. Flow to the tank was controlled using manually operated parallel
valves. Carbon was added to the tank using the volumetric screw feeder. An electrically
operated tank mixer was used to maintain a homogenous mixture of carbon and slurry.
A diaphragm pump was used to pump the carbon and slurry mixture from the tank to
the reactor via the existing slurry control valve and slurry flow meter. The pump was set
to deliver a constant flow rate.
Carbon feed rates were determined in the same manner as described previously in
Section 2.3.2.
2.3.3 Slaking Tank Carbon Slurry Feed System
Carbon was fed to Unit A as a carbon/lime slurry by mixing carbon with lime and
water in the lime slurry feed tank during each slaking cycle. The amount of carbon
added during each slaking cycle was designed to maintain a constant carbon content in
the slurry. The target slurry feed rate during all testing was 9 gallons per minute.
During two of the runs, flows were adjusted to correct for "abnormal" stack SO2 levels;
however, all runs.averaged between 8.2 and 9.6 gpm, and the condition averages were
2-12
-------
DC Motor with
Variable Speed
Controller
AC-DC
Converter
110V AC
Power Supply
K-Tron
S200 Feeder
with Screw
Control
From
Economizer
SOpsig
Plant Air
(25cfm)
Air Pressure
Regulator
1"Dia. Fox
Valve Eductor
Injection
Probe
To
Cyclone
Figure 2-2. Dry Carbon Injection System
-------
between 8.8 and 9.2 gpm. The amount of carbon added during each slaking cycle was
recorded so that the average carbon injection rate could be confirmed.
The carbon slurry mixture was injected into the reactor through the existing slurry
feed and atomization system. Existing slurry storage, slurry mixing, and slurry transport
equipment was used; therefore, no additional equipment was required. The carbon
retention time in the slurry for any single test condition is estimated to range from 3 to
8 hours, with an average of approximately 5 hours. Three hours represents the minimum
slurry volume maintained in the slurry feed tank prior to addition of fresh lime and
carbon at the start of a slaking cycle. Eight hours represents the maximum residence
time before adding fresh lime and carbon.
2.4 Description of Tested Carbons
Two different carbons were used to investigate whether carbon type was critical to
Hg removal by an SD/ESP-equipped MWC. Information on activation method, surface
area, pore radius, grind, and tamped density is summarized for both carbons in
Table 2-3. The first carbon used in the testing (Darco PC-100) was a thermally
activated, bituminous coal-based carbon with medium surface area and high tamped
density. The second carbon (Darco FGD) was thermally activated from lignite and had
a lower surface area, smaller average particle size, and lower tamped density than the
coal-based carbon.
2.5 Sampling Locations
2.5.1 Economizer Outlet Flue Gas Sample Location
A general schematic of the economizer outlet flue gas sampling location is shown
in Figure 2-4. The flue gas exits the economizer through a circular duct with an ID of
76 inches. Two.pairs of flue gas sample ports are located on this duct. One pair is
located approximately 156 inches (2.05 equivalent diameters) from the nearest upstream
2-14
-------
To SD Reactor
U)
K-Tron S200
System
(See Figure 2-2)
Existing Pump
Pipe ^
Existing
Pump
Figure 2-3. Short-Duration Slurried Carbon Injection System
-------
To
Cyclone
156"
120"
-d
Extractive
Probe
76"
e-f
a i
14.5"
Flow
a= 12" Unit A
22" Unit B
Grate Level
Concrete Floor
Figure 2-4. Economizer Outlet Flue Gas Sample Location
2-16
-------
TABLE 2-3 DESCRIPTION OF ACTIVATED CARBONS TESTED
CAMDEN COUNTY MWC
Source
Material
Coal
Lignite
Brand Name
Darco PC-
100
Darco FGD
Activation
Method
Thermal
Thermal
Surface
Area
(m2/g)
950
600
Average
Pore Radius
(1(T9 m)
1.5
3.0
Grind
% Thru 200
Mesh
97.1
99.9
% Thru 325
Mesh
72.8
98.2
Tamped
Density
(kg/m*)
690
470
to
-------
150'
2C
)0'
ฃ
><:
\v^
Expanded To
Flue inner diame
"N jr
) s'i
l/Xi
/
ran /
Figure 2-5. Stack Flue Gas Sample Location
2-18
-------
disturbance and approximately 120 inches (1.6 equivalent duct diameters) from the
nearest downstream disturbance. The other pair of ports (which were added for this
program) are located 6 inches to the right and 12 inches above the other pair of ports
for Unit A, and 6 inches to the right and 22 inches above the other pair for Unit B. All
of the ports at this location were 4 inches in diameter. A 24-point sampling matrix was
used for this location for both units. Sufficient work space was available at this location,
therefore, no additional preparations were required to traverse two trains simultaneously
with each pair of ports.
In addition to the two pairs of sampling ports described above, an additional port
located approximately 4 feet above the grate with a 4-inch pipe nipple was available for
non-traversing tests. This port was used for fly ash sampling using an EPA Method 5
type train.
2.5.2 Stack Flue Gas Sample Location
After the flue gas exits the ESP, it passes through an induced draft fan located at
the base of the stack. The gas enters the flue and is emitted into the atmosphere
approximately 366 feet from ground level. The test platform is located approximately
200 ft from ground level as shown in Figure 2-5.
The flues for Units A and B, along with Unit C and an additional flue for possible
plant expansion, are located in the same stack shell. A stack sampling grate is located
within the stack shell to provide access to the flues. Each of the flues has a 72-inch ID.
The flues for Units A and B have one pair of ports with 6-inch diameter flanged nipples,
located approximately 75 inches from the grate level. These ports were used for metals
sampling on Unit A and for metals and CDD/CDF sampling on Unit B. One additional
port, with a 4-inch diameter flanged nipple, is located approximately 52 inches above the
grate level. This port was used for PSD sampling on Unit A and for VOC sampling on
UnitB.
2-17
-------
TABLE 2-4 SAMPLING MATRIX
CAMDEN COUNTY MWC (1992)
Parameters
Mercury and Other Metals
Paniculate Matter
CDD/CDF
VOC
Particle Size Distribution
Fly Ash Carbon
Carbon in Fly Ash
SO2, O2 (Inlet)
SO2, O2, CO2, HC1, NO,, CO, THC, H2O,
Opacity (Stack)
Steam Flow and Furnace Temperature
Economizer Outlet Temperature
Lime Slurry and Dilution Water Flow
SD Outlet. Temperature
ESP Voltage, Amperage, Spark Rate
Method
EPA Multi-Metals Method
EPA Method 23
SW 846 Method 0030 (VOST)
Instack Cascade Impactor
EPA Method 5
ASTM D 3178-84
Plant CEMs QA
(40 CFR, Part 60, Appendix F)
Plant Process Monitors
2-20
-------
2.6 Sampling and Analytical Methods
Sampling methods used during the emission tests are listed in Table 2-4.
Summary descriptions of the sampling methods and corresponding analytical methods are
provided in Section 6. These sampling and analytical methods are also contained in
EPA or American Society of Testing and Materials (ASTM) reference documents. The
method used for Hg, other metals, and PM (EPA multi-metals method) is documented in
the Environmental Protection portion of the Code of Federal Regulations (40 CFR),
Part 266, Appendix IX.9 The methods used for PM carbon samples (EPA Method 5)
and CDD/CDF measurements (EPA Method 23) are contained in 40 CFR, Part 60,
Appendix A.10'11 The method used for VOST sampling and analysis (SW 846
Method 0030) is documented in EPA's Test Methods for Evaluating Solid Waste.12 The
method used for determining the amount of carbon in fly ash is contained in the
1984 Annual Book of ASTM Standards, Part 26.13 The plant Continuous Emission
Monitors (CEMs) were operated in accordance with the quality assurance requirements
of 40 CFR, Part 60, Appendix F.14
Table 2-5 shows the sampling times, minimum sample volumes, and detection
limits of the methods. Detailed descriptions of the sampling methods and the
corresponding analyses are provided in Section 6.
2-19
-------
TABLE 2-5 SAMPLING TIMES, MINIMUM SAMPLING VOLUMES,
AND DETECTION LIMITS
CAMDEN COUNTY MWC (1992)
Sampling
Train
PM/Metals
CDD/CDF
VOST
Sampling
Time
(hours)'
lc
2
1
Minimum
Sample
Volume
(dscf)
30
90
20 liters
per pair of
tubes
Analyte
PM
Hg
Cd
Pb
CDD/CDF
Volatile
Organics
Detection Limit
Flue Gasb
0.003 gr/dscf
0.05 /ig/dscm
0.4 ^g/dscm
1.2 jUg/dscm
0.03 ng/dscm
0.025-0.5 ^g/dscm
Analytical
10-50 mg
0.0002 ^g/ml
0.001 ^g/mld
0.003 /*g/mld
0.05 ng
1-20 ng per
pair of tubes
a An average sampling rate of 0.5 ft3/min was used to calculate sampling time.
b Flue gas detection limit is calculated conservatively by summing the front-half and back-half
detection limits. Solution volume for front-half and back-half fractions are typically 300 ml
and 150 ml, respectively.
0 During times when CDD/CDF sampling was also conducted, run time was two hours.
d Based on graphite furnace atomic absorption spectroscopy (GFAA).
e Detection limits for penta, hexa, and hepta isomers are approximately 5 times the above
value and the detection limits for the octa isomers are approximately 10 times the above
value.
2-21
-------
TABLE 3-1. SUMMARY OF TEST CONDITIONS AND MERCURY TEST RESULTS
CAMDEN COUNTY MWC (1992)
Phase-
Condition
I-B1
I-B2
I-B3
I-B4
I-B5
II-B6
Run
Number
1
2
3
Avg
4
5
6
Avg
7
8
9
Avg
10
11
12
Avg
13
14
15
Avg
10
11
12
Avg
Carbon
Type
None
FGD
PC-100
PC-100
FGD
None
Injection
Method
~
Dry
Dry
Dry
Dry
Carbon
Injection
Rate
(mg/dscm
@7%0j)
0
0
0
0
73
79
78
77
89
73
88
83
477
456
418
450
430
444
450
441
0
0
0
0
ESP Inlet
Temp
(ฐF)
269
267
262
266
274
266
275
272
264
272
273
270
265
290
291
282
275
277
262
271
348
350
349
349
Total Carbon
at Cyclone
Inlet
(mg/dscm
@ 7% O^
79a
79a
79a
79"
154a
160a
1593
158a
154a
138a
153a
148a
579a
558a
520a
552a
534a
548a
554a
546a
74
101
55
77
Mercury
Cone
at Inlet
(fig/dscm
@ 7% Oj)
356
1363
711
810
972
593
835
800
593
639
586
606
491
440
512
481
680
820
644
715
365
249
349
321
Mercury
Cone
at Outlet
(/ig/dson
@ 7% O,)
175
210
54
146
296
63
149
169
134
29
102
88
21
14
17
17
9
13
12
12
301
177
261
246
Removal
Efficiency
(%)
50.8
84.6
92.4
75.9
69.5
89.4
82.2
80.4
77.4
95.5
82.6
85.2
95.7
96.8
96.6
96.4
98.6
98.4
98.2
98.4
17.5
28.9
25.2
23.9
-------
3.0 INTERPRETATION OF RESULTS
This section summarizes the flue gas and process data collected during the
Camden County MWC testing (Section 3.1) and discusses the relationship between
carbon injection and emissions of Hg (Section 3.2), other metals (Section 3.3),
CDD/CDF (Section 3.4), VOC (Section 3.5), and acid gases (Section 3.6), and the
impact of carbon injection on ESP performance (Section 3.7). Where appropriate, the
test results from Camden County are compared to the results from SD/FF testing at
OMSS.
3.1 Data Summary
A summary of key operating data is presented in Table 3-1. The table includes
carbon type, injection rate and method, ESP temperature, total carbon concentration at
the cyclone inlet, Hg inlet and outlet concentrations, and Hg removal efficiency.
Additional process and emissions data are presented in Sections 4 and 5.
3.2 Mercury
As described in Section 2.2, activated carbons produced from lignite and
bituminous coals were injected at different rates into the flue gas as a dry powder and
with the lime slurry feed to the SD. To evaluate the effectiveness of carbon injection
and SD/ESP operating conditions on emissions, the following operating parameters were
studied: carbon type, injection rate and method, inherent carbon content of the
combustor fly ash, carbon retention time in the SD slurry, ESP temperature, and PM
emission rate.
3-1
-------
TABLE 3-1, CONTINUED
Phase-
Condition
II-B13
II-A1
II-A2
II-A3
II-A4
II-A5
Run
Number
37
38
39
Avg
1
2
3
Avg
4
5
6
Avg
7
8
9
Avg
22
23
Avg
31
32
33
Avg
Carbon
Type
FGD
None
FGD
FGD
FGD
FGD
Injection
Method
Slurry
_
Slurry
Slurry
Slurry
Slurry
Carbon
Injection
Rate
(mg/dscm
@7% Oj)
183
194
200
192
0
0
0
0
344
346
343
344
402
356
386
381
442
391
417
269
280
249
266
ESP Inlet
Temp
OF)
266
263
264
264
277
270
273
273
265
265
266
265
278
269
288
278
285
283
284
283
283
284
283
Total Carbon
at Cyclone
Inlet
(mg/dscm
@ 7% Oj)
233
265
248
249
100
86
198
128
427
468
450
448
579
412
629
540
640
567
604
381
404
356
380
Mercury
Cone
at Inlet
(/ig/dscm
@ 7% OT)
382
377
974
578
268
430
610
436
302
403
1412
706
530
458
690
559
643
816
730
335
294
364
331
Mercury
Cone
at Outlet
(/ig/dscm
@ 7% O,)
78
81
158
106
121
290
322
244
55
78
261
131
43
108
156
102
49
90
70
40
51
52
48
Removal
Efficiency
(%)
79.7
78.5
83.8
80.7
54.9
32.6
47.2
44.9
81.9
80.7
81.5
81.4
91.9
76.4
77.4
81.9
92.3
89.0
90.7
88.0
82.6
85.6
85.4
Inlet PM concentration was not measured during Phase I; inherent carbon concentrations lor Phase 1 estimated based on the average inlet PM
measured during Phase II and the measured fly ash percent carbon for the test condition.
-------
TABLE 3-1, CONTINUED
Phase-
Condition
II-B7
II-B8
II-B9
II-B10
II-B11
II-B12
Run
Number
13
14
15
Avg
16
17
18
Avg
19
20
21
Avg
25
26
27
Avg
28
29
30
Avg
34
35
36
Avg
Carbon
Type
FGD
FGD
FGD
None
FGD
FGD
Injection
Method
Dry
Dry
Dry
-
Dry
Slurry
Carbon
Injection
Rate
(mg/dscm
@7%O2>
313
329
324
322
173
149
190
171
30
46
43
40
0
0
0
0
357
342
387
362
324
325
336
328
ESP Inlet
Temp
(ฐF)
352
352
344
349
267
263
262
264
266
266
265
266
269
266
258
264
271
273
269
271
26<>
269
275
271
Total Carbon
at Cyclone
Inlet
(mg/dscm
@ 7% 03)
387
429
418
411
305
276
306
295
111
141
129
127
83
98
91
91
506
504
505
505
385
368
403
385
Mercury
Cone
at Inlet
(jig/dsan
@ 7% O^)
964
506
778
749
545
455
525
508
485
957
463
635
663
433
384
493
626
635
664
642
299
521
300
373
Mercury
Cone
at Outlet
(/ig/dson
@ 7% O^
107
22
59
63
40
23
24
29
103
170
124
132
388
279
207
291
20
16
16
17
50
77
69
65
Removal
Efficiency
(%)
88.9
95.6
92.4
92.3
92.7
95.0
95.4
94.4
78.8
82.2
73.2
78.1
41.5
35.6
46.1
41.0
96.8
97.4
97.7
97.3
83.2
85.3
77.0
81.8
-------
g
a
.2
i
3
c*
ฃป
1
1UU
90
80
70
60
50
40
30
20
10
n
*
.*..
.
i i i i i
100 200 300 400 500
Injected Carbon Concentration (mg/dscm)
Lignite (FGD) + Coal(PC-lOO)
Figure 3-1. Effect of Carbon Type on Mercury Reduction
(Dry Injection and 270ฐ F ESP Inlet Temperature)
100
90
80
gT 70
<^
| 60
1
"S 50
tf
40
30
20
10
0
100 200 300 400
Injected Carbon Concentration (mg/dscm)
500
Figure 3-2. Effect of Injected Carbon Concentration on Mercury Reduction
(FGD Carbon, Dry Injection and 270ฐ F ESP Inlet Temperature)
3-6
-------
3.2.1 Impact of Carbon Type
The influence of carbon type was examined during testing with lignite-based
carbon (Conditions B2 and B5) and coal-based carbon (Conditions B3 and B4).
Conditions B2 and B3 were conducted at a low carbon feed rate of approximately
80 mg/dscm. Conditions B4 and B5 were conducted at a high carbon feed rate of
approximately 450 mg/dscm. The carbon was injected as a dry powder during each
condition.
Figure 3-1 shows the calculated Hg removal efficiency during each test run. At
the low carbon feed rate, the calculated removal efficiency was 70 to 89% with the
lignite-based carbon and 78 to 96% with the coal-based carbon. At high carbon feed
rates, the removal efficiency was 98 to 99% with the, lignite-based carbon and 95 to 97%
with the coal-based carbon. Because there was no clear distinction in the removal
efficiency of these two carbons, the remaining tests were conducted using the more
economical lignite-based carbon. The similarity in performance of these two carbons
when injected as a dry powder is consistent with the results of the OMSS testing.4
3.2.2 Impact of Carbon Injection Rate
Figure 3-2 graphs Hg removal efficiency as a function of carbon injection rate.
The data points show the removal efficiencies measured during individual runs
conducted at an ESP temperature of approximately 270ฐF and while injecting either dry,
lignite-based carbon, or no carbon. The specific test conditions are Bl, B2, B5, B8, B9,
BIO, and Bll. The carbon injection rates range from 40 to 450 mg/dscm.
As evident from the figure, increasing carbon injection increases the Hg reduction
and decreases the variability of Hg reduction between individual runs of the same
condition. These tendencies were also observed during the OMSS test program.
However, the carbon feed rates at Camden County were significantly higher than at
OMSS, where the highest feed rate with dry carbon was approximately 70 mg/dscm.4
3-5
-------
450
6s 40ฐ
1, 350
| 300
250
5 20ฐ
15ฐ
50
100 200 300 400
Injected Carbon Concentration (mg/dscm)
500
Figure 3-3. Effect of Injected Carbon Concentration on Mercury Emissions
(FGD Carbon, Dry Injection and 270ฐ F ESP Inlet Temperature)
composite PM sample was collected at the economizer exit during each test condition
and analyzed for carbon content. The carbon levels measured during each day were
between 1.1 and 2.2% of the dried sample weight. The percent carbon found in each
daily sample was then multiplied by the measured PM loading at the economizer exit for
each run on that day. The resulting estimate of inherent carbon in mg/dscm was then
added to the rate at which activated carbon was injected to estimate the total carbon
level in the flue gas. One shortcoming of this approach is that only a single estimate of
the PM carbon content is obtained for each day and any run-to-run variations in
combustion conditions that could result in increased carbon levels during an individual
run are not measured. Also, some large fly ash carbon is removed by the spray dryer
inlet cyclone and this carbon loss is unaccounted for.
As shown in Figure 3-4, the 40 to 70% reduction in emitted Hg in the absence of
carbon injection could be explained by the presence of approximately 100 mg/dscm of
unburned carbon associated with the emitted PM. At OMSS, the carbon content of the
3-8
-------
For dry carbon injection rates above 150 mg/dscm, Hg removals were 93% or
greater, and exhibited relatively small increases in Hg reduction. At these feed rates, the
variability between runs of a given test condition was 3% or less. At carbon feed rates
of less than 150 mg/dscm, the Hg removal efficiencies were noticeably lower and the
run-to-run variability between individual runs was as much as 20% during a single test
condition.
The greatest variability in Hg reduction was observed during Conditions Bl and
BIO, with no carbon injection. In particular, during Runs 2 and 3 of Condition Bl,
removal efficiencies were 85 and 93%, which are nearly double the value of other runs
with no carbon injection. It was initially believed that these high values may reflect poor
combustion conditions caused by high moisture content in the waste stream. However,
similar "wet waste" were experienced during the Phase II tests, and no abnormally high
Hg captures were observed. Review of the three previous quarterly Hg emission tests of
Unit B shows reductions during three-run tests of 41 to 43%, 41 to 55%, and 30 to 73%
(all based on EPA Method 101A). These data suggest that Hg removals without carbon
injection for the tested unit is typically between 40 and 50%, but can be both higher or
lower.
The effect of carbon injection rate on stack Hg concentrations is shown in
Figure 3-3. The trends in these data are similar to the Hg reduction data. Specifically,
at carbon injection rates above 150 mg/dscm, stack concentrations show relatively little
run-to-run variation, while at lower carbon injection rates, there is significant variability
in the nm-to-run data.
3.2.3 Impact of Inherent Carbon
Part of the variability in Hg reductions during the EPA-funded and previous tests
may result from differences in the amount of unburned carbon in the PM emitted from
the combustor. To estimate the amount of unburned carbon present in the flue gas, a
3-7
-------
i* 40ฐ
1) 35ฐ
I 300
ed
a 250
8
5 20ฐ
ง 150
% 100
50
0
600
0 100 200 300 400 500
Total Carbon Concentration (mg/dscm)
Figure 3-5. Effect of Total Carbon Concentration on Mercury Emission
(FGD Carbon and 270ฐ F ESP Inlet Temperature)
a
.2
t5
100
80
70
60
g 50
tf
& 40
| 30
20
10
0
0 100 200 300 400 500
Injected Carbon Concentration (mg/dscm)
Dry@270ฐF + Wet@270ฐF
Figure 3-6. Effect of Carbon Feed Method on Mercury Reduction
(FGD Carbon and 270ฐ F ESP Inlet Temperature)
3-10
-------
90
80
70
o 60
1
"2 50
40
30
20
10
600
0 100 200 300 400 500
Total Carbon Concentration (mg/dscm)
Figure 3-4. Effect of Total Carbon Concentration on Mercury Reduction
(FGD Carbon, Dry Injection and 270ฐ F ESP Inlet Temperature)
PM emitted from the combustor (0.5 to 1.0%) was approximately one-half the level at
Camden County and the Hg reduction without carbon injection was also approximately
one-half the level (25%).4
Figure 3-5 is a plot of the stack Hg concentration versus total carbon
concentration.
3.2.4 Impact of Carbon Injection Method
Figure 3-6 shows the relationship between carbon injection method and Hg
removal. At medium carbon injection rates (150 to 200 mg/dscm), removal efficiencies
were 92 to 95% with dry injection (Condition B8) and 79 to 84% when the carbon was
injected as a slurry (Condition B13). At high carbon injection rates (320 to
390 mg/dscm), removal efficiencies were 97 to 98% with dry injection (Condition Bll)
3-9
-------
i*
1
e
I
a
I
6
i
8
s
zuu
180
160
140
120
100
80
60
40
20
n
+
+ * +
...
a -
" a" a
i i i i i
100
200
300
400
500
Injected Carbon Concentration (mg/dscm)
Dryฎ 270ฐ F + Wet @ 270ฐ F
Figure 3-7. Effect of Carbon Feed Method on Mercury Emissions
(FGD Carbon and 270ฐ F ESP Inlet Temperature)
Conditions A2, A3, and B12 were used to assess the impact of carbon retention
time in the slurry. The average carbon feed concentration for these conditions was 320
to 400 mg/dscm. As shown in Figure 3-8, the Hg removal efficiency for the testing on
Units A and B were very similar with both units averaging 82%. As a result, it appears
the decreased Hg adsorbance of the Darco FGD carbon when mixed with slurry occurs
rapidly and does not change with slurry retention time in excess of the minimum times
tested at Camden.
3.2.6 Impact of ESP Temperature
Figure 3-9 shows the relationship between ESP temperature and Hg removal
efficiency when operating without carbon injection and at high carbon injection
concentrations. As discussed in Section 3.2.2, when operating without carbon injection
and an ESP temperature near 270ฐF (Conditions Bl and BIO), Hg removals averaged
roughly 50%. At the higher ESP temperature of 350ฐF (Condition B6), the Hg removal
3-12
-------
and 77 to 85% with slurried carbon (Condition B12). Figure 3-7 shows the relationship
between carbon feed method and Hg stack concentrations. These data suggest that the
feed method does affect Hg removal efficiency and emissions.
This observation is in contrast to the OMSS results, which found that feed method
did not have a significant impact on Hg emissions and Hg removal. The cause of this
difference is uncertain, but may be due to the different carbon type used or the type of
PM control device. The carbon used for the OMSS SD/FF slurry testing was a coal-
based carbon, rather than the lignite-based carbon used during slurry testing at Camden
County.4
As discussed in Section 2.4, the lignite-based carbon is characterized as having
larger average pore diameters than the coal-based carbon. Lignites are also generally
more hygroscopic than bituminous coals. Both of these factors may contribute to greater
wetting or plugging of the carbon surface, and thus reduced reactivity.
The difference in PM control device may also be significant. For an ESP, as used
at Camden County, most of the Hg adsorption occurs while carbon is suspended in the
flue gas (a residence time of 10 to 20 seconds). For a FF, as used at OMSS, the carbon
has additional time to dry and adsorb Hg while it is held in the filter cake.
3.2.5 Impact of Carbon Retention Time in Lime Slurry
The carbon retention time in the lime slurry was different for Unit A and Unit B.
On Unit A, carbon was added to the lime slaking tank approximately once every five
hours. Carbon retention time in the slurry is estimated to be three to eight hours. On
Unit B, carbon was added to the slurry in a mixing tank installed just prior to the SD.
Retention time of the carbon hi this system is estimated at 8 to 10 minutes.
3-11
-------
was 18 to 29%. At high carbon feed rates and an ESP inlet temperature of 270ฐF
(Condition Bll), Hg removals were 97 to 98%. At similar carbon feed rates, but an ESP
temperature of 350ฐF (Condition B7), Hg removals were 89 to 96%. These data suggest
that the ability of carbon to absorb Hg is directly related to flue gas temperature, but
that even at relatively high temperatures of 350ฐF, activated carbon injection can achieve
significant Hg reductions.
3.2.7 Impact of PM Control Efficiency
Particulate matter removal efficiencies during the Camden County testing were
greater than 99.9% for all but five runs. These five runs occurred during
Conditions BIO, B13, A3 and A5. Condition BIO was conducted without carbon
injection. Conditions B13, A3 and A5 were conducted while injecting carbon as a slurry.
As can be seen on Figure 3-10, there is no apparent relationship between PM and Hg
removal efficiency during these four test conditions.
3.2.8 Multivariate Regression Analysis
A stepwise multivariate regression analysis was used to assess the statistical
significance of individual process variables and to develop predictive equations of Hg
removal efficiency and outlet concentration. The process variables examined in this
analysis included injected carbon concentration, total carbon concentration (i.e., injected
carbon plus unburned carbon in the fly ash), carbon injection method (dry or slurry), SD
outlet temperature, inlet Hg concentration, and PM control efficiency. In a stepwise
multivariate regression analysis, the model first identifies the single independent (i.e.,
process) variable that is the strongest predictor of the dependent variable (outlet Hg
concentration or removal efficiency). If the independent variable is statistically
significant (the 95% confidence level based on the t-statistic was used in the analysis),
the model then identifies the next most significant variable, which when combined with
the first variable best predicts the dependent variable. A t-statistic based confidence
level is used for statistical analysis of small populations (less than 30 test data points).
3-14
-------
1UU
90
80
ง 70
| *ฐ
I 50
a 40
30
20
10
n
-
I
1 ^
A
1
- 2
S 8 IT
1 I 1
ซ c
2 ^
5 r ซ
ป
-
-
3 |
Short Residence Time (approximately 8-10 minutes)
Long Residence Time (approximately 3-8 hours)
Figure 3-8. Effect of Carbon Retention Time in Lime Slurry on Mercury Reduction
(FGD Carbon, Dry Injection and 270" F ESP Inlet Temperature)
*T
1
ซ
i
i
1
1UU
90
on
70
60
50
40
30
10
0
*
- - - ----- -
; -
--"
~~l~ ~" - " - -
i 1 1 1 i i
100 200 300 400
Total Carbon Concentration (mg/dscm)
270CF ESP Inlet * 350ฐF ESP Inlet
500
Figure 3-9. Effect of ESP Temperature on Mercury Reduction
3-13
-------
the removal of Hg when carbon was not injected due to the presence of inherent carbon
in the fly ash and the difference in carbon utilization rates observed for dry versus slurry
injection. Based on review of the residual error estimates from the initial regression
analysis, Run 3 of Condition I-B1 was determined to be a statistical outlier and was
excluded from use in the final regression analyses.
Removal Efficiency
The final regression analysis identified three statistically significant process
variables influencing Hg control efficiency: carbon feed rate, SD outlet temperature, and
carbon injection method. The best predictive model for Hg percent reduction based on
the dry carbon injection data was:
ln(100-%RED) = 9.76 - 0.145 (CFC)0-5 - 2390/1) (Equation 3-1)
where %RED is % reduction in Hg, CFC is Carbon Feed Concentration in mg/dscm,
and T is temperature in Kelvin. The "goodness of fit" (R2) of this model is 0.83. The
predictive equation for slurry injection of carbon was based on the two Unit B test
conditions using slurry injection and the removal efficiency of 52% derived from
Equation 3-1 at zero carbon injection and a temperature of 270ฐF. This equation is:
ln(100 - %RED) = 9.76 - 0.0578 (CFC)0-5 - 2390/1) (Equation 3-2)
The calculated Hg removal efficiencies derived from Equation 3-1 at 270ฐF and
350ฐF and from Equation 3-2 at 270ฐF are shown in Figure 3-11, along with the actual
data. The carbon injection concentration required to achieve an average reduction in Hg
concentration of 90% during the Camden testing, based on injection of dry carbon at
270ฐF, is approximately 115 mg/dscm. Due to variations in process operation, however,
the Hg reduction achieved during an individual test at a given carbon feed concentration
varies. As shown in Figure 3-12, for dry injection at 270ฐF, 90% (i.e., the span between
the 5% and 95% confidence interval lines) of the Hg reduction data at a carbon feed
3-16
-------
100
80
60
3 40
1
20
99.5
99.6 99.7 99.8 99.9
Paniculate Matter Reduction (%)
100
A3
A5
A
*
BIO
B13
Figure 3-10. Effect of Participate Matter Control on Mercury Reduction
(FGD Carbon, Wet Injection and 270ฐ F ESP Inlet Temperature)
This variable is then tested for statistical significance and this "stepwise" process
continues until no other independent variables are found to statistically improve the
model.
In this analysis, the percent Hg reduction values were converted to emissivity
values (100 minus percent reduction). Because emissivity and emissions data are
generally lognormally distributed, the natural log of the emissivity and outlet Hg levels
were used as the dependant variables. To account for decreasing carbon utilization as
the carbon feed rate increases, the square root of the carbon feed rate was used to
linearize these data.
To estimate a mathematical model for predicting Hg control efficiency and outlet
concentrations, the data set was divided into two subsets - one consisting of the data
from testing with no carbon and dry carbon injection, and the second consisting of the
data with no carbon and slurry carbon injection. These data subsets were used to reflect
3-15
-------
concentration of 115 mg/dscm are projected to be between approximately 80% and 95%.
Alternatively, at a carbon feed concentration of 250 mg/dscm, there is still a 5%
probability that the Hg reduction during an individual test will be less than 90%.
To account for the variability in the Hg reductions without carbon injection (as
well as differences in the unburned carbon content of individual MWCs), the intercept
constant in Equation 3-1 was adjusted to reflect a baseline (i.e., no carbon injection) Hg
reduction of 30%. This reduction reflects the lower end of the test data at Camden
County, and is consistent with the average Hg reduction at OMSS and several other mass
burn MWCs equipped with SD/FF and SD/ESP systems that do not inject carbon. As
shown by Figure 3-11, the predicted carbon feed rate needed to achieve an average
reduction of 90% with dry injection at 270ฐF is approximately 180 mg/dscm. This
injection rate is roughly three times the predicted rate needed to achieve 90% Hg
reduction using the SD/FF data from the OMSS testing.4 Note also that 90% removal
of Hg is predicted for dry carbon at 350ฐF at a carbon feed concentration of
approximately 230 mg/dscm, and that injection of slurried carbon is noticeably less
effective, resulting in predicted average reductions of approximately 80% at injected
carbon concentrations of 230 mg/dscm.
The absence of inlet Hg concentration as a statistically significant variable for
predicting Hg removal efficiency is in contrast to the OMSS data and is believed to
reflect the difference in control capability of systems equipped with a FF versus an ESP.
With a FF, carbon will adsorb Hg both while entrained in the flue gas and after it is
collected in the filter cake. When inlet Hg levels vary (e.g., due to a short-duration spike
in Hg concentration), the carbon on the filter cake is able to limit the impact of the
spike at the outlet. In this situation, the efficiency of the control system (i.e., entrained
carbon and filter cake carbon) increases when the inlet Hg level increases. The ability
of the filter cake to buffer spikes in inlet Hg levels is similar to the ability of the filter
cake to moderate fluctuations in inlet acid gas levels. With an ESP, most of the Hg
reduction occurs while the carbon is entrained in the flue gas and is controlled by the
likelihood of contact between carbon particles and Hg prior to the collection of carbon
3-18
-------
.2 60
1
"8 50
B*
1 40
S 30
20
10
0
t I /
f
-B- Dry @ 270ฐ F
-+--Dry@350ฐF
-A- Wet @ 270ฐ F
Dry @ 270ฐ F w/ 30% Baseline Control
500
100 200 300 400
Injected Carbon Concentration (mg/dscm)
Figure 3-11. Regression Analysis Results for Mercury Reduction
95% Confidence Limit
Mean Reduction
5% Confidence Limit
I 100 200 300 400
Injected Carbon Concentration (mg/dscm)
Figure 3-12. Predicted Mercury Reduction Variability with Dry Injection at 270ฐ F
3-17
-------
Inlet Mercury Concentration
1100/ig/dscm
800jig/dscm
500/ig/dscm
200/ig/dscm
100 200 300 400
Injected Carbon Concentration (mg/dscm)
Figure 3-13. Regression Analysis Results for Mercury Emissions
(FGD Carbon, Dry Injection and 270ฐ F ESP Inlet Temperature)
A summary of the metals removal efficiencies for each test condition is shown in
Table 3-2. For Cd, Pb, As, Ba, and Cu, metals removal efficiencies exceeded 99%
during all test conditions. For Cr and Mn, removal efficiencies exceeded 99% except
during the high temperature run (B7) and for Mn during the medium feed rate carbon
slurry test condition (B13). For Mo and Ni, removal efficiencies showed significant
variability, ranging from a low of 72% for Mo during the high temperature test condition
up to 98%. Removal efficiencies for Sb, Be, Co, and V could not be precisely
determined due to concentrations at the ESP outlet that were below the analytical
detection limit. The values shown for these four metals in Table 3-2 were estimated
based on the analytical detection limit for each metal and a typical flue gas flow rate.
Removal efficiencies for Ag and Tl could not be estimated because concentrations of
these metals were below the analytical detection limit at both the inlet and outlet
sampling location. Poor matrix spike recoveries were experienced for Se; therefore, Se
data are not presented in Table 3-2.
3-20
-------
on the ESP plates. Once the carbon particle is collected on an ESP plate, the potential
for contact with Hg is greatly reduced. As a result, the control efficiency of this system
(i.e., entrained carbon only) is independent of the inlet Hg level.
Outlet Concentration
The stepwise regression analysis identified four statistically significant process
variables influencing outlet Hg concentration: carbon feed rate, SD outlet temperature,
carbon injection method, and inlet Hg concentration. The best predictive model for this
model based on the dry injection data was:
ln(HgOut) = 9.67 - 0.136(CFQW + 0.00114(HgIn) - 1960{ฑ\ (Equation 3-3)
where HgOut and Hgln are the Hg Outlet and Inlet concentrations in mg/dscm and
CFC and T are as defined in Equation 3-1. The R2 of this model is 0.81.
Figure 3-13 shows the predicted outlet concentrations from this model based on
an ESP operating temperature of 270ฐF and inlet Hg concentrations of 200, 500, 800, and
1,100 ^g/dscm. Note that most of the reduction in outlet concentration occurs at carbon
injection rates of up to approximately 100 mg/dscm. At carbon injection rates above this
level, the reduction in outlet concentrations is much more gradual.
33 Other Metals
Flue gas concentrations of the 16 other metals listed in Section 2.2.2 were
evaluated during six test conditions. Five of these test conditions were conducted at
270ฐF: no carbon injection (BIO), dry carbon injection at a low and a high feed rate (B2
and Bll, respectively), and slurry injection of carbon at a medium and a high feed rate
(B13 and B12, respectively). The sixth test condition was conducted at 350ฐF with dry
carbon injection (B7).
3-19
-------
These data indicate that the 13 detected metals, with the possible exception of
Mo, are emitted primarily as participate and that control of emissions of these metals is
achieved predominantly by the PM control device. There also appears to be an affect of
ESP temperature on the control of Cr, Mn, and Ni, but given the small size of the data
set, this partial relationship may be due to random chance. Injection of activated carbon
did not have a quantifiable impact on emissions of any of the metals.
3.4 CDD/CDF
Economizer outlet and stack concentrations of CDD/CDF were measured during
Conditions BIO (no carbon injection), Bll (dry carbon at 270ฐF), and B12 (slurry carbon
at 270ฐF). Figure 3-14 shows the calculated CDD/CDF reduction for each of the three
runs at these conditions. During Condition BIO without carbon injection, the total
CDD/CDF removal efficiency across the SD/ESP was 78 to 80%. During
Condition Bll with a high injection rate (approximately 360 mg/dscm) of dry carbon, the
removal efficiency was 95 to 98%. During Condition B12 with a high injection rate of
slurried carbon, the removal efficiency was 96 to 97%. These data suggest that, unlike
Hg, the CDD/CDF collection efficiency of dry and slurried carbon injection is similar.
As shown in Figure 3-15, total CDD/CDF emission levels drop from 40 to
60 ng/dscm without carbon injection to less than 10 ng/dscm for dry carbon injection
and less than 15 ng/dscm for slurry injection. The higher CDD/CDF outlet levels during
slurry injection of carbon appears to reflect the higher concentration of CDD/CDF
measured at the economizer outlet during two of the Condition B12 runs of
approximately 375 ng/dscm compared with 130 to 220 ng/dscm for the other seven runs.
3-22
-------
TABLE 3-2. AVERAGE SD/ESP REMOVAL EFFICIENCY (%)
FOR SELECTED TEST CONDITIONS AT
CAMDEN COUNTY MWC (1992)'
Condition No.
Carbon Feed Rate
Carbon Feed Method
ESP Temperature (ฐF)
B2
Low
Dry
270
B7
High
Dry
350
BIO
None
None
270
Bll
High
Dry
270
B12
High
Slurry
270
B13
Med
Slurry
270
Removal Efficiency (%)
Total PM
Cadmium
Lead
Antimony
Arsenic
Barium
Beryllium
Chromium
Cobalt
Copper
Manganese
Molybdenum
Nickel
Vanadium
99.95"
99.5
99.7
>99.5C
99.9
99.8
>85C
99.0
>98C
99.2
99.1
82.7
93.7
>98.5C
99.98
99.8
99.9
>99.5C
99.9
99.8
>85C
98.4
>98C
99.8
97.8
72.1
96.1
>98.5C
99.90
99.6
99.6
>99.5C
99.8
99.8
>85C
99.5
>98C
99.7
99.2
87.6
98.5
>98.5C
99.96
99.8
99.8
>99.5C
99.9
99.8
>85C
99.7
>98C
99.8
99.6
91.0
97.4
>98.5C
99.96
99.9
99.9
>99.5C
99.9
99.6
>85C
99.6
>98C
99.9
99.3
84.5
98.3
>98.5C
99.82
99.9
99.9
>99.5C
99.8
99.7
>85C
99.3
98C
99.4
98.2
80.2
96.0
>98.5C
selenium results not presented due to poor matrix spike recoveries.
b Estimated. Inlet PM level not measured. Control efficiency based on
value equal to average of all measured runs.
c Outlet emission rate less than detection limit. Percent reduction based
limit.
assumed inlet
on detection
3-21
-------
3.5 Volatile Organic Compounds
Sampling for VOC was conducted during Conditions BIO (no carbon injection)
and Bll (dry carbon injection). Figure 3-16 shows the percent reduction across the
SD/ESP for the seven compounds found in most of the samples. Measured
concentrations of these compounds are contained in Section 4.9. As shown in
Figure 3-16, there appears to be a reduction in the level of carbon disulfide, benzene,
and chlorobenzene across the SD/ESP, and an increase (i.e., negative reduction) in
trichlorofluoromethane, methylene chloride, and toluene. Because of the low
concentrations of several of the detected compounds, the quantitative removal or
formation across the SD/ESP is uncertain. Of significance to this study, however, there
is no apparent impact of carbon injection on the behavior of any of these compounds.
Reduction (%)
ง 1 0 I
I1
B
i
.1; '
A
M if ฐ,
F i " !
F 1 i
I E
* 1
f
i I
C i i
TY.ซปlซ t~*n*4\nr* Coo/I Dnto
riign uaroon reea Kate
A = Trichlorofluoromethane E = Toluene
B = Carbon Disulfide F = Chlorobenzene
C = Methylene Chloride G = m,p-Xylene
D = Benzene
Figure 3-16. Effect of Carbon on VOC Reduction
3-24
-------
^
e
|
i
p
Q
p
S
8
100
98
96
94
92
90
88
86
84
82
80
78
76
74
72
in
f
| $
-
-
-
-
-
-
BIO - No Carbon
Bll - Dry Carbon
I B12 - Wet Carbon
-
-
-
i i i
BIO
Bll
Condition Number
B12
Figure 3-14. Effect of Carbon Injection on CDD/CDF Reduction
(FGD Carbon and 270ฐ F ESP Inlet Temperature)
70
60
I 50
***
|,0
a
g 30
a
u
20
10 -
BIO - No Carbon
Bll - Dry Carbon
B12-Wet Carbon
BIO
Bll
Condition Number
B12
Figure 3-15. Effect of Carbon Injection on CDD/CDF Concentration
(FGD Carbon and 270ฐ F ESP Inlet Temperature)
3-23
-------
-s
cT
v
1
8
V)
1UU
95
90
85
80
75
70
65
60
55
*n
A A
A
A A^ A * ^
A A
- ft
A
. """'
*
. v
-
.
- S
i i i i i i
500
0 100 200 300 400
Injected Carbon Concentration (mg/dscm)
Unit B, Dryฎ 270ฐ F * Unit A, Wet @ 270ฐ F
Figure 3-17. Effect of Carbon Injection on SO2 Reduction
0
s>
>
IL
g^*
w^
.2
ง
a
1
a
5
2
4OU
250
240
230
220
210
200
190
180
170
iฃn
A
A
A A A A
A A
A *
-
A
A
"
A
' *
0 g
A
i i i > i i
0 100 200 300 400 500
Injected Carbon Concentration (mg/dscm)
Unit B, Dryฎ 270ฐ F A Unit A, Wet @ 270ฐ F
Figure 3-18. Effect of Carbon Injection on NOX Concentration
3-26
-------
3.6 Acid Gases
During each test condition, emissions of SO2, HC1, and NOX were monitored using
the plant's continuous emission monitoring systems. Figures 3-17 and 3-18 are plots of
SO2 removal efficiency across the SD/ESP and of stack NOX concentrations, respectively.
The Unit B data are from test conditions using dry carbon injection and a target ESP
operating temperature of 270ฐF (Conditions Bl, B2, B5, B8, BIO, and Bll). The Unit A
data are based on slurry carbon injection and include the first two test conditions (Al
without carbon and A2 with carbon injection).
Based on the general increase in SO2 removal versus carbon injection rates shown
in Figure 3-17, it appears that carbon injection increases SO2 removal. However, the size
of the data set, the effects of SO2 inlet concentrations and the scatter in the data are
such that this apparent relationship may be due to random chance.
Based on the data shown in Figure 3-18, there is no apparent relationship
between carbon feed rate and NOX emissions. A review of HC1 data, although not
shown, also did not indicate any relationship with carbon feed rate.
3.7 Impact of Carbon Injection on ESP Performance
To evaluate whether carbon injection might detrimentally affect the emissions
control performance of the ESP, carbon was feed into Unit A continuously for 12 days.
Prior to and during carbon feeding, testing was conducted to assess emissions of Hg, Cd,
Pb, and PM, and to assess any changes in stack opacity levels, ESP operating
characteristics, and the size distribution of emitted paniculate. Test Condition Al was
conducted without carbon injection, A2 through A4 were conducted on the first, third,
and eighth days after the start of carbon injection. D.uring each of these four test
conditions, the ESP was operated with four ESP fields in service. Following completion
3-25
-------
of testing for Condition A4, the fourth ESP field was taken out of service. Condition A5
was conducted after the ESP had operated for approximately 80 hours with three fields
in service.
Figure 3-19 is a plot of ESP performance as indicated by average PM, Cd, and Pb
removal efficiencies and the percent of total PM less than 2 ^m during each test
condition. As shown in the figure, there was no consistent change in any of these
parameters during the first four test conditions, indicating that carbon injection did not
alter ESP performance. During Condition A5, with the fourth ESP field out of service,
there was not a noticeable change in PM removal efficiency. However, the removal
efficiency for Cd and Pb decreased, and the percent of emitted PM less than 2 /*m
increased. These changes are consistent with the expected enrichment of volatile metals
onto fine particulate and the reduced ability of the ESP to collect fine paniculate when
the fourth ESP field was taken out of service. Stack opacity, ESP voltage, and ESP
current did not vary significantly during the entire test period.
1UU.U
99.8
1 Efficiency (%)
% 5g
jป i>
ง 99.2
E
&
99.0
OB a
.*.-*.-
ซ:.-.;- *-ป ^iS*"^
- - 7;:r:ri^"""" "\..
-
Al A2 A3 A4
Condition
- - - PM Removal Efficiency
----- Cd Removal Efficiency
* Pb Removal Efficiency
' % Particulate Matter Less Than 2 /im
K-
AS
JV
40
30 $
i
A
20 to
I
10
Figure 3-19. Changes in ESP Performance Over Time
3-21
-------
TABLE 4-1 CARBON FEED SYSTEM DATA FOR UNIT B
CAMDEN COUNTY MWC (1992)
Phase-
Condition
I-B1
I-B2
I-B3
I-B4
I-B5
H-B6
H-B7
n-B8
H-B9
Date
5/11/92
5/12/92
5/13/92
5/14/92
5/15/92
6/2/92
6/3/92
6/4/92
6/5/92
Carbon Type
None
FGD
PC-100
PC-100
FGD
None
FGD
FGD
FGD
Carbon Feed
Method
~
Dry
Dry
Dry
Dry
-
Dry
Dry
Dry
Ron
1
2
3
Carbon Feed
Rate(Ib/hr)*
0
0
0
Average 0
4
5
6
12.1
12.1
12.1
Average 12.1
7
8
9
12.5
12.5
12.5
Average 12.5
10
11
12
61.4
61.4
61.4
Average 61.4
13
14
15
Average
10
11
12
Average
13
14
15R
Average
16
17
18
Average
19
20
21
60.0
60.0
60.0
60.0
0
0
0
0
47.6
51.0
513
50.0
25.6
25.9
25.8
25.8
4.9
6.6
6.7
Average 6.1
4-2
-------
4.0 CARBON INJECTION PARAMETRIC TESTING
Testing was conducted on Unit B to evaluate the impact of carbon injection
system and SD/ESP operating parameters on emission control performance. Variables
included carbon type, feed rate, and feed method, and ESP operating temperature. A
total of 13 test conditions were conducted, with each test condition consisting of three
runs conducted on the same day.
4.1 Carbon Feed System Data
Table 4-1 presents the data for the carbon feed systems used to feed carbon
during selected tests on Unit B. These data include type of carbon fed, the carbon feed
method (i.e., slurry or dry), and carbon feed rates. Carbon was injected for 10 of the
13 tests on Unit B. Of the 10 test conditions when carbon was injected, 8 injected dry
carbon into the flue gas ductwork upstream of the cyclone. The two remaining tests
were conducted with carbon injected into the spray dryer with the lime slurry. For these
two conditions, the carbon was added to the lime slurry in a feed tank located just prior
to the spray dryer.
During Run 15 on June 3, 1992, it was discovered that the dry carbon feeder had
run out of carbon sometime during the last 10 minutes of the test. For this reason, the
run was repeated as Run 15R.
4.2 Combustor Operating Data
Key combustor operating data for each test run are presented in Table 4-2.
Included are run and condition averages for boiler steam flow, furnace temperature, and
flue gas temperature at the economizer outlet. All of these data were collected from
plant instruments.
4-1
-------
TABLE 4-2. UNIT B COMBUSTOR OPERATING DATA
CAMDEN COUNTY MWC (1992)
Condition
Bl
B2
B3
B4
B5
B6
B7
B8
Run
1
2
3
Average
4
5
6
Average
7
8
9
Average
10
11
12
Average
13
14
15
Average
10
11
12
Average
13
14
15
Average
16
17
18
Average
Boiler Steam
Flow
(ib x lOVhr)
100.0
92.3
94.4
95.5
87.3
98.2
97.8
94.5
99.4
93.5
99.4
97.5
97.6
94.5
99.3
97.1
98.1
96.9
93.8
963
98.1
100.7
99.8
99.6
99.7
93.9
96.7
96.2
93.9
102.4
97.5
97.9
Furnace
Temperature
(T)
1121
1079
1141
1114
1151
1149
1143
1148
1133
1121
1129
1128
1100
1108
1122
1110
1168
1147
1203
1172
1144
1146
1136
1142
1136
1135
1181
1151
1164
1201
1152
1173
Economizer
Outlet
Temperature
(ฐF)
486
484
473
481
501
491
488
493
472
481
484
479
478
488
484
483
474
470
467
470
476
482
472
477
476
475
468
473
482
481
468
477
4-4
-------
TABLE 4-1, CONTINUED
Phase-
Condtion
n-Bio
n-Bii
-B12
DDI1!
-B13
Date
6/8/92
6/9/92
6/11/92
ฃ /1*y Ify)
o/iz/yz
Carbon Type
None
FGD
C/Tk
ruL/
I?
-------
As shown in Table 4-2, the average boiler steam flow during each test condition
ranged from 93,600 to 99,600 Ib/hr, except during Condition BIO when two runs were
less than 90,000 Ib/hr. The furnace temperature condition average ranged from 1110 to
1187ฐF. The average of the flue gas temperature at the economizer outlet ranged from
470 to 493ฐF.
4.3 Sprav Dryer Absorber/Electrostatic Precipitator Operating Data
Operating data for the SD and ESP are presented in Table 4-3. These data
include lime slurry flow rate, SD and ESP outlet temperatures, ESP secondary voltage,
secondary current to each ESP field, and the stack flue gas opacity, and the measured
percent carbon in the fly ash. Dilution water flow rate and spark rate across each ESP
field were also measured, but are not summarized here since there were no unusual
variations during any of the test runs. For each condition, run averages and condition
averages are shown. Plant instruments were used to collect all data, with the exception
of the ESP outlet temperature which was measured by Radian.
The higher SD outlet temperatures of Conditions B6 and B7 reflect the elevated
ESP operating temperature selected for these two conditions. The lime slurry flow rates
for Conditions B6 and B7 were run at higher values to compensate for the higher ESP
inlet temperature (i.e., SO2 capture decreases with increasing temperature and increases
with increasing lime slurry flow rates). No unusual variations were noted in the ESP
voltage or currents during any of the test runs. There is no apparent correlation between
opacity and the amount of carbon in fly ash. The cause of the elevated opacity readings
during Run 19 is unknown.
4.4 Mercury
Table 4-4 presents the Hg results for the testing on Unit B. The table shows Hg
concentrations for each sample fraction, for the total train, and the percent reduction
across the SD/ESP. The HC1 rinses of the KMnO4/H2SO4 impingers were also
4-6
-------
TABLE 4-2, CONTINUED
Condition
B9
BIO
Bll
B12
B13
Run
19
20
21
Average
25
26
27
Average
28
29
30
Average
34
35
36
Average
37
38
39
Average
Boiler Steam
Flow
(Ib x itf/br)
96.8
98.6
85.4
93.6
84.1
87.9
99.4
90.4
97.6
98.1
98.2
98.0
99.2
99.8
99.5
99.5
95.7
92.3
96.8
94.9
Furnace
Temperature
CF)
1139
1153
1115
1136
1109
1127
1171
1136
1173
1192
1197
1187
1159
1175
1192
1176
1143
1161
1187
1164
Economizer
Outlet
Temperature
(ฐF)
472
469
471
471
470
469
481
473
487
487
479
484
478
472
483
478
470
466
473
470
4-5
-------
TABLE 4-3, CONTINUED
Run
Lime
Slurry
Flow Rate
(gpm)
SD
Outlet
Temp
<ฐF)
ESP
Outlet
Temp
<ฐF)
ESP
Voltage
(KV)
ESP
TR1-1
Current
(mA)
ESP
TR1-2
Current
(mA)
ESP
TR1-3
Current
(mA)
ESP
TR1-4
Current
(mA)
Opacity
<%)
Carbon
In Fly
Ash
(%)
Phase I, Condition B5
13
14
15
Average
8.1
8.2 .
8.1
8.1
275
277
262
271
281
278
273
277
47
46
46
46
192
230
285
236
434
437
439
437
426
431
411
423
448
448
448
448
0.0
0.0
0.0
0.0
1.86
Phase n, Condition B6
10
11
12
Average
9.6
10.6
9.1
9.8
348
350
349
349
353
348
348
350
46
46
46
46
126
135
115
125
424
405
395
408
Phase II, Condition B7
13
14
15
Average
8.8
9.0
8.8
8.9
352
352
344
349
347
346
342
345
45
45
45
45
122
150
151
141
418
435
426
426
440
440
440
440
448
448
448
448
1.0
1.0
1.0
1.0
1.56
447
448
448
448
448
448
448
448
1.0
1.0
1.0
1.0
1.53
Phase n, Condition B8
16
17
18
Average
9.2
9.1
9.3
9.2
267
263
262
264
274
272
270
272
48
48
48
48
229
247
274
250
440
440
440
440
389
379
374
380
448
448
448
448
1.0
1.0
1.0
1.0
1.89
Phase D, Condition B9
19
20
21
Average
9.2
9.1
9.2
9.2
266
266
265
266
268
265
269
268
48
48
48
48
289
284
273
282
440
440
440
440
376
372
372
373
448
448
448
448
3.2
1.5
1.0
1.9
1.69
-------
TABLE 4-3. UNIT B SPRAY DRYER ABSORBER/
ESP OPERATING DATA
CAMDEN COUNTY MWC (1992)
Run
Lime
Slurry
Flow Rate
(gpm)
SD
Outlet
Temp
<ฐF)
ESP
Outlet
Temp
<ฐF)
ESP
Voltage
(KV)
ESP
TR1-1
Current
(nปA)
ESP
TR1-2
Current
(mA)
Phase I, Condition B}
1
2
3
Average
8.2
8.2
8.2
8.2
269
267
262
266
274
277
263
271
47
48
46
47
293
292
292
292
440
439
443
441
ESP
TR1-3
Current
(mA)
419
408
403
410
ESP
TR1-4
Current
(mA)
448
448
448
448
Opacity
(%)
Carbon
In Fly
Ash
(%)
0.0
0.0
0.0
0.0
1.41
Phase I, Condition B2
4
5
6
Average
8.1
8.2
8.1
8.1
274
266
275
272
254
269
272
265
47
48
48
48
211
268
282
254
433
443
441
439
416
417
413
415
448
448
448
448
0.1
0.3
0.5
0.3
1.45
Phase I, Condition B3
7
8
9
Average
8.3
8.2
8.2
8.2
264
272
273
270
271
270
277
273
46
47
47
47
278
269
277
275
440
440
440
440
412
422
418
417
448
448
448
448
1.0
1.3
1.8
1.4
1.16
Phase I, Condition B4
10
11
12
Average
8.2
8.2
8.2
8.2
265
290
291
282
273
287
292
284
46
46
46
46
241
167
217
208
440
431
439
437
419
436
435
430
448
448
448
448
0.0
0.0
0.1
0.0
1.82
-------
TABLE 4-4. UNIT B MERCURY RESULTS
Condition
Bl
B2
B3
B4
BS
B6
B7
Run
1
i
2
3
AVG
4
5
6
AVG
7
8
9
AVG
10
11
12
AVG
13
14
15
AVG
10
11
12
AVG
13
14
15R*
AVG
Mercury Concentration (u
Inlet
Filter &
Probe
Rinse
231
498
412
380
537
330
560
476
229
396
322
316
339
287
331
313
397
606
219
407
120
137
120
126
395
248
345
329
IINO3/
H202
Impingers
121
820
294
412
434
262
274
323
362
238
241
280
145
143
177
144
273
206
410
297
240
98
221
187
557
248
431
412
KMn04/
II2S04
Impingers
4.9
45.0
4.9
18.2
0.5
0.8
0.5
0.6
0.7
5.9
23.0
9.9
6.8
9.5
3.5
8.2
9.1
8.3
15.0
10.8
4.6
13.7
7.4
8.6
11.3
8.9
2.2
7.5
Total
356
1363
711
810
972
593
835
800
593
639
586
606
491
440
512
465
680
820
644
715
365
249
349
321
964
506
778
749
g/dscmat7%O2)
Outlet
Filter &
Probe
Rinse
0.13
5.01
0.78
1.97
2.69
0.11
0.41
1.07
ND
0.22
ND
0.07
0.05
ND
ND
0.03
0.18
0.38
0.17
0.24
0.11
ND
0.04
0.05
1.23
0.32
1.27
0.94
IINO3/
II202
Impingers
151
132
46
110
256
62
128
149
116
18
80
71
14.1
10.5
14.5
12.3
4.32
7.08
3.10
4.84
283
165
252
233
91.7
18.6
49.2
53.2
KMnO4/
II2SO4
. Impingers
23.7
73.4
7.5
34.9
37.1
0.7
20.5
19.4
17.5
10.4
21.8
16.6
6.9
3.4
2.9
5.2
4.9
5.9
8.5
6.4
17.9
12.8
8.6
13.1
14.1
3.3
8.9
8.8
Total
175
210
54
147
296
63
149
169
134
29
102
88
21.0
13.9
17.4
17.5
9.4
13.3
11.8
11.5
301
177
261
246
107
22.3
59.4
62.9
Removal
Efficiency
(%)
50.8
84.6
92.4
75.9
69.5
89.4
82.2
80.4
77.5
95.5
82.6
85.2
95.7
96.8
96.6
96.3
98.6
98.4
98.2
98.4
17.6
28.7
25.2
23.8
88.9
95.6
92.4
92.3
7s
ป
O
ND = No! Delected
* Run 15R was conducted due to possible problems caused by an interruption in carbon feed toward the end of Run 15.
-------
TABLE 4-3, CONTINUED
Run
Lime
Slurry
How Rate
(gpm)
SD
Outlet
Temp
<ฐF)
ESP
Outlet
Temp
<ฐF)
ESP
Voltage
(KV)
ESP
TR1-1
Current
(mA)
ESP
TR1-2
Current
(mA)
ESP
TR1-3
Current
(mA)
ESP
TR1-4
Current
(mA)
Opacity
<%)
Carbon
In Fly
Ash
(%)
Phase n, Condition BIO
25
26
27
Average
9.2
9.1 .
9.2
9.2
269
266
258
264
276
274
274
275
47
49
47
48
287
333
295
305
445
444
438
442
372
370
356
366
448
448
448
448
1.0
1.0
1.1
1.0
1.59
Phase II, Condition Bll
28
29
30
Average
9.2
9.2
9.1
9.1
271
273
269
271
279
280
277
279
48
48
47
48
269
282
286
279
441
443
438
441
385
382
374
380
448
448
448
448
1.0
1.0
1.7
1.2
2.20
Phase II, Condition B12
34
35
36
Average
7.9
7.9
7.9
7.9
269
269
275
271
277
279
282
279
Phase II, Condition B13
37
38
39
Average
8.0
7.9
8.0
7.9
266
263
264
265
273
273
272
272
48
48
47
48
260
280
249
263
438
440
441
439
47
48
48
48
273
283
280
278
441
442
445
443
387
382
385
385
382
388
389
386
448
448
448
448
1.0
1.0
1.0
1.0
1.20
448
448
448
448
1.0
1.0
1.0
1.0
1.16
-------
analyzed, but the levels of Hg were generally less than the detection limit. Because of
the consistently low Hg level in these samples, the Hg found in this fraction has not been
included in the table and is not discussed further.
The inlet Hg concentrations during each test condition averaged from 321 to
810 ^g/dscm. The maximum inlet concentration for an individual run was 1363 ^g/dscm
during Condition Bl, Run 2. The minimum run concentration was 249 ^g/dscm during
Condition B6, Run 11. The average filter concentration levels during each condition
ranged from 39 to 67% of the total Hg collected. The average HNO3/H2O2 impinger
concentration level ranged from 31 to 58%, with the KMnO4/H2SO4 impinger containing
roughly 4% of the total Hg collected.
The condition average outlet. Hg levels ranged from 11.5 to 292 ^ag/dscm. The
maximum outlet concentration for an individual run was 389 /ig/dscm during
Condition BIO, Run 25. The minimum concentration for an individual run was
9.4 /ig/dscm during Condition B5, Run 13. The filters contained less than 8% of the
total Hg collected at the outlet. The average HNO3/H2O2 impinger concentration levels
ranged from 42 to 93% of the total Hg content, while KMnO4/H2SO4 impinger levels
ranged from 4 to 56%.
Percent reduction averages ranged from 24%, with no carbon injection during
Condition B6, to 98% with the high carbon injection rate during Condition B5.
4.5 Cadmium and Lead
Flue gas concentrations of Cd and Pb were determined during six test conditions.
Five of these six test conditions were conducted at 270ฐF: no carbon injection (BIO), dry
carbon injection at a low and high feed rate (B2 and Bll, respectively), and slurry
carbon injection at a medium and high feed rate (B13 and B12, respectively). The sixth
test condition was at 350ฐF and dry carbon injection at a high rate (B7). The results for
each metal are shown in Table 4-5 and include front-half and back-half results for both
4-12
-------
TABLE 4-4. (continued)
Condition
B8
B9
BIO
Bll
B12
B13
Run
16
17
18
AVG
19
20
21
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Mercury Concentration (ug/dscm at 7% O2)
Inlet
Filter &
Probe
Rinse
262
206
184
217
224
414
192
277
351
165
205
241
295
465
193
318
139
193
195
176
117
290
485
297
IINO3/
11202
Impingers
262
238
339
280
258
527
262
349
300
253
171
241
327
161
427
305
157
324
98
193
234
79
485
266
KMnO4/
II2SO4
Impingers
20.6
10.3
2.2
11.0
2.6
16.9
8.9
9.5
11.4
14.3
7.4
11.0
3.7
9.8
44.7
19.4
3.0
4.5
7.0
4.8
30.6
7.7
5.2
14.5
Total
545
455
525
508
485
957
463
635
663
433
384
493
626
635
664
642
299
521
300
373
382
377
974
578
Outlet
Filter &
Probe
Rinse
2.58
0.49
0.46
1.18
4.85
0.42
1.55
2.27
0.17
1.68
1.99
1.28
1.10
1.24
0.75
1.03
1.09
1.74
2.21
1.68
7.51
19.88
9.73
12.37
IINO3/
11202
Impingers
29.9
17.9
20.3
22.7
92.9
166
116
125
358
265
199
274
11.0
6.8
10.0
9.3
39.6
69.0
56.8
55.1
57.0
44.3
141
80.7
KMn04/
H2SO4
Impingers
5.3
4.3
3.3
4.3
5.4
3.1
6.4
4.9
30.8
12.3
6.4
16.5
8.0
8.3
4.8
7.1
9.6
5.8
10.1
8.5
13.2
16.8
7.0
12.4
Total
37.8
22.7
24.1
28.2
103
170
124
132
389
279
207
292
20.1
16.4
15.6
17.4
50.3
76.5
69.1
65.3
77.7
81.0
158
105
Removal
Efficiency
(%)
93.1
95.0
95.4
94.5
78.7
82.2
73.2
78.1
41.3
35.7
46.0
41.0
96.8
97.4
97.7
97.3
83.2
85.3
77.0
81.8
79.7
78.5
83.8
80.7
-------
the SD inlet and ESP outlet sampling locations. Due to in advertent archiving of
samples, the SD inlet back-fractions from Runs 26 and 30, and the ESP outlet back-half
fraction from Run 27 were not analyzed.
For Cd, average reduction efficiencies across the SD/ESP were 99.6% during
Condition BIO without carbon injection and 99.5 to 99.9% with carbon injection. For Pb,
average reduction efficiencies were 99.6% without carbon injection and 99.7 to 99.9%
with carbon injection. Removal efficiencies without carbon injection were in excess of
99.9% during two of the runs and 98.8% during the third run. The average metals
concentrations at the ESP outlet were 4 to 8 /ig/dscm for Cd and 14 to 68 /^g/dscm for
Pb for each test condition.
Of the total Cd and Pb concentrations measured at the SD inlet, over 99.8% was
in the front-half except during Run 36. During this run, the back-half accounted for 10%
of the total Cd and 13% of the Pb. These higher values may have been caused by
penetration of paniculate through or around the sampling train filter. At the ESP outlet
sampling location, the front-half generally accounted for over 70% of the total Cd and
90% of the total Pb, but was lower on several runs.
4.6 Other Metals
Flue gas concentrations of antimony, arsenic, barium, beryllium, chromium, cobalt,
copper, manganese, molybdenum, nickel, silver, thallium, and vanadium were also
determined during the same six test conditions discussed in Section 4.5. The results for
each metal are shown in Table 4-6 and include front-half and back-half results for both
the SD inlet and ESP outlet sampling locations. As with Cd and Pb, the SD inlet back-
fractions from Runs 26 and 30, and the ESP outlet back-half fraction from Run 27 were
not analyzed. The results of selenium QC spike recoveries were not satisfactory and
data for selenium are not reported.
4-14
-------
TABLE 4-5. UNIT B CADMIUM AND LEAD RESULTS8
CAMDEN COUNTY MWC (1992)
Condition
B2
B2
B2
B2
B7
B7
B7
B7
BIO
BIO
BIO
BIO
Bll
Bll
Bll
Bll
B12
B12
B12
B12
B13
B13
B13
B13
Run
4 ,
5
6
AVG
13
14
1SR
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Cadmium (ug/dscm at 7% O2)
Inlet
Front
Half
1485
1481
1550
1505
3055
1348
1327
1910
1137
1102
1027
1089
1159
1251
1321
1244
1217
3446
1330
1998
1478
1275
1783
1512
Estimated Detection Limit) 5.91
Back
Half
0.32
0.30
0.24
0.28
ND
ND
ND
ND
0.74
NA
0.39
0.57
ND
ND
NA
ND
ND
ND
142
47.3
0.32
0.31
2.13
0.92
0.12
Total
1486
1482
1550
1506
3055
1348
1327
1910
1138
1102
1027
1089
1159
1251
1321
1244
1217
3446
1472
2045
1479
1275
1785
1513
6.02
Outlet
Front
Half
8.96
4.58
3.97
5.84
4.09
2.00
1.19
2.43
0.64
1.15
9.63
3.81
1.17
1.38
3.88
2.14
1.78
0.98
0.82
.19
.15
.99
.28
.47
0.40
Back
Half
0.63
2.29
2.31
1.74
0.17
0.60
0.50
0.42
0.42
0.28
-NA
0.35
ND
1.52
0.18
0.57
0.38
0.25
0.45
0.36
ND
0.41
1.54
0.65
0.10
Total
9.59
6.87
6.28
7.58
4.26
2.60
1.69
2.85
1.06
1.43
9.63
4.04
1.17
2.90
4.06
2.71
2.16
1.23
1.27
1.55
1.15
2.40
2.82
2.12
Removal
Efficiency
(%)
99.35
99.54
99.59
99.50
99.86
99.81
99.87
99.85
99.91
99.87
99.06
99.63
99.90
99.77
99.69
99.78
99.82
99.%
99.91
99.92
99.92
99.81
99.84
99.86
Lead (ug/dscm at 7% O2)
Inlet
Front.
Half
27421
12535
27416
22458
23362
24837
24130
24110
14472
20939
15403
16938
23188
18759
17269
19739
20871
25341
15960
20724
30647
17573
19384
22535
0.49| NA| 59.1
Back
Half
2.40
4.33
0.97
2.57
0.72
1.77
0.88
1.12
4.55
NA
3.34
3.94
0.38
1.70
NA
1.04
ND
0.49
2483
828
1.98
3.48
12.4
5.95
Total
27424
12540
27417
22460
23363
24838
24131
24111
14476
20939
15407
16941
23188
18761
17269
19739
20871
25342
18442
21552
30649
17576
19396
22541
l_ 0.35 L 59.4
Outlet
Front
Half
99.9
54.7
23.1
59.2
22.6
18.6
15.5
18.9
5.57
6.79
186
66.2
21.9
20.1
57.6
33.2
21.9
6.97
11.4
13.4
11.7
53.5
10.8
25.3
1.48
Back
Half
3.59
1.40
20.5
8.49
0.45
0.56
0.50
0.51
0.30
ND
NA
0.15
0.56
0.90
0.59
0.68
ND
0.44
0.49
0.31
ND
ND
1.66
0.55
Total
103
56.1
43.6
67.7
23.0
19.2
16.0
19.4
5.87
6.79
186
66.3
22.5
21.0
58.2
33.9
21.9
7.41
11.8
13.7
11.7
53.5
12.4
25.9
Removal
Efficiency
(%)
99.62
99.55
99.84
99.67
99.90
99.92
99.93
99.92
99.%
99.97
98.79
99.57
99.90
99.89
99.66
99.82
99.90
99.97
99.94
99.94
99.%
99.70
99.94
99.87
0.30| 1.781 NA
f-
ป-ป
U)
BND 1** Detected
NA Not Analyzed
NC - Not Calculated
-------
TABLE 4-6, CONTINUED
Condition
B2
B2
B2
B2
B7
B7
B7
B7
BIO
BIO
BIO
BIO
Bll
Bll
Bll
Bll
B12
B12
B12
B12
B13
B13
B13
B13
Run
4 ,
5
6
AVG
13
14
1SR
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Estimated Detection Limit
Barium (ug/dscm at 7% O2)
Inlet
Front
Half
2742
3077
3338
3052
3235
3903
3102
3413
2274
2865
2054
2398
2213
1965
1117
1765
1044
2027
168
1080
1983
1835
2908
2242
11.80
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
3.20
NA
1.54
2.37
1.26
1.34
NA
1.30
ND
ND
40.8
13.6
ND
ND
4.65
1.55
1.18
Total
2742
3077
3338
3052
3235
3903
3102
3413
2277
2865
2055
2399
2215
1967
1117
1766
1044
2027
209
1093
1983
1835
2912
. 2243
13.0
Outlet
Front
Half
5.76
5.85
5.51
5.71
5.93
5.32
4.66
5.30
2.93
3.62
6.35
4.30
2.34
2.63
3.26
2.74
2.66
2.30
2.02
2.33
4.59
8.72
4.48
5.93
0.99
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
2.20
NA
1.10
0.88
0.97
0.88
0.91
ND
ND
ND
ND
ND
ND
ND
ND
Total
5.76
5.85
5.51
5.71
5.93
5.32
4.66
5.30
2.93
5.82
6.35
5.03
3.22
3.60
4.14
3.65
2.66
2.30
2.02
2.33
4.59
8.72
4.48
5.93
0.991 1.98
Removal
Efficiency
(%)
99.79
99.81
99.83
99.81
99.82
99.86
99.85
99.84
99.87
99.80
99.69
99.79
99.85
99.82
99.63
99.79
99.74
99.89
99.04
99.79
99.77
99.52
99.85
99.74
Beryllium (ug/dscm at 7% O2)
Inlet
Front
Half
3.31
3.53
26.2
11.0
ND
4.08
4.14
2.74
2.69
2.98
3.00
2.89
2.11
2.86
2.64
2.54
2.52
ND
2.84
1.79
ND
ND
ND
NA
NAI 1.98
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
ND
ND
ND
ND
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.20
Total
3.31
3.53
26.2
11.0
ND
4.08
4.14
2.74
2.69
2.98
3.00
2.89
2.11
2.86
2.64
2.54
2.52
ND
2.84
1.79
ND
ND
ND
ND
Outlet
Front
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Total
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
2.171 0.201 0.201 0.40
Removal
Efficiency
(%)
>99.99
>99.99
> 99.99
>99.99
NC
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
> 99.99
>99.99
>99.99
NC
>99.99
>99.99
NC
NC
NC
NC
NA
O\
ND
NA
Not Delected
Not Analyzed
NC - Not Calculated
-------
TABLE 4-6. UNIT B OTHER METAL RESULTS8
CAMDEN COUNTY MWC (1992)
Condition
B2
B2
B2
B2
B7
B7
B7
B7
BIO
BIO
BIO
BIO
Bll
Bll
Bll
Bll
B12
B12
B12
B12
B13
B13
B13
B13
Run
4 .
5
6
AVG
13
14
15R
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Estimated Detection Limit
Antimon
Inlet
From
Half
4227
4786
3457
4157
3414
3371
6205
4330
3204
3527
23%
3042
2424
4288
3251
3321
2435
4257
3813
3502
3786
4828
3877
4163
177
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
27.4
13.7
ND
31.3
NA
15.6
12.2
15.2
213
80.1
45.1
34.8
31.0
36.9
11.8
Total
4227
4786
3457
4157
3414
3371
6205
4330
3204
3527
2423
3052
2424
4319
3251
3331
2447
4273
4025
3582
3831
4862
3908
4200
189
v(ug/dscmat7%O2)
Outlet
Front
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
9.88
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
9.88
Total
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
19.8
Removal
Efficiency
(%)
> 99.99
>99.99
> 99.99
> 99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
> 99.99
> 99.99
>99.99
Arsenic (ug/dscm at 7% O2)
Inlet
Front
Half
1257
1367
1073
1232
1132
852
603
862
579
838
676
697
738
670
1219
876
1217
588
1685
1163
559
637
523
573
NA| 1.18
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
0.52
0.26
ND
ND
NA
ND
1.48
ND
78.0
26.5
ND
ND
3.10
1.03
0.12
Total
1257
1367
1073
1232
1132
852
603
862
579
838
677
698
738
670
1219
876
1219
588
1763
1190
559
637
526
574
Outlet
Front
Half
1.41
1.03
0.59
1.01
1.23
0.68
0.74
0.88
ND
ND
4.69
1.56
ND
ND
1.38
0.46
0.66
0.38
0.43
0.49
0.75
1.48
0.54
0.92
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Total
1.41
1.03
0.59
1.01
1.23
0.68
0.74
0.88
ND
ND
4.69
1.56
ND
ND
1.38
0.46
0.66
0.38
0.43
0.49
0.75
1.48
0.54
0.92
Removal
Efficiency
(%)
99.89
99.92
99.95
99.92
99.89
99.92
99.88
99.90
> 99.99
>99.99
99.31
99.78
>99.99
>99.99
99.89
99.95
99.95
99.93
99.98
99.%
99.87
99.77
99.90
99.84
1.301 0.401 0.101 0.491 NA
f-
c;
*ND - Not Delected
NA - Not Analyzed
NC Not Calculated
-------
TABLE 4-6, CONTINUED
Condition
B2
B2
B2
B2
B7
B7
B7
B7
BIO
BIO
BIO
BIO
Bll
Bll
Bll
Bll
B12
B12
B12
B12
B13
B13
B13
B13
Run
4 .
5
6
AVG
13
14
1SR
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Estimated Detection Limit
Cobalt
Inlet
Front
Half
81.1
69.5
72.7
74.4
88.1
101
276
155
71.3
132
83.9
95.8
86.4
116
82.3
94.9
60.9
50.7
97.5
69.7
57.7
126
98.9
94.0
11.8
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
ND
ND
ND
ND
NA
ND
ND
ND
1.60
0.53
ND
ND
ND
ND
1.18
Total
81.1
69.5
72.7
74.4
88.1
101
276
155
71.3
132
83.9
95.8
86.4
116
82.3
94.9
60.9
50.7
99.1
70.2
57.7
126
98.9
94.0
13.0
ug/dscm at 7% O2)
Outlet
Front
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.99
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.99
Total
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.98
Removal
Efficiency
(%)
>99.99
>99.99
>99.99
>99.99
>99.99
> 99.99
> 99.99
>99.99
>99.99
>99.99
>99.99
>99.99
> 99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
Copper (ug/dscm at 7% O2)
Inlet
Front
Half
4456
4216
7748
5473
4313
4435
3964
4237
2894
3857
4279
3677
3478
3662
3149
3430
23480
5676
4788
11315
3606
3862
3489
3652
NAI 23.6
Back
Half
1.71
ND
ND
0.57
ND
ND
ND
ND
ND
NA
ND
ND
ND
1.97
NA
0.98
ND
ND
301
100
ND
ND
6.20
2.07
2.36
Total
4458
4216
7748
5474
4313
4435
3964
4237
2894
3857
4279
3677
3478
3664
3149
3431
23480
5676
5089
11415
3606
3862
3495
3654
Outlet
Front
Half
29.4
44.5
10.2
28.1
8.61
9.18
6.34
8.04
3.43
8.02
18.6
10.0
4.46
4.36
6.89
5.23
8.19
2.44
2.84
4.49
4.31
8.87
6.02
6.40
Back
Half
2.05
3.31
41.0
15.5
ND
ND
ND
ND
ND
4.23
NA
2.12
5.71
2.97
ND
2.89
ND
4.81
1.45
2.09
4.03
ND
39.7
14.6
26.01 1.981 1.98
Total
31.5
47.8
51.2
43.5
8.61
9.18
6.34
8.04
3.43
12.3
18.6
11.4
10.2
7.33
6.89
8.13
8.19
7.25
4.29
6.58
8.34
8.87
45.7
21.0
3.95
Removal
Efficiency
(%)
99.29
98.87
99.34
99.20
99.80
99.79
99.84
99.81
99.88
.99.68
99.56
99.69
99.71
99.80
99.78
99.76
99.97
99.87
99.92
99.94
99.77
99.77
98.69
99.43
NA
*ND Not Delected
NA - Not Analyzed
NC - Not Calculated
-------
TABLE 4-6, CONTINUED
Condition
B2
B2
B2
B2
B7
B7
B7
B7
BIO
BIO
BIO
BIO
Bll
Bll
Bll
Bll
B12
B12
B12
B12
B13
B13
B13
B13
Run
4
5
6
AVG
13
14
1SR
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Chromium (ug/dscm at 7% O2)
Inlet
Front
Half
1485
1367
1550
1467
1222
1419
1431
1357
827
1058
10269
4051
1476
1340
1117
1311
1044
568
1330
980
667
1931
1008
1202
Estimated Detection Limit! 1 1.8
Back
Half
3.54
1.82
1.67
2.34
4.31
8.87
4.65
5.95
2.58
NA
2.05
2.32
ND
1.34
NA
0.67
2.96
ND
18.6
7.19
2.52
3.48
4.26
3.42
1.18
Total
1489
1369
1551
1470
1226
1428
1435
1363
830
1058
10271
4053
1476
1341
1117
1311
1047
568
1349
988
670
1935
1012
1205
Outlet
Front
Half
3.46
8.39
2.95
4.93
3.67
3.73
3.88
3.76
2.36
2.65
7.06
4.02
2.49
2.35
3.38
2.74
1.98
1.95
2.27
2.07
4.45
6.27
5.25
5.32
13.01 0.99
Back
Half
2.43
2.03
26.9
10.5
32.5
6.65
14.2
17.8
3.79
4.14
NA
3.97
0.88
2.21
.31
.47
.37
.46
.01
1.28
ND
ND
4.86
1.62
0.99
Total
5.89
10.4
29.9
15.4
36.1
10.4
18.1
21.5
6.15
6.79
7.06
6.67
3.36
4.56
4.70
4.21
3.35
3.42
3.28
3.35
4.45
6.27
10.1
6.94
1.98
Removal
Efficiency
(%)
99.60
99.24
98.08
98.95
97.05
99.27
98.74
98.42
99.26
99.36
99.93
99.84
99.77
99.66
99.58
99.68
99.68
99.40
99.76
99.66
99.34
99.68
99.00
99.42
Vanadium (ug/dscm at 7% O2)
Inlet
Front
Half
297
353
381
344
270
337
310
306
207
276
248
243
232
286
244
254
217
162
301
227
216
328
213
253
NAI 23.6
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
ND
ND
ND
ND
NA
ND
ND
ND
3.19
1.06
ND
ND
ND
ND
Total
297
353
381
344
270
337
310
306
207
276
248
243
232
286
244
254
217
162
305
228
216
328
213
253
2.361 26.0
Outlet
Front
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Total
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Removal
Efficiency
(%)
>99.99
>99.99
>99.99
> 99.99
>99.99
>99.99
>99.99
>99.99
>99.99
> 99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
1.981 1.981 1.951 NA
*ND Not Delected
NA - Not Analyzed
NC - Not Calculated
-------
TABLE 4-6, CONTINUED
Condition
82
B2
B2
B2
B7
B7
B7
B7
BIO
BIO
BIO
BIO
Bll
Bll
Bll
Bll
B12
B12
B12
B12
B13
B13
B13
B13
Run
4
5
6
AVG
13
14
15R
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Estimated Detection Limit
Nickel I
Inlet
Front
Half
446
433
381
420
395
390
362
382
248
485
359
364
295
322
437
351
243
253
328
275
216
348
291
285
23.6
Back
Half
2.40
3.19
ND
1.86
ND
5.68
7.58
4.42
6.20
NA
2.65
4.43
ND
2.05
NA
1.03
ND
ND
5.14
1.71
ND
5.02
ND
1.67
2.36
Total
448
436
381
422
395
396
370
387
254
485
362
367
295
324
437
352
243
253
333
277
216
353
291
287
26.0
ug/dscm at 7% O2)
Outlet
Front
Half
17.9
5.60
8.07
10.5
ND
ND
ND
ND
2.07
2.29
1.41
1.93
1.90
4.29
1.44
2.54
5.05
ND
1.83
2.29
ND
ND
8.07
2.69
1.98
Back
Half
1.79
6.87
35.9
14.8
16.9
18.6
9.83
15.1
4.72
4.59
NA
4.65
ND
5.46
16.9
7.46
ND
1.74
5.99
2.58
ND
ND
26.9
8.96
1.98
Total
19.7
12.5
43.9
25.4
16.9
18.6
9.83
15.1
6.79
6.88
1.41
6.58
1.90
9.75
18.3
10.0
5.05
1.74
7.82
4.87
ND
ND
35.0
11.7
3.95
Removal
Efficiency
(%)
95.60
97.14
88.48
93.98
95.72
95.29
97.34
96.09
97.33
98.58
99.61
98.21
99.36
96.99
95.80
97.16
97.92
99.31
97.65
98.24
> 99.99
>99.99
87.98
95.93
NA
'NO Not Detected
NA - Not Applicable
NC * Not Calculated
-------
TABLE 4-6, CONTINUED
Condition
B2
B2
B2
B2
B7
B7
B7
B7
BIO
BIO
BIO
BIO
Bll
Bll
Bll
Bll
B12
B12
B12
B12
B13
B13
B13
B13
Run
4
5
6
AVG
13
14
15R
AVG
25
26
27
AVG
28
29
30
AVG
34
35
36
AVG
37
38
39
AVG
Estimated Detection Limit
Manganese (ug/dscm at 7% O2)
Inlet
Front
Half
3542
3647
3338
3509
3594
4258
3102
3651
2377
3857
3252
3162
3057
3930
4673
3887
2522
1825
3635
2661
2524
4248
2520
3097
11.8
Back
Half
19.4
8.89
3.70
10.7
82.7
213
77.6
124
21.7
NA
18.0
19.8
2.95
5.00
NA
3.98
4.70
5.27
168
59.5
16.8
16.4
523
186
1.18
Total
3561
3656
3341
3519
3677
4471
3180
3776
2399
3857
3270
3175
3060
3935
4673
3889
2527
1830
3804
2720
2541
4265
3043
3283
Outlet
Front
Half
3.97
11.2
3.59
6.25
2.54
45.2
233
16.7
1.57
1.68
834
3.86
1.54
4.70
3.13
3.12
4.23
1.19
132
2.25
36.1
68.8
6.53
37.2
13.01 0.99
Back
Half
25.6
21.6
320303
23.6
116
18.6
63.4
65.9
17.2
47.6
NA
32.4
2.78
23.5
5.95
10.7
203
17.4
6.94
15
959
19.9
433
243
0.99
Total
29.6
32.8
320307
31.2
119
63.9
65.7
82.6
18.7
493
834
253
432
28.2
9.08
13.9
24.7
18.6
8.26
17.2
45.7
88.7
50.1
613
1.98
Removal
Efficiency
(%)
99.17
99.10
-9486.24
99.11
96.78
9837
97.93
97.81
99.22
98.72
99.74
99.20
99.86
99.28
99.81
99.64
99.02
98.98
99.78
9937
98.20
97.92
9836
98.13
NA
Molybdenum (ug/dscm at 7% O2)
Inlet
Front
Half
274
125
113
171
ND
903
ND
30.2
83.7
101
608
264
179
134
953
136
84.4
58.8
973
80.2
119
145
118
127
59.1
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
ND
ND
ND
ND
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
5.91
Total
274
125
113
171
NA
903
NA
30.2
83.7
101
608
264
179
134
953
136
84.4
58.8
973
80.2
119
145
118
127
Outlet
Front
Half
243
25.4
25.6
25.1
25.4
25.3
233
24.7
15.0
17.6
10.9
143
12.4
11.8
10.6
11.6
10.9
13.2
10.7
11.6
25.0
273
23.0
25.2
Back
Half
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
ND
ND
ND
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
65.01 4.94| 4.94
Total
243
25.4
25.6
25.1
25.4
25.3
23.3
24.7
15.0
17.6
10.9
14.5
12.4
11.8
10.6
11.6
10.9
13.2
10.7
11.6
25.0
27.5
23.0
25.2
Removal
Efficiency
(%)
91.13
79.71
7737
8530
NC
72.05
NC
18.24
82.07
82.61
98.20
9431
93.06
91.23
88.86
91.48
87.05
77.47
89.01
8530
78.97
81.00
8031
80.22
9.881 NA
*ND Not Detected
NA - Not Analyzed
NC - Not Calculated
bThere was Ma contamination of the back half outlet fraction for Run 6 due to bkwback of Ma from the KMnO4 impingers into the HNO3 impinge re following the post-run leak check.
This value is not used in the averages.
-------
TABLE 4-7. UNIT B PARTICULATE
MATTER RESULTS
CAMDEN COUNTY MWC (1992)
Phase-
Condition
I-B1
I-B2
I-B3
I-B4
I-B5
II-B6
H-B7
H-B8
Run
Number
1
2
3
Avg
4
5
6
Avg
7
8
9
Avg
10
11
12
Avg
13
14
15
Avg
10
11
12
Avg
13
14
. 15
Avg
16
17
18
Avg
Inlet PM
(g/dscm @ 7% O2)
NAง
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4.77
6.48
3.53
4.93
4.80
6.52
6.16
5.83
6.97
6.69
6.14
6.60
Outlet PM
(g/dscm @ 7% O2)
0.0040
0.0036
0.0042
0.0039
0.0036
0.0022
0.0022
0.0027
0.0012
0.0024
0.0041
0.0026
0.0014
0.0021
0.0024
0.0020
0.0075
0.0026
0.0019
0.0040
0.0012
0.0010
0.0011
0.0011
0.0013
0.0013
0.0015
0.0014
0.0019
0.0014
0.0013
0.0015
Removal
Efficiency
(%)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
99.98
99.99
99.97
99.98
99.97
99.98
99.97
99.98
99.97
99.98
99.98
99.98
4-22
-------
4.7 Particulate Matter
Table 4-7 presents the PM concentrations for each run, as well as condition
averages. Because of the need for expedited Hg analysis of the EPA SW-846
Method 0012 front-half fraction collected at the SD inlet during Phase I, gravimetric
analyses of the probe rinse and filter catch were not performed. As a result, inlet PM
data are not available for these runs.
The average inlet concentrations for the Phase II-B test conditions ranged from
4.76 to 6.60 g/dscm, and the corresponding outlet averages ranged from 0.0011 to
0.0088 g/dscm. All of the individual runs achieved greater than 99.7% reduction of
paniculate matter.
4.8 CDD/CDF
Table 4-8 presents the CDD/CDF results for Conditions BIO, Bll, and B12. The
table presents economizer outlet and stack concentrations of each congener; the total
CDD, total CDF, and combined CDD/CDF concentrations; and the removal efficiencies
for CDD, CDF, and combined CDD/CDF.
Inlet CDD concentrations during individual runs ranged from 18 to 103 ng/dscm.
Inlet CDF concentrations ranged from 114 to 302 ng/dscm. Total CDD/CDF
concentrations averaged 46.8 ng/dscm during Condition BIO, 5.6 ng/dscm during
Condition Bll, and 10.5 ng/dscm during Condition B12. Total CDD/CDF removal
efficiencies were greater than 95% for runs with carbon injection. Removal efficiencies
were between 77 and 80% without carbon injection.
4-21
-------
TABLE 4-8. CDD/CDF RESULTS
CAMDEN COUNTY MWC (1992)
CONGENER
MOXINS
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1 234678 HpCDD
Other HpCDD
OcUCDD
Total CDD
FURANS
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
1 23678 HxCDF
123789 HxCDF
234678 HxCDF
Other HxCDF
1 234678 HpCDF
1 234789 HpCDF
Other HpCDF
Octa CDF
Total CDF
Total CDD + CDF
Condition BIO (no carton injection. 270 F ESP inlet temperature)
Inlet
(ng/dscm
@7%O2)
0.5S8
11.8
1 1.01
5.40
0.589
0.693
0.589
5.47
4.96
4.24
8.37
43.7
4.13
110
4.75
4.44
50.8
3.41
ND
ND
1.00
18.3
5.89
1.76
4.75
4.55
213
257
Run 25
Outlet
(ng/dfcm
@ 7% O2)
0.147
2.13
0.162
1.38
0.0735
0.0808
0.0661
0.588
0.353
0.331
0.676
5.99
1.10
31.2
1.03
1.10
11.1
0.514
0.581
0.360
0.110
2.92
0.882
0.140
0.522
0.375
52.0
58.0
Removal
Efficiency
(%)
73.7
82.0
84.0
74.4
87.5
88.3
88.8
89.2
92.9
92.2
91.9
86.3
73.3
71.5
78.4
75.2
78.1
84.9
0.0
0.0
89.0
84.1
85.0
92.1
89.0
91.8
75.6
77.5
Run 2*
Inlet
(ng/dscm
@7%02)
0.589
5.60
1.11
5.78
0.445
0.622
0.511
4.42
3.56
3.11
7.22
33.0
4.22
86.9
5.22
4.67
53.5
ND
ND
ND
0.945
17.9
5.33
1.33
1.89
3.00
185
218
Outlet
(ng/dicm
@7%02)
0.112
1.17
0.136
0.665
0.0568
0.0608
0.0520
0.471
0.240
0.224
0.424
3.61
0.801
23.2
0.881
0.801
8.73
0.464
0.408
0.112
0.104
2.59
0.648
0.0881
0.304
0.176
39.3
42.9
Removal
Efficiency
(*)
81.0
79.1
87.8
88.5
87.2
90.2
89.8
89.4
93.2
92.8
94.1
89.0
81.0
73.3
83.1
82.8
83.7
0.0
0.0
0.0
89.0
85.5
87.8
93.4
83.9
94.1
78.7
80.3
Run 27
Inlet
(ng/dfcm
@7%02)
0.469
4.75
0.796
3.18
0.354
0.371
0.354
2.90
2.21
2.12
4.07
21.6
4.07
79.0
3.80
3.62
40.3
2.56
2.65
1.77
0.486
11.98
4.77
0.751
0.0442
1.50
157
179
Outlet
(ng/dicm
@ 7% 02)
0.0929
0.894
0.139
0.790
0.0557
0.0639
0.0552
0.522
0.314
0.499
0.575
4.00
0.813
20.1
0.755
0.697
8.42
0.436
0.436
0.232
0.0987
2.22
0.697
0.174
0.116
0.232
35.4
39.4
Removal
Efficiency
(*)
80.2
81.2
825
75.2
84.2
82.8
84.4
82.0
85.8
76.5
85.9
81.5
80.0
74.6
80.1
80.8
79.1
83.0
83.6
86.9
79.7
81.4
85.4
76.8
-163
84.5
77.5
78.0
-------
TABLE 4-7, CONTINUED
Phase-
Condition
H-B9
II-B10
n-Bii
H-B12
n-B13
Run
Number
19
20
21
Avg
25
26
27
Avg
28
29
30
Avg
34
35
36
Avg
37
38
39
Avg
NA - Not available. Inlet PM
expedited mercury analysis of t
Inlet PM
(g/dscm @ 7% O2)
4.80
5.66
5.10
5.19
5.36
6.27
5.84
5.82
6.78
7.38
5.39
6.52
5.09
3.62
5.58
4.76
4.24
6.12
4.12
4.83
Outlet PM
(g/dscm @ 7% O2)
0.0012
0.0018
0.0039
0.0023
0.0016
0.0015
0.0138
0.0056
0.0016
0.0011
0.0039
0.0022
0.0025
0.0011
0.0018
0.0018
0.0021
0.142
0.0102
0.0089
Removal
Efficiency
(%)
99.97
99.97
99.92
99.96
99.97
99.98
99.76
99.90
99.98
99.99
99.93
99.96
99.95
99.97
99.97
99.96
99.95
99.77
99.75
99.82
levels were not determined during Phase I due to the need for
he front half fraction.
4-23
-------
TABLE 4-8, CONTINUED
ODgCDCT
DIOXINS
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1234678 HpCDD
Other HpCDD
OcU CDD
Total CDD
FURANS
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
234678 HxCDF
Other HxCDF
1234678 HpCDF
1234789 HpCDF
Other HpCDF
OctaCDF
Total CDF
Total CDD + CDF
londiUra Bi! Mum carbon Injection. i>0 P. Est inlet temperature)
Inlet
(ng/dicm
@ 7* O2)
0.621
6.45
1.24
6.69
1.15
1.15
0.956
10.1
13.4
12.4
48.7
103
3.44
85.4
5.54
5.35
61.7
6.69
ND
ND
1.91
36.3
24.8
5.26
8.12
29.6
274
377
Run 34
Outlet
(ni/dicn
@7%O2)
0.0308
0.164
0.0578
0.200
0.0452
0.0251
0.0213
0.197
0.170
0.157
0.484
1.55
0.163
4.48
0.213
0.182
2.74
0.0364
0.182
0.213
0.151
0.987
0.490
0.0816
0.308
0.370
10.6
12.2
Removal
Efficiency
<%)
95.0
97.5
95.4
97.0
96.1
97.8
97.8
98.1
98.7
98.7
99.0
98.5
95.3
94.8
96.1
96.6
95.6
99.5
0.0
0.0
92.1
97.3
98.0
98.4
96.2
98.7
96.1
96.8
Run 35
Inlet
(nc/dscn
@7%02)
0.562
6.32
1.49
8.03
1.15
1.38
1.07
11.3
10.7
9.98
20.7
72.6
4.25
93.3
5.97
5.39
69.0
7.11
7.11
6.08
2.18
37.2
20.7
4.13
11.9
27.5
302
374
Outlet
(ng/dicm
@7%02)
0.0349
0.153
0.0401
0.193
0.0388
0.0278
0.0168
0.201
0.226
0.0776
0.621
1.63
0.175
4.290
0.220
0.201
2.36
0.0343
0.162
0.194
0.142
0.956
0.388
0.0647
0.110
0.214
9.51
11.1
Removal
Efficiency
(*)
93.8
97.6
97.3
97.6
96.6
98.0
98.4
98.2
97.9
99.2
97.0
97.8
95.9
95.4
96.3
96.3
96.6
99.5
97.7
96.8
93.5
97.4
98.1
98.4
99.1
99.2
96.8
975
Run 36
Inlet
(ng/dscn
@ 7% O2)
0.439
4.81
0.879
5.04
0.563
0.697
0.611
5.77
5.83
5.63
17.2
47.5
2.77
66.0
3.63
3.34
40.8
ND
ND
ND
0.898
16.3
9.55
1.72
4.97
5.83
156
203
Outlet
(ng/dscm
@ 7% 02)
0.0528
0.159
0.0390
0.120
0.0244
0.0198
0.0139
0.140
0.112
0.106
0.317
1.10
0.152
3.35
0.192
0.145
1.78
0.126
0.126
0.0727
0.0225
0.645
0.264
0.0370
0.0621
0.126
7.09
8.20
Renoval
Efficiency
(*)
88.0
96.7
95.6
97.6
95.7
97.2
97.7
97.6
98.1
98.1
98.2
97.7
94 5
94.9
94.7
95.7
95.6
0.0
0.0
0.0
97.5
96.0
97.2
97.8
98.7
97.8
95.4
96.0
-------
TABLE 4-8. CONTINUED
Congener
DIOXINS
2378 TCDD
Other TCDD
12378 PCDD
Other PCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Other HxCDD
1 234678 HpCDD
Other HpCDD
Octa CDD
Total CDD
FURANS
2378 TCDF
Other TCDF
12378 PCDF
23478 PCDF
Other PCDF
1 23478 HxCDF
123678 HxCDF
123789 HxCDF
234678 HxCDF
Other HxCDF
1 234678 HpCDF
1234789 HpCDF
Other HpCDF
Octa CDF
Total CDF
Total CDD + CDF
Condition Bll (dry carbon injection, 270 F ESP inlet temperature)
Run 28
Inlet
(ng/dscm
@ 7% 02)
0.422
4.14
0.718
3.84
0.490
0.581
0.524
4.10
3.88
3.18
8.21
30.1
2.74
66.8
3.42
2.85
38.2
2.74
ND
ND
0.752
13.6
66.1
1.37
3.42
3.31
146
176
Outlet
(ng/dicm
@7%02)
ND
ND
0.0250
0.106
0.0225
0.0200
0.0140
0.130
0.100
0.0873
0.206
0.711
0.231
1.39
0.119
0.0873
0.917
0.0936
0.0873
0.0561
0.0200
0.429
0.168
0.0293
0.0830
0.0624
3.77
4.48
Removal
Efficiency
(*)
100.0
100.0
96.5
97.2
95.4
96.6
97.3
96.8
97.4
97.3
97.5
97.6
91.6
97.9
96.5
96.9
97.6
96.6
0.0
0.0
97.3
96.8
97.5
97.9
97.6
98.1
97.4
97.5
Run 29
Inlet
(ng/dicm
@7%02)
0.286
2.94
0.55
2.95
0.25
0.29
0.24
2.2
1.8
1.75
4.6
17.9
2.49
55.6
2.67
2.86
28.6
1.84
1.94
1.38
0.42
8.3
3.6
0.66
1.9
1.7
114
132
Outlet
(ng/dscn
@ 7% 02)
0.0169
0.102
0.0207
0.124
0.0157
0.0245
0.0163
0.182
0.144
0.1X5
0.307
1.08
0.0880
2.17
0.115
0.100
1.22
0.100
0.107
0.0690
0.0295
0.510
0.219
0.0476
0.122
0.119
5.03
6.10
Removal
Efficiency
(*)
94.1
96.5
96.3
95.8
93.7
91.4
93.2
91.6
92.2
92.8
93.3
94.0
96.5
96.1
95.3
96.5
95.7
94.6
94.5
95.0
92.9
93.8
93.9
92.8
93.7
92.8
95.6
95.4
Run 30
Inlet
(ng/dicm
@ 7% 02)
0.317
3.98
0.667
3.97
0.328
0.339
0.305
3.21
2.04
1.92
4.07
21.1
3.05
65.9
3.05
2.94
31.3
ND
ND
1.470
0.475
9.4
3.96
0.72
1.99
1.58
126
147
Outlet
(ng/dscm
@ 7% 02)
ND
0.090
0.0264
0.166
0.0162
0.0198
0.0126
0.149
0.138
0.126
0.336
1.08
0.108
2.17
0.138
0.120
1.30
0.038
0.084
0.1020
0.0780
0.537
0.228
0.0456
0.1160
0.120
5.19
6.27
Removal
Efficiency
(*)
1 00.0
97.7
96.0
95.8
95.1
94.2
95.9
95.4
93.2
93.4
91.8
94.9
96.5
96.7
95.5
95.9
95.8
0.0
0.0
93.1
83.6
94.3
94.2
93.7
94.2
92.4
95.9
95.7
-------
TABLE 4-9. FREQUENCY OF VOC DETECTED
IN TUBE PAIRS
CAMDEN COUNTY MWC (1992)
Number of Traps With Detectable Levels (Out of 24 Traps)
Compound
Bromomethane
Trichlorofluoromethane
1, 1-Dichloroethene
Carbon Bisulfide
Acetone
Methylene Chloride
Chloroform
1,1,1-Trichloroethane
Carbon Tetrachloride
Benzene
4-Methyl-2-Pentanone
Toluene
Tetrachloroethene
2-Hexanone
Chlorobenzene
m,p-Xylene
o-Xylene
Styrene
1, 1,2,2-Tetrachloroethane
Inlet
5-
8
0
24
1
16
2
0
1
24
0
2
0
2
17
18
4
1
0
Outlet
1
14
1
21
0
23
2
4
1
23
1
22
4
1
1
22
3
0
1
4-28
-------
4.9 Volatile Organic Compounds
Sampling for VOC was conducted for 19 target compounds during the 6 runs of
Conditions BIO and Bll. During both conditions, VOC were measured at the
economizer outlet and in the stack. During an individual run, 4 pairs of traps were
collected. Thus, for the 6 runs, there were a total of 24 inlet and 24 outlet pairs of traps.
As shown in Table 4-9, of the 19 target compounds, only 7 were detected in
greater than 40% of the sampling trap pairs. The remaining 12 compounds were
detected in less than 20% of the sampling trap pairs.
Table 4-10 presents the VOC inlet and outlet concentrations and removal
efficiency results for the 7 compounds that were detected in over 40% of the trap pairs.
Concentrations were calculated by summing the compound mass found in all 4 trap pairs,
dividing by the total metered volume, and correcting to 7% O2. If a compound was not
detected in a trap pair, the compound was assumed to be present at the detection limit.
It appears that three of the compounds (trichlorofluoromethane, methylene
chloride, and toluene) increased between the inlet and outlet, while three others (carbon
disulfide, benzene, and chlorobenzene) appear to be reduced. For the seventh
compound (m,p-xylene), the reported levels are higher at the outlet in three runs and
lower in the others. Except for benzene and carbon disulfide, however, the average
detected levels in the inlet samples were less than five times the practical quantitation
limit of the analytical method. As a result, the analytical data are subject to a relatively
high degree of uncertainty. Relative to the impact of carbon injection, the removal
efficiency for each of the compounds appears to be similar whether carbon is injected or
not.
4-27
-------
4.10 Flv Ash Carbon Content
Table 4-11 presents the carbon analysis of the Unit B fly ash samples. A single
composite sample was collected during each condition at the economizer outlet using an
EPA Method 5 sampling train. The condition results were between 1.20 and 2.36%
carbon on a dry basis.
4.11 Volumetric Flow and Moisture by EPA Methods 1 and 4
Unit B inlet and outlet gas flow rates and moisture contents were determined
using the procedures in EPA Methods 1 and 4, respectively. The results are presented in
Table 4-12. These values are based on measurements from the EPA multi-metals
sampling train. The plant CEMS were used to measure O2. Volumetric flow rates are
expressed in dry standard cubic meters per minute (dscmm) at measured O2
concentrations, and the moisture content is expressed in volume percent. Average inlet
results were 8 to 10% O2, 14 to 21% moisture, and flow rates of 2200 to 2700 dscmm.
Average outlet results were 10 to 13% O2, 16 to 21% moisture, and flow rates of 2800 to
3400 dscmm. During the test conditions conducted at a target SD outlet temperature of
270ฐF, (all but Conditions B6 and B7), the moisture gain across the SD averaged 2%.
During the two test conditions conducted at a target SD outlet temperature of 350ฐF, the
average moisture gain was negligible.
The inlet flue gas moisture content measured during Run 6 based on Method 4
calculations was 13.9%, which appears to be anomalously low compared to the inlet
moisture levels during other runs and to the measured outlet moisture level during the
same run. The value shown in Table 4-12 of 16.6% was estimated by subtracting 2.0%
from the outlet flue gas moisture content for the run (i.e., the average difference
between the inlet and outlet moisture contents during the other runs conducted at a
target SD exit temperature of 270ฐF). A similar adjustment of 2.0% was used to
estimate the flue gas moisture content at the stack during Runs 1 and 3. During Run 1,
the outlet impingers were not weighed prior to recovery. During Run 3, the original
4-30
-------
TABLE 4-10. VOLATILE ORGANIC COMPOUND RESULTS8
CAMDEN COUNTY MWC (1992)
Inlet Concentration (iif/dicm at 7% O2)
Compound
rrichlorofluororaethane
Carbon Disulfide
Methylene Chloride
lenzene
Toluene
Chlorobenzene
ra,p-Xylene
Condition BIO
Run 25
4.2
3.6
0.8
10.3
0.7
1.4
1.2
Run 26
1.5
5.7
1.0
7.0
0.6
0.9
1.8
Run 27
0.6
5.0
1.6
4.8
0.6
0.8
1.9
Condition Bll
Run 28
0.9
6.1
0.9
2.8
0.6
0.7
1.1
Run 29
0.6
6.2
0.8
9.5
0.6
1.3
3.3
Run 30
0.6
6.3
0.6
6.2
0.6
1.3
0.6
Outlet Concentration (ug/dscm at 7% O2)
Compound
rrichlorofluoromethane
Carbon Disulfide
Methylene Chloride
Jenzene
Toluene
Chlorobenzene
ra,p-Xylene
Compound
rrichlorofluoromethane
Carbon Disulfide
Methylene Chloride
Benzene
Toluene
Chlorobenzene
nt.p-Xylene
Condition 810
Run 25
7.7
1.6
2.0
5.0
1.2
0.6
1.6
Run 26
1.9
1.1
1.4
2.0
1.1
0.6
1.3
Run 27
0.7
1.0
1.5
2.1
1.6
0.6
1.6
Condition Bll
Run 28
1.8
1.4
1.2
2.7
1.7
0.6
3.3
Run 29
1.2
1.8
1.2
1.7
1.2.
0.7
1.7
Run 30
1.0
1.1
1.0
1.7
1.4
0.6
1.2
Removal Efficiency (%)
Condition BIO
Run 25
82.1
56.7
-162.0
51.0
-73.3
56.0
-30.5
Run 26
-27.1
80.7
-42.6
71.5
-75.0
32.2
28.3
Run 27
-22.5
79.3
10.6
56.8
-170.0
28.6
12.5
Condition Bll
Run 28
-103.4
77.7
-43.9
3.2
-175.0
11.1
-189.5
Run 29
-92.9
71.6
-47.2
81.6
-102.5
49.4
47.2
Run 30
-67.5
82.8
-77.5
73.0
-145.0
55.6
-105.0
'Method detection limit (MDL) for each of the detected compounds is 10 ng per tube pair, which equates to an
approximate flue gas concentration of 0.6 ^g/dscm at 7% O2. If the compound was not detected
during analysis of a trap pair, the compound was assumed to be present at the detection limit.
-------
TABLE 4-12. UNIT B VOLUMETRIC FLOW AND
MOISTURE RESULTS
CAMDEN COUNTY MWC (1992)
Phase-
Condition
I-B1
I-B2
I-B3
I-B4
I-B5
H-B6
D-B7
H-B8
Run
Number
1
2
3
Avg
4
5
6
Avg
7
8
9
Avg
10
11
12
Avg
13
14
15
Avg
10
11
12
Avg
13
14
1SR
Avg
16
17 .
18
Avg
Inlet
02
(%)
9.3
10.2
8.6
9.4
9.1
9.0
9.4
9.2
8.9
9.8
9.2
9.3
9.2
9.4
9.1
9.2
8.8
9.5
8.6
9.0
9.2
9.0
9.0
9.1
8.7
9.1
8.2
8.7
9.3
8.1
8.4
8.6
Flow Rate
(dscnun)
1467
1388
1346
1400
1447
1435
1485
1456
1364
1380
1468
1404
1410
1312
1432
1384
1322
1280
1178
1260
1215
1297
1410
1307
1309
1382
1313
1335
1340
1425
1144
1303
Stack
Moisture
(%)
16.8
17.4
18.4
17.5
16.8
16.9
16.6'
16.8
16.6
14.2
16^2
15.7
15.7
16.4
15.7
15.9
15.3
14.2
15.3
14.9
16.2
16.9
15.8
16.3
16.5
16.7
17.0
16.7
15.6
17.5
16.5
16.5
Outlet
o,
(%)
11.7
12.6
11.5
11.9
11.7
11.7
12.0
11.8
11.4
12.2
11.7
11.8
12.1
12.0
12.0
12.0
11.8
12.7
11.8
12.1
11.5
11.3
11.1
11.3
11.0
11.3
10.6
11.0
11.5
10.5
10.7
10.9
Flow Rate
(dscnun)
1747
1687
1692
1709
1793
1754
1797
1781
1698
1707
1767
1724
1840
1780
1805
1808
1671
1652
1457
1594
1781
1914
1954
1883
1857
1787
1721
1789
1715
1562
1435
1571
Stack
Moisture
(%)
18.8'
18.6
19.3'
18.9
18.8
19.1
18.6
18.8
18.4
17.7
19.1
18.4
17.1
17.6
17.7
17.4
16.5
15.8
16.6
16.3
16.5
17.0
14.8
16.1
16.3
16.7
16.1
16.3
18.6
19.6
18.2
18.8
4-32
-------
TABLE 4-11. UNIT B FLY ASH CARBON RESULTS
CAMDEN COUNTY MWC
Phase-Condition
I-B1
I-B2
I-B3
I-B4
I-B5
II-B6
II-B7
H-B8
H-B9
n-Bio
n-Bii
H-B12
H-B13
Carbon Content
(% by weight, dry basis)
1.41
1.45
1.16
1.82
1.86
1.56
1.53
1.89
1.69
1.56
2.20
1.20
1.16
4-31
-------
outlet moisture result of 11.8% was substantially below the level measured during the
other runs. During Run 25, the silica gel impinger used at the inlet sampling location
broke following successful final leak check of the train and absorbed water in the
impinger bucket. To estimate the actual moisture level, a silica gel weight gain of 8.3 g
was used, based on the average weight gain during other runs.
4.12 Continuous Emissions Monitoring Data
The CEM data are presented in Table 4-13. The CEM data includes spray dryer
inlet O2 and SO2, and stack NOX, HC1, CO2, H2O, O2, CO, and SO2. All concentrations
shown in the table are presented at actual O2 levels. However, the SO2 removal
efficiency was calculated after adjustment of the inlet and outlet concentrations to 7%
O2. Stack methane and THC were also recorded by the CEM system; however, neither
were found above a 0.1 ppm concentration level and are not reported.
Average unconnected inlet SO2 levels during each condition ranged between 45
and 93 ppm and average uncorrected outlet concentrations were between 9 and 17 ppm.
The average corrected SO2 removal were 64 to 83%. During the two test conditions with
no carbon injection, SO2 reductions were 64 to 65%. With carbon injection, the SO2
reductions were 67 to 82%. The higher SO2 removal efficiencies with carbon injection
may be the result of higher inlet SO2 levels rather than increased SO2 removals
associated with carbon injection. Outlet NOX levels during each test condition were
between 110 and 136 ppm, and average HC1 levels were less than 5 ppm during all test
conditions expect B12. The cause of the higher measured HC1 level (18 ppm) during
Condition B12 is unknown.
Average CO concentrations for each test condition ranged from 9 to 22 ppm.
Comparison of average CO concentrations with the fly ash carbon content data in
Table 4-11 does not indicate any significant relationship between these two parameters.
4-34
-------
TABLE 4-12, CONTINUED
Phase-
Condition
II-B9
n-Bio
H-B12
II-B13
Run
Number
19
20
21
Avg
25
26
27
Avg
28
29
30
Avg
34
35
36
Avg
37
38
39
Avg
Inlet
0:
(%)
8.6
8.2
8.7
8.5
9.6
9.2
8.2
9.0
8.9
8.8
8.1
8.6
8.9
8.7
8.7
8.8
8.7
8.7
8.6
8.7
Flow Rate
(dsanm)
1382
.1199
1353
1311
1275
1205
1350
1277
1376
1361
1318
1352
1449
1426
1399
1425
1334
1224
1230
1263
Stack
Moisture
.(*)
17.8
18.3
18.5
18.2
18.1k
17.6
18.9
18.2
17.2
17.8
17.6
17.5
15.9
15.2
16.1
15.7
16.7
16.0
16.1
16.2
Outlet
02
(%)
10.9
10.6
11.1
10.9
11.7
11.5
10.8
11.3
11.2
10.9
10.5
10.9
11.0
11.0
10.7
10.9
10.9
10.9
10.8
10.9
Flow Rate
(dsonun)
1531
1476
1417
1475
1478
1412
1534
1475
1643
15%
1511
1583
1641
1619
1525
1595
1558
1431
1518
1502
Stack
Moisture
(%)
19.6
19.7
20.0
19.7
20.5
20.8
21.3
20.9
20.4
20.4
21.6
20.8
18.3
17.6
18.1
18.0
18.5
17.1
17.6
17.1
Value calculated based on repotted moisture gains appeared eroneous. Value shown is calculated by adding
2.0% to the inlet or subtracting 2.0% from the outlet.
The silica gel impinger broke after run was completed. The weight gain by this impinger is estimated at 8.3 g
based on the average of the other runs.
4-33
-------
TABLE 4-13, CONTINUED
Phase-
Condition
n-Bio
U-Bll
U-B12
n-Bi3
Run
25
26
27
Avg
28
29
30
Avg
34
35
36
Avg
37
38
39
Avg
Inlet*
02
(%)
9.6
9.3
8.2
9.0
9.0
8.8
8.1
8.6
8.9
8.7
8.7
8.7
8.7
8.7
8.6
8.7
S02
(ppm)
45.0
57.1
66.1
56.1
52.8
94.6
76.3
74.6
84.9
59.3
83.7
76.0
82.6
60.1
54.9
65.9
Outlet1
NO,
(ppm)
127.1
126.2
137.5
130.2
139.7
132.4
136.4
136.2
133.4
125.1
137.3
131.9
131.6
121.5
121.1
124.8
HC1
(ppm)
4.6
10.5
8.9
8.0
6.5
10.1
4.4
7.0
26.2
13.5
14.4
18.0
4.6
2.2
2.0
2.9
C02
(%)
10.3
10.5
11.2
10.6
10.8
11.0
11.5
11.1
11.0
11.0
11.3
11.1
11.1
11.1
11.1
11.1
H2O
(%)
17.4
17.5
18.8
17.9
17.2
17.3
17.8
17.4
16.2
15.8
16.3
16.1
16.5
16.1
15.8
16.1
02
(%)
11.7
11.5
10.8
11.3
11.2
10.9
10.5
10.8
11.0
11.0
10.7
10.9
10.9
10.9
10.8
10.9
CO
(ppm)
17.3
12.3
8.6
12.7
7.9
12.7
7.1
9.2
13.2
11.4
6.0
10.2
9.0
18.0
13.7
13.6
SO2
(ppm)
15.9
14.7
17.0
15.9
16.2
20.3
16.7
17.7
18.1
15.3
7.3
16.9
16.9
15.1
14.5
15.5
SO,
Reduction'
(%)
56.4
68.1
67.8
64.1
62.4
74.1
73.1
69.9
74.0
68.2
75.3
72.5
75.0
69.3
67.9
70.7
*Concnetrations are reported at actual O2 levels, dry basis.
'Based on concentration corrected to 7% O2, dry basis.
4-36
-------
TABLE 4-13. UNIT B CEM RESULTS
CAMDEN COUNTY MWC (1992)
Phase-
Condition
I-B1
I-B2
I-B3
I-B4
II-B6
n-B?
TT-B8
D-B9
Run
1
2
3
Avg
4
5
6
Avg
7
8
9
Avg
10
11
12
Avg
13
14
15
Avg
10
11
12
Avg
13
14
15
Avg
16
17
18
Avg
19
20
21
Avg
Inlef
02
(*)
9.3
10.2
8.5
9.3
9.1
9.0
9.4
9.2
8.9
9.8
9.2
9.3
9.2
9.4
9.1
9.2
8.8
9.5
8.6
9.0
9.2
9.0
9.0
9.1
8.7
9.1
8.2
8.7
9.3
8.1
8.4
8.6
8.6
8.2
8.7
8.5
SOj
(ppm)
50.1
49.5
36.0
45.2
50.7
55.1
31.9
45.9
58.2
46.4
60.9
55.2
110.2
70.3
42.6
74.4
64.1
89.8
41.7
65.2
51.8
75.4
151.6
92.9
53.1
53.4
46.8
51.1
60.3
62.0
68.9
- 63.7
47.2
61.6
48.9
52.5
Outlet*
NO,
(ppm)
125.1
126.9
122.6
124.9
129.6
132.5
124.4
128.9
132.3
123.4
128.2
128.0
122.4
124.0
132.1
126.2
109.7
111.6
109.0
110.1
131.8
126.6
129.6
129.3
122.8
126.3
124.0
124.4
118.4
135.6
121.1
125.1
130.7
130.3
124.0
128.4
HCI
(ppm)
1.4
1.2
0.6
1.0
1.6
1.7
0.8
1.4
1.1
2.0
2.8
1.9
2.5
5.1
2.5
3.4
1.0
3.3
0.7
1.7
3.0
4.7
15.5
7.7
9.0
4.8
4.1
5.9
2.5
2.0
1.3
1.9
0.3
0.4
0.0
0.2
C02
(%)
10.3
9.4
10.5
10.1
10.3
10.3
10.0
10.2
10.6
9.7
10.3
10.2
9.9
10.0
10.0
10.0
10.1
9.3
10.2
9.9
10.4
10.7
10.8
10.6
10.9
10.7
11.4
11.0
10.5
11.4
11.2
11.0
11.1
11.4
10.9
11.1
H2O
(%)
17.2
16.0
17.7
17.0
16.7
16.5
16.1
16.4
16.2
15.3
15.9
15.8
14.7
15.0
14.9
14.9
14.4
13.9
14.9
14.4
14.4
14.9
13.9
14.4
14.5
14.7
15.0
14.7
15.9
17.6
16.2
16.6
17.3
.17.3
18.0
17.5
02
(%)
11.7
12.6
11.5
11.9
11.7
11.7
12.0
11.8
11.4
12.2
11.7
11.8
12.1
12.0
12.0
12.0
11.8
12.7
11.8
12.1
11.5
11.3
11.1
11.3
11.0
11.3
10.6
11.0
11.5
10.5
10.7
10.9
10.9
10.6
11.1
10.9
CO
(ppm)
8.1
43.5
11.5
21.0
12.8
10.5
12.7
12.0
9.0
15.7
12.0
12.2
12.4
19.6
11.2
14.4
26.4
19.5
16.3
20.7
9.7
11.4
7.2
9.4
12.1
15.6
13.5
13.7
20.8
8.5
9.7
13.0
12.8
11.9
42.0
22.2
S02
(ppm)
12.0
. 11.1
12.4
11.8
11.1
12.3
10.2
11.2
11.5
11.2
13.4
12.0
12.7
10.6
6.0
9.8
8.6
12.2
5.0
8.6
9.4
11.8
30.8
17.3
11.1
9.1
10.4
10.2
11.0
13.8
11.6
12.1
12.5
12.1
12.8
12.4
sn
au2
Reduction11
(%)
69.8
71.1
54.8
65.2
71.9
71.2
58.8
67.3
74.9
68.8
72.0
71.9
84.7
80.5
81.4
82.2
82.1
81.2
83.9
82.4
77.3
80.7
75.2
77.7
74.2
79.0
72.7
75.3
77.6
72.6
79.2
76.5
67.4
76.0
67.6
70.3
4-35
-------
TABLE 5-1 UNIT A CARBON FEED SYSTEM DATA
CAMDEN COUNTY MWC (1992)
Condition
Al
A2
A3
A4
A5
Date
5/29/92
5/30/92
6/1/92
6/6/92
6/10/92
Carbon Type
None
FGD
FGD
FGD
FGD
Carbon Feed
Method
None
Slurry8
Slurry"
Slurry"
Slurry8
Run
1
2
3
Carbon Feed
Rate
(Ib/hr)
0
0
0
Average 0
4
5
6
49.0
51.1
51.1
Average 50.4
7
8
9
62.3 .
56.4
56.4
Average 58.4
22
23
51.0
51.0
Average 51.0
31
32
33
42.7
42.7
42.7
Average 42.7
"Fed manually from 50 Ib bags into the lime slurry feed tank during lime slaking periods.
5-2
-------
5.0 ELECTROSTATIC PRECIPITATOR PERFORMANCE TESTING
Testing to evaluate the long-term impact of carbon injection on ESP performance
was conducted over a 13-day period on Unit A These tests included one day of baseline
testing without carbon injection (Condition Al), three days of testing with four ESP
fields in service (Conditions A2, A3, and A4), and one day of testing with three ESP
fields in service (Condition A5). Carbon was continuously added to the SD lime slurry
feed tank during slaking from Day 2 through Day 13. Three test runs were conducted
during Conditions Al, A2, A3, and A5. During Condition A4, only two runs were
completed due to operating problems that precluded a third run.
5.1 Carbon Feed System Data
Table 5-1 summarizes key data for the carbon feed system used during testing on
Unit A These data include type of carbon fed, the carbon feed method (i.e., slurry or
dry), and carbon feed rates.
All carbon injected into Unit A was mixed with the lime slurry. Carbon was fed
manually from 50 Ib bags into the lime slurry feed tank that was dedicated to the Unit A
SD throughout the testing. The carbon was fed only during normal slaking periods,
which usually lasted less than an hour and occurred once every four to five hours.
Slaking (and thus carbon addition to the slurry) was not done during a sampling run so
as not to interfere with or bias a run. The objective was to operate with a constant
carbon injection rate between Day 2 (Condition A2) and Day 14 (Condition A5). As
indicated in Table 5-1, however, average injection rates during each condition ranged
from 42.7 to 58.4 Ib/hr.
Records were kept of the amounts and times at which carbon was added so that
carbon injection rates could be monitored. Table 5-2 gives the complete record of
carbon added to the lime slurry feed tank during the 13 days of carbon addition. Carbon
feed rates were calculated for the time between each slaking period by dividing the sum
5-1
-------
TABLE 5-2, CONTINUED
Slake
Number
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
Date
6/4/92
6/4/92
6/4/92
6/5/92
6/5/92
6/5/92
6/5/92
6/5/92
6/6/92
6/6/92
6/6/92
6/6/92
6/7/92
6/7/92
6/7/92
6/7/92
6/7/92
6/8/92
6/8/92
6/8/92
6/68/92
6/8/92
6/9/92
6/9/92
6/9/92
6/9/92
6/9/92
6/10/92
6/10/92
6/10/92
Slake
Start Time
1230
1715
2223
0331
0840
1333
1824
2322
0415
0929
1421
2015
0128
0654
1152
1644
2139
0228
0711
1152
1646
2145
0237
0723
1206
1655
2157
0244
0733
1334
Slake
Stop Time
1320
1805
2323
0431
0943
1430
1925
0022
0515
1024
1518
2135
0229
0756
1251
1745
2228
0318
0803
1250
1742
2235
0326
0811
1258
1749
2248
0344
0840
1409
Carbon
Added
(Ib)
261
256
248
250
251
247
248
248
251
250
248
300
255
262
257
265
258
265
257
256
258
258
256
255
254
256
251
256
250
257
Time
Between
Starts
(mm)
284
285
308
308
309
293
291
298
293
314
292
354
313
326
298
292
295
289
283
281
294
299
292
286
283
289
302
287
289
361
Average
Carbon
Feed Rate
(Ib/hr)
55.1
53.9
483
48.7
48.7
50.6
51.1.
49.9
51.4
47.8
51.0
50.8
48.9
482
51.7
54.5
515
55.0
54.5
54.7
52.7
51.8
52.6
53.5
53.9
53.1
49.9
53.5
51.9
42.7
5-4
-------
TABLE 5-2 UNIT A LONG-TERM CARBON FEED DATA
CAMDEN COUNTY MWC
Slake
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Date
5/29/92
5/30/92
5/30/92
5/30/92
5/30/92
5/30/92
5/31/92
5/31/92
5/31/92
5/31/92
5/31/92
6/1/92
6/1/92
6/1/92
6/1/92
6/1/92
6/2/92
6/2/92
6/2/92
6/2/92
6/2/92
6/3/92
6/3/92
6/3/92
6/3/92
6/3/92
6/4/92
6/4/92
Slake
Start Time
2235
0307
0730
1236
1832
2320
0400
0855
1353
1731
2130
0205
0656
1200
1718
2205
0255
0754
1247
1739
2236
0316
0740
132Q
1710
2209
0300
0746
Slake
Stop Time
2323
0355
0810
1326
1926
2359
0455
0940
1439
1807
2220
0305
0755
1300
1820
2305
0355
0856
1343
1839
2336
0416
0840
1320
1810
2309
0400
0846
Carbon
Added
(Ib)
200
203
197
250
303
203
258
259
255
202
255
306
302
308
299
253
298
301
248
250
248
253
250
259
259
259
257
256
Time
Between
Starts
(mm)
--
272
263
306
365
288
280
295
298
218
239
275
291
304
318
287
290
299
293
292
297
280
264
280
290
299
291
286
Average
Carbon
Feed Rate
(Ib/hr)
44.8
44.9
49.0
51.1
423
553
52.7
513
55.6
64.0
66.8
623
60.8
56.4
52.9
61.7
60.4
50.8
51.4
50.1
542
56.8
55.5
53.6
52.0
53.0
53.7
5-3
-------
TABLE 5-3. UNIT A COMBUSTOR
OPERATING DATA
CAMDEN COUNTY MWC (1992)
Condition
Al
A2
A3
A4
AS
Run
I
2
3
Average
4
5
6
Average
7
8
9
Average
22
23
Average
31
32
33
Average
Boiler
Steam
Flow
(lb x KP/hr)
100.0
101.3
95.9
99.0
98.1
91.3
94.6
94.7
89.6
102.7
102.9
98.4
72.1
84.1
79.6
93.3
94.4
99.3
95.7
Furnace
Temperature
(ฐF)
1157
1165
1155
1159
1145
1128
1110
1128
1130
1167
1190
1162
1052
1113
1082
1126
1138
1157
1140
Economizer
Outlet
Temperature
(ฐF)
482
465
479
475
479
480
494
484
441
469
501
470
470
474
472
485
494
495
492
5-6
-------
of the carbon added during each slaking cycle by the time elapsed between slaking
cycles.
5.2 Combustor Operating Data
Key combustor operating data for each test run are presented in Table 5-3.
Included are boiler steam flow, furnace temperature, and flue gas temperature at the
economizer outlet. For each condition, run averages and the condition averages are
shown. All of these data were collected from plant instruments.
As shown in Table 5-3, the boiler steam flow averages for Conditions Al, A2, A3,
and A5 ranged from 95,600 to 99,000 Ib/hr. Because of problems maintaining the
desired combustor operating conditions due to wet refuse, the steam production rate
during Condition A4 averaged 80,000 Ib/hr. The furnace temperature condition
averages ranged from 1128 to 1162ฐF during Conditions Al, A2, A3, and A5, but
decreased to an average of 1085ฐF during Condition A4, again due to wet refuse fed
during this test. The average of the flue gas temperature at the economizer outlet
ranged from 470 to 492ฐF and did not vary significantly between test conditions.
5.3 Spray Dryer Absorber/Electrostatic Precipitator Operating Data
Operating data for the SD and ESP are presented in Table 5-4. These data
include lime slurry flow rate, SD and ESP outlet temperatures, ESP secondary voltage,
secondary current to each ESP field, and the stack flue gas opacity. Dilution water flow
and ESP field spark rate data were also collected, but are not summarized here. For
each condition, run and condition averages are shown. Plant instruments collected all
the data, with the exception of the ESP outlet temperature which was measured by
Radian. The fourth ESP field was not in operation during Condition A5.
5-5
-------
5.4 Mercury
Table 5-5 presents the Hg test results. The average inlet Hg concentrations
during each condition ranged from 331 to 729 ^g/dscm. The distribution of Hg in the
three sampling train fractions for each condition varied from 28 to 71% in the filter, 27
to 67% in the HNO3/H2O2 impingers, and less than 5% in the KMnO4/H2SO4.
The outlet Hg levels during each condition averaged 48 to 131 /zg/dscm when
carbon was injected and 245 //g/dscm when carbon was not injected. Variations in outlet
Hg concentrations are consistent with changes in inlet Hg concentrations, carbon
injection rates, and fly ash carbon content (see Table 3-1). With the exception of
Run 31, the filter fraction contained less than 9% of the total Hg collected by the
sampling train. The average Hg percentages in the HNO3/H2SO2 impingers ranged from
47 to 69% with carbon injection and was 87% when carbon was not injected. The Hg
percentages in the KMnO4/H2SO4 impingers ranged from 24 to 52% with carbon
injection and was 10% when carbon was not injected.
The average Hg reduction was 45% with no carbon injection during Condition Al
and ranged from 81 to 91% during Conditions A2 through A5 when carbon was injected.
5.5 Cadmium and Lead
Concentrations of Cd and Pb during these tests are shown in Table 5-6. For Cd,
average removal efficiencies across the SD/ESP were 99.80% during Condition Al, 99.74
to 99.88% during Conditions A2 through A4, and 99.62% during Condition A5. For Pb,
average removal efficiencies were 99.93% for Condition Al, 99.78 to 99.98% for
Conditions A2 through A4, and 99.40% for Condition A5.
Of the total Cd and Pb concentrations measured at the inlet sampling location,
over 99.8% were associated with the front-half fraction, except during Runs.2, 4, and 8.
During these three runs, the back-half accounted for 3 to 7% of the total catch for both
5-8
-------
TABLE 5-4. UNIT A SPRAY DRYER ABSORBER/ESP OPERATING DATA
CAMDEN COUNTY MWC (1992)
Run
Lime Slurry
Flow Rate
(gpm)
Phase n, Condition Al
. 1
2
3
Average
9.1
9.6
9.0
9.2
SD Outlet
Temp
(F)
ESP Outlet
Temp
(F)
ESP Voltage
(KV)
277
270
273
274
284
277
277
280
46
47
47
47
ESP
TR1-1
Current
(mA)
288
280
260
276
ESP
TR1-2
Current
(mA)
ESP
TR1-3
Current
(mA)
ESP
TR1-4
Current
(mA)
441
446
443
443
456
454
451
454
450
455
451
452
Opacity
(%)
0.0
0.0
0.1
0.0
Phase n, Condition A2
4
5
6
Average
9.2
9.0
8.2
8.8
265
265
266
265
273
269
272
272
47
46
47
47
270
270
288
276
448
448
448
448
456
456
454
455
456
. 456
456
456
0.0
0.0
0.0
0.0
Phase n, Condition A3
7
8
9
Average
9.1
9.1
9.0
9.1
274
262
281
272
278
269
288
278
47
46
46
46
281
282
220
261
448
448
448
448
452
451
449
451
453
450
448
450
0.0
0.0
0.0
0.0
Phase n, Condition A4
22
23
Average
9.1
9.1
9.1
280
278
279
285
283
284
46
46
46
275
288
' 282
441
444
442
451
452
452
454
456
455
0.0
0.0
0.0
Phase n, Condition AS
31
32
33
Average
OOS = Out (
9.1
9.1
9.1
9.1
Df Service.
276
276
275
276
283
283
284
283
47
47
47
47
284
272
285
280
448
448
448
448
454
451
455
453
oosa
OOS
OOS
OOS
0.0
0.0
0.0
0.0
-------
TABLE 5-6. UNIT A CADMIUM AND LEAD RESULTS*
CAMDEN COUNTY MWC (1992)
Condition
Al
Al
Al
Al
A2
A2
A2
A2
A3
A3
A3
A3
A4
A4
A4
A5
A5
A5
A5
Run
1
2 .
3
AVG
4
5
6
AVG
7
8
9
AVG
22
23
AVG
31
32
33
AVG
Cadmium (ug/dscm at 7% O2)
Inlet
Front
Half
842
1669
1350
1287
993
1114
1004
1037
1218
747
2099
1355
1860
12%
1578
1142
1084
2150
1459
Estimated Detection Limit! 10.1
Back
Half
1.28
51.4
0.46
17.7
33.1
1.11
1.40
11.9
0.45
50.9
0.48
17.3
0.60
0.39
0.50
0.22
ND
0.29
0.17
Total
843
1720
1350
1304
1026
1115
1005
1049
1218
798
2100
1372
1861
1296
1578
1142
1084
2151
1459
Outlet
Front
Half
1.46
2.23
2.50
2.07
1.47
2.00
2.41
1.%
3.56
1.66
2.33
2.52
2.35
1.00
1.67
6.55
0.59
8.87
5.34
0.201 10.31 0.72
Back
Half
0.42
0.39
0.47
0.43
0.25
0.66
1.20
0.71
0.48
0.46
0.35
0.43
0.22
0.20
0.21
0.43
0.40
0.60
0.47
0.18
Total
1.88
2.62
2.97
2.49
1.72
2.66
3.61
2.66
4.04
2.12
2.68
2.95
2.56
1.21
1.88
6.98
0.99
9.46
5.81
Removal
Efficiency
(%)
99.78
99.85
99.78
99.81
99.83
99.76
99.64
99.75
99.67
99.73
99.87
99.79
99.86
99.91
99.88
99.39
99.91
99.56
99.60
Lead (ug/dscm at 7% O2)
Inlet
Front
Half
19637
20857
23007
21167
13820
23453
17565
18279
21814
10523
32446
21594
32679
36711
34695
16129
17921
15835
16628
0.901 NA| 101
Back
Half
6.17
626
2.15
211
345
14.5
3.23
121
3.64
849
3.44
285
ND
5.18
2.59
2.36
1.75
3.71
2.61
0.61
Total
19643
21482
23009
21378
14165
23468
17569
18401
21818
11371
32450
21880
32679
36716
34697
16131
17923
15839
16631
Outlet
Front
Half
13.3
12.1
14.1
13.2
14.5
27.7
28.3
23.5
80.5
20.8
27.6
43.0
9.00
5.48
7.24
103
84.7
109
98.8
Back
Half
1.01
1.20
1.77
1.32
0.97
2.15
1.70
1.61
1.35
1.11
0.89
1.11
ND
ND
0.00
1.46
ND
ND
0.49
Total
14.3
13.3
15.9
14.5
15.5
29.9
30.0
25.1
81.8
21.9
28.5
44.1
9.0
5.48
7.24
104
84.7
109
99.2
Removal
Efficiency
(%)
99.93
99.94
99.93
99.93
99.89
99.87
99.83
99.86
99.62
99.81
99.91
99.80
99.97
99.99
99.98
99.35
99.53
99.31
99.40
1021 2.691 0.541 3.221 NA
I
H-t
o
*ND - Nol Detected.
-------
TABLE 5-5. UNIT A MERCURY RESULTS
CAMDEN COUNTY MWC (1992)
Condition
Al
A2
A3
A4
A5
Run
1
2
3
AVG
4
5
6
AVG
7
8
9
AVG
22
23 b
AVG
31
32
33
AVG
Mercury Concentrations (ug/dscm at ?/ O2)
Inlet
Filter &
Probe
Rinse
168
83
353
202
158
90
341
196
236
139
401
259
189
302
245
217
206
279
234
HNO3/
H2O2
Impingers
97
320
245
221
128
235
1058
473
291
289
286
289
452
497
475
116
69
82
89
KMnO4/
H2SO4
Impingers
3.4
26.4
11.7
13.8
15.8
78.2
14.2
36.1
2.7
30.5
2.5
11.9
1.8
16.6
9.2
1.3
19.9
2.5
7.9
Total
268
430
610
436
302
403
1412
706
530
458
690
559
643
816
729
335
294
364
331
Outlet
Filter &
Probe
Rinse
0.6
10.6
10.3
7.2
0.5
0.6
5.1
2.1
ND .
1.2
ND
0.4
0.3
2.2
1.3
8.3
4.5
3.0
5.3
HNO3/
H2O2
Impingers
107
251
280
213
36.7
36.9
198
90.6
27.9
37.4
80.0
48.4
25.4
47.0
36.2
26.8
32.5
34.2
31.1
KMnO4/
H2SO4
Impingers
14.2
27.9
32.4
24.8
17.6
40.0
58.1
38.6
14.9
69.3
75.6
53.3
23.5
40.7
32.1
5.06
14.1
15.2
11.5
Total
121
290
322
245
54.8
77.6
261
131
42.7
108
156
102
49.2
89.9
69.5
40.2
51.1
52.4
47.9
Removal
Efficiency
(%)
54.8
32.5
47.2
44.8
81.9
80.7
81.5
81.4
91.9
76.4
77.4
81.9
92.3
89.0
90.7
88.0
82.6
85.6
85.4
(J\
ND = Not Detected.
bRun 24 was terminated due to plant operating problems.
-------
TABLE 5-7. UNIT A PARHCULATE MATTER RESULTS
CAMDEN COUNTY MWC (1992)
Condition
Al
A2
A3
A4
A5
Run
1
2
3
AVG
4
5
6
AVG
7
8
9
AVG
22
23
AVG
31
32
33
AVG
INLET
PM
(g/dscm
@ 7% O2)
5.11
4.42
10.2
6.57
5.78
8.62
7.57
732
7.49
236
103
6.73
8.81
7.86
833
736
8.19
7.04
7.53
OUTLET
PM
(g/dscm
@ 7% O2)
0.0034
0.0035
0.0043
0.0037
0.0019
0.0038
0.0035
0.0031
0.0059
0.0075
0.0048
0.0061
0.0057
0.0016
0.0036
0.0078
0.0051
0.0034
0.0054
Removal
Efficiency
(%)
99.93
99.92
99.96
99.94
99.97
99.96
99.95
99.96
99.92
99.68
99.95
99.85
99.94
99.98
99.96
99.89 . .
99.94
99.95
99.93
5-12
-------
metals and may have been caused by penetration of paniculate through or around the
filter. The total metal concentrations for each test condition at the inlet were relatively
consistent, ranging from 1,049 /ig/dscm to 1,578 ;zg/dscm for Cd, and 16,628 to
34,695 /
-------
5.8 Particle Size Distribution
Two sets of PSD samples were collected using an 8-stage Andersen impactor
during Conditions Al, A3, A4, and A5. One set of samples was collected during
Condition A2. Both of the PSD trains operated during Condition A4 experienced
operating problems: a loose impinger connection was discovered on one of the trains at
the end of the run, and the other had problems with operation of the sampling pump.
Post-test review of the collected data from the first train indicated that the flue gas
moisture content was lower than for other trains and that the isokinetic flow rate was
high. As a result, the samples collected by the first train were rejected. Post-test review
of data from the second train indicated that all QA/QC criteria were met. Therefore,
the data from this train were accepted.
Selected data from each of the accepted PSD trains are presented in Table 5-9.
The data include the start and stop times for each sampling period; the cumulative mass
fraction collected following the second, fourth, sixth, and eighth impactor stages; and the
total PM loading. As indicated by these data, the particle size distribution samples
collected during Conditions Al through A4 are generally consistent with each other, with
the exception of Run PSD-5. The cause of this difference during Run PSD-5 is
unknown. The total PM loading from all six trains run during Conditions Al through
A4 are similar. During Condition A5, with the fourth ESP field out of service, there was
an increase in the fraction of PM less than 9 ftm in diameter and in the total quantity of
PM collected. Comparison of the average measurements from Conditions Al through
A4 versus Condition A5 indicates that the emissions of PM greater than 9 ^m were
approximately 0.0007 g/dscm during all of the tests. However, emissions of PM less than
9 pm during Condition A5 were approximately 0.002 g/dscm compared to 0.0005 g/dscm
during Conditions Al through A4.
5-14
-------
TABLE 5-8. UNIT A FLY ASH CARBON RESULTS
CAMDEN COUNTY (MWC) 1992
Test Condition
Al
A2
A3
A4
A5
Carbon Content
(% by weight, dry basis)
1.95
1.42
2.36
2.25
1.52
5-13
-------
5.9 Volumetric Flow and Moisture bv EPA Methods 1 and 4
Unit A inlet and outlet gas flow rates and moisture contents were determined
using the procedures in EPA Methods 1 and 4, respectively. The results are presented in
Table 5-10. The values were measured using the EPA multi-metals sampling train. The
flow rates are expressed in dscmm at actual O2 levels, and the moisture content is
expressed in percent by volume. Average inlet results were 8 to 11% O2, 14 to 17%
moisture, and flow rates of 1236 to 1490 dscmm. Average outlet results were 11 to 13%
O2, 17 to 19% moisture, and flow rates of 1550 to 1624 dscmm.
During Conditions A4 and A5, the inlet O2 monitor appeared to be reporting high
results (11 to 13% O2). Review of the plant calibration data for both of these days and
discussions with plant personnel indicated that these readings were potentially erroneous.
The values shown for inlet O2 concentrations for both of the conditions were calculated
by subtracting 2.4% from the outlet O2 reading for the same run. This adjustment factor
was based on the average difference in O2 (caused by air infiltration to the SD/ESP)
between the inlet and outlet sampling locations during the other test conditions.
5.10 Continuous Emission Monitoring Data
The CEM data are presented in Table 5-11. The CEM data include spray dryer
absorber inlet O2 and SO2 concentrations, and stack outlet NO,,, HC1, CO2, H2O, O2, and
SO2 concentrations. All concentrations are presented at actual O2 levels. However, the
SO2 removal efficiency was calculated after normalizing the inlet and outlet SO2
concentrations to 7% O2. Outlet methane and THC were also recorded by the CEM
system; however, neither were found above a 0.1 ppm concentration level and are not
reported.
5-16
-------
TABLE 5-9 UNIT A PARTICLE SIZE DISTRIBUTION DATA
CAMDEN COUNTY MWC
Condition
(Date)
Al
(5/29)
A2
(5/30)
A3
(6/1)
A4
(6/6)
A5
(6/10)
Run No.
PSD-1
PSD-2
Sampling
Period
(Start-Stop
Time)
13:50-16:50
18:50-21:50
Average
PSD-3
10:40-18:40
Average
PSD-5
PSD-6
10:45-19:45
15:00-20:00
Average
PSD-22B
15:30-22:10
Average
PSD-30A
PSD-30B
10:20-18:20
10:15-18:15
Average
Cumulative Mass Fraction Less Than Indicated
Particle Size (fan)*
9.0
0.398
0.398
0398
0.364
0364
0.741
0.269
0.505
0.256
0.256
0.760
0.768
0.764
4.0
0306
0343
0325
0.236
0.236
0.589
0.195
0392
0.205
0.205
0.548
0.638
0.593
13
0.213
0.230
0222
0.145
0.145
0.267
0.104
0.186
0.185
0.185
0.350
0376
0363
0.5
0.111
0.047
0.079
0.036
0.036
0.100
0.034
0.067
0.140
0.140
0.139
0.205
0.172
PM Loading
(g/dscm)
0.00120
0.00193
0.00157
0.00071
0.00071
0.00128
0.00102
0.00115
0.00122
0.00122
0.00138
0.00399
0.00269
8 Theoretical particle cut sizes vary with sample collection rate for individual train, sizes shown are approximate.
-------
TABLE 5-11. UNIT A CEN RESULTS
CAMDEN COUNTY MWC (1992)
Phase-
Condition
II-A1
II-A2
II-A3
II-A4
II-A5
Run
1
2
3
Avg
4
5
6
Avg
7
8
9
Avg
22
12
Avg
13
14
15
Avg
Inlet1
(%)
9.2
8.1
9.1
8.8
8.8
9.7
9.9
9.5
10.1
7.9
10.7
9.6
11.3
10.1
10.7
9.8
9.8
9.0
9.5
S02
(ppm)
100.9
130.6
99.7
110.4
95.8
68.4
105.2
89.8
55.2
114.6
49.7
73.2
29.2
42.8
36.0
32.0
78.1
64.4
58.2
Outlet*
NO,
(ppm)
156.7
160.0
147.5
154.8
161.4
143.4
141.8
148.9
146.0
130.1
126.8
134.3
131.7
141.2
136.4
134.6
146.3
154.1
145.0
HO
(ppm)
7.3
7.7
9.5
8.2
5.3
1.3
3.8
3.5
0.7
0.9
1.0
0:9
0.0
0.0
0.0
0.9
3.0
1.9
1.9
C02
10.4
11.2
10.4
10.7
10.7
9.8
9.8
10.1
9.9
11.6
9.8
10.5
8.3
9.5
8.9
9.8
9.7
10.6
10.0
HjO
15.5
15.2
15.9
15.5
17.4
15.8
17.5
16.9
15.4
16.6
15.1
15.7
15.3
17.1
16.2
16.1
16.1
17.9
16.7
ฐ2
11.6
10.7
11.5
11.3
11.3
12.2
12.2
11.9
12.0
10.3
12.1
11.5
13.7
12.5
13.1
12.2
12.2
11.4
11.9
CO
(ppm)
9.4
5.1
15.0
9.8
8.8
15.2
7.6
10.5
10.5
5.2
14.0
9.9
22.6
11.9
17.3
12.5
15.2
13.3
13.7
S02
(ppm)
7.8
10.8
10.9
9.8
3.8
0.7
6.9
3.8
1.8
4.3
2.3
2.8
1.4
0.8
1.1
0.2
5.1
2.7
2.7
Reduction1*
90.3
89.6
86.2
88.7
95.0
98.7
91.7
95.1
96.0
95.4
94.6
95.4
93.6
97.6
95.6
99.2
91.6
94.8
95.2
V1
>-ป
00
Concentrations are reported at actual O2 levels, dry basis.
bBased on concentration corrected to 7% O2, dry basis.
-------
TABLE 5-10. UNIT A VOLUMETRIC FLOW AND MOISTURE RESULTS
CAMDEN COUNTY MWC (1992)
Condition
Al
A2
A3
A4
AS
RUN
1
2
3
AVG
4
5
6
AVG
7
8
9
AVG
22
23
AVG
31
32
33
AVG
Inlet
02
(%)
9.2
8.1
9.1
8.8
8.8
9.7
9.9
9.5
10.1
7.9
10.7
9.6
113'
10.1'
10.7
9.8'
9.8 '
9.0"
9.5
Flow Rate
(dscmm)
1329
1197
1183
1236
1236
1388
1424
1350
1510
1282
1507
1433
1263
1272
1267
1504
1446
1518
1490
Moisture
(%)
14.0
14.5
15.0
14.5
15.5
14.7
17.1
15.8
16.5
15.2
14.6
15.4
15.2
17.7
16.5
16.3
15.4
17.2
163
Outlet
02
(%)
11.6
10.7
11.5
11.3
11.3
12.2
12.2
11.9
12.0
10.3
12.1
11.5
13.7
12.5
13.1
12.2
12.2
11.4
11.9
Flow Rate
(dscmm)
1587
1518
1545
1550
1511
1647
1695
1618
1670
1474
1729
- 1624
1521
1550
1535
1526
1663
1614
1601
Moisture
(%)
16.8
17.0
18.1
17.3
19.7
16.3
20.1
18.7
17.6
19.4
17.5
18.1
17.5
18.2
17.8
19.1
18.5
20.2
19.3
Oxygen data from inlet CEM appeared to be
Inlet O2 levels were calculated by subtracting
erroneous for Conditions A4 and A5.
2.4% from outlet O2 levels.
5-17
-------
Condition average inlet SO2 concentrations were between 36 and 110 ppm, and
average outlet concentrations were between 1 and 10 ppm. The average SO2 removal
was 89% with no carbon injection and 96% with carbon injection. Outlet NOX
concentrations ranged between 133 and 155 ppm. Outlet HC1 concentrations averaged
8 ppm without carbon injection and 2 ppm with carbon injection.
The average outlet SO2 concentrations measured from Unit A (1 to 10 ppm) are
lower than those measured from Unit B (8 to 17 ppm). This difference raised questions
regarding a potential error in one of the CEMs. However, based on review of SD/ESP
and CEM performance data for both units with plant personnel, it was concluded that
the measured SO2 levels from both units were correct.
5-19
-------
Glass
Temperature
Sensor /
Temperature Sensor
Temperature Sensor
Reverse-Type
Pilot Tube
7s
to
Implngers with Absorbing Solutl
Wall Glass Probe
Uner
Probe Tip / /
Glass Filter Holder
Heated Area
Ice Bath
Optional Empty Knockout
(Not Used) 5%HN03/10%H202 Em 4% KMnq/10% H^SO, Silica Ge.
(Not Used) Temperature cmpiy
Atmosphere
Manometer
Vacuum
Une
Figure 6-1. Schematic of Multiple Metals Sampling Train
-------
6.0 FLUE GAS SAMPLING AND ANALYTICAL PROCEDURES
This section describes the flue gas sampling and analytical procedures used during
the testing at the Camden County Resource Recovery Facility (CCRRF).
6.1 Paniculate Matter and Multiple Metals
The EPA multi-metals method was used to determine concentrations of PM and
Hg in flue gas during all tests.9 The same method was also used to determine the
concentration of other selected metals (cadmium, lead, antimony, arsenic, barium,
beryllium, boron, chromium, cobalt, copper, manganese, molybdenum, nickel, selenium,
silver, thallium, and vanadium) during several test conditions. Sampling was conducted
simultaneously at the economizer outlet and in the stack. At the economizer outlet, a
24-point sampling matrix was used. For the stack location, a 12-point sampling matrix
was used. During the three test conditions during which CDD/CDF were also sampled,
the sampling duration was two hours. During all other test conditions, the runs were one
hour in duration.
6.1.1 Sampling Equipment Preparation
The multiple metals sampling train is shown in Figure 6-1. The train consists of a
glass nozzle and probe, a heated filter assembly with a glass fiver filter and Teflonฎ filter
support, a series of impingers, and the standard EPA Method 5 (40 CFR, Part 60,
Appendix A) meterbox and vacuum pump. The sample is not exposed to any metals
surfaces in the train. The contents of the sequential impingers include an optional
knockout impinger for collecting moisture (this impinger was not used during the
Camden County testing), two impingers with a 5% nitric acid (HNO3)/10% hydrogen
peroxide (H2O2) solution, an empty impinger to protect against impinger solution
contamination, two impingers with a 4% potassium permanganate (KMnO4)/10%
sulfuric acid (H2SQ4) solution, and an impinger containing silica gel. The second
impinger containing HNO3/H2O2 was of the Greenburg-Smith design; the other
6-1
-------
The sampling trains were leak checked at the start and finish of sampling. Leak
checks were also performed before and after every port change. The acceptable pre-test
leak rate was less than 0.02 cubic feet per minute (cfm) at approximately 15 inches of
Hg.
After successful completion of the pre-test leak check and when all train
components were at their required temperatures, the initial dry gas meter reading was
recorded and the test was initiated. Sampling train data for each sampling point were
recorded on standard data forms.
Recovery procedures began as soon as the probe was removed from the stack and
the post-test leak check was completed. To facilitate transfer from the sampling location
to the recovery trailer, the sampling train was disassembled into three sections: the
nozzle/probe liner, the filter holder, and the impingers. Each of these sections were
capped with Teflonฎ tape before removal to the recovery trailer.
Once in the trailer, the sampling train was recovered as six separate front-half and
back-half fractions. A diagram illustrating front-half and back-half sample recovery
procedures is shown in Figure 6-2.
6.1.3 Particulate Matter Analysis
The general gravimetric procedure described in Section 4.3 of EPA Method 5 was
used to determine the amount of collected PM. The key difference was the use of a
metal-free probe brush to avoid potential metals contamination of the probe wash
sample. All sample drying, desiccation, and weighing activities were performed in
Radian's Perimeter Park Laboratory.
The filters and precleaned beakers were dried to a constant weight before use.
The same balance was used for weighing the samples prior to and after testing. The
acetone rinses were evaporated to dryness under a clean hood at 70ฐF in a tared beaker.
6-4
-------
impingers had straignt tubes. The impingers were connected together with clean glass
U-tube connectors and were arranged in an impinger bucket.
Equipment preparation included calibration and leak checking of all sampling
train equipment as specified in EPA Method 5. This equipment includes the probe
nozzles, pilot tubes, metering system, probe heater, temperature gauges, leak check
metering system, and barometer.
6.1.2 Sampling Equipment Operation and Recovery
Prior to sampling, preliminary measurements were made to ensure isokinetic
sampling. These included determining the traverse point locations and performing a
preliminary velocity traverse, cyclonic flow check, and moisture determination. These
measurements were used to calculate a "K factor," which was used to determine an
isokinetic flue gas sampling rate.
Measurements were made of the duct inside diameter, port length, and the
distances to the nearest upstream and downstream flow disturbances. These
measurements were used to verify the sampling point locations required by EPA
Method 1 guidelines. The insertion depths were then marked on the sampling probe
using an indelible marker.
After assembling the train, the heaters for the probe liner and filter box were
turned on. The system was then brought to the appropriate temperature, and a pre-test
leak check of the sampling train was conducted. The filter skin temperature was
maintained at 120 ฑ14ฐC (248 ฑ25ฐF). The probe temperature was maintained above
100ฐC (212ฐF).
6-3
-------
The residue was desiccated for 24 hours in a desiccator containing fresh silica gel at
room temperature. The filter was also desiccated under the same conditions to a
constant weight. Each replicate weighing had to agree to within 0.5 mg or 1%
(whichever is greater) between two consecutive weighings, conducted at least 6 hours
apart. Weight gain was reported to the nearest 0.1 mg.
Following weighing, the desiccated filter and acetone rinse samples were sent to
Radian's Summit Park laboratory for metals analysis. The filter and acetone rinse
samples collected from the economizer exit sampling location during the Phase I
Characterization Test were sent directly to Summit Park for expedited Hg analysis. As a
result, PM loadings at the economizer exit are not available for these runs.
6.1.4 Metals Analytical Procedures
A diagram illustrating the sample preparation and analytical procedure for the
target metals is shown in Figure 6-3. As shown in this figure, metals analyses were
conducted on four distinct fractions:
Front-half (filter, acetone probe rinse, and nitric acid probe rinse);
HNO3/H2O2 impingers;
KMnO4/H2SO4 impingers; and
HC1 rinse.
The first two fractions were analyzed for Hg and for other metals. The last two fractions
were analyzed for Hg only. All metals analyses were conducted in Radian's Summit
Park Laboratory.
The acetone probe rinse and HNO3 probe rinse for each train were combined to
yield the front-half sample fraction. The front-half fractions were then digested with
concentrated HNO3 and hydrofluoric (HF) acid in a microwave-heated pressure vessel.
6-6
-------
Probe Liner
and Nozzle
Rinse with
0.1NHNO3lnto
Tared Container
Brush Uner
with Nonmetallic
Brush and Rinse
with0.1NHNO3
at Least
3X
o\
Check Liner
to see If
Particulate
Removed: If
not Repeat
Step Above
I
Front Half of
Filter Housing
Brush with
Nonmetallic
Brush and
Rinse with
0.1NHNO3into
Tared Container
Filter
Carefully
Remove Filter
from Support with
Teflon-Coated
Tweezers and
Place in Petrl Dish
Brush Loose
Particulate
from Holder
onto Filter
Fitter Support
and Back Half
of Filter
Housing
Istlmpinger
(Empty at
beginning
of test)
2nd & 3rd
Impingers
Rlns
wJ
0.1
HM
into!
Cont
B3X
th
N
ฐ3^
ared
ainer
Weigh
Implnger
Contents
I
Calculate
Moisture
Gain
I
Empty
Contents
into
Tared
Container
Weigh
Implnger
Contents
I
Calculate
Moisture
Gain
I
Seal Petrl
Dish with
Teflon Tape
I
Weigh to
Calculate
Rinse Volume
I
APR
(2)
Rinse Impingers
3Xwith
0.1N
HNO3
I
Recover Into
Sample
Container
I
Weigh to
Calculate
Rinse Amount
4th Impinger
(Empty at
beginning
of test)
I
Weigh
Implnger
Contents
I
Calculate
Moisture
Gain
I
Empty
Contents
Into
Tared
Container
Rinse Impinger
3X with ~100 ml
0.1 N
HNO3
I
Recover Into
Sample
Container
I
Weigh to
Calculate
Rinse Amount
5th & 6th
Impingers
(Acidified
(KMnO,)
Last Implnger
Weigh
Impinger
Contents
Calculate
Moisture
Gain
Empty
Contents
into
Tared
Container
Rinse 3X
with 100 ml
Permanganate
Reagent
I
Weigh for
Moisture
Calculate
Moisture
Gain
I
Discard
I
Recover into
Sample
Container
I
I
If visible residue
is present in impingers,
remove with 50 ml
8N HCI solution.
Weigh to Calculate
Sample and
Rinse Volume
KM
(5)
Weigh to Calculate
Sample and Rinse
Volume
IR
(6)
F
(1)
HN
(3)
KR
(4)
Figure 6-2. Metals Sample Recovery Scheme
-------
The microwave digestion took place over a period of approximately 10 to 12 minutes in
intervals of 1 to 2 minutes at 600 watts. The fraction was diluted to a specified volume
with deionized (DI) water and divided for analysis.
Aliquots were taken from each of the remaining four fractions (front-half digest,
HNO3/H2O2, KMnO4/H2OSO4, and HC1) for analysis of Hg by bold vapor atomic
absorption (CVAAS) (EPA Methods 7470 and 7471). Each of these samples were
prepared for analysis as indicated in Figure 6-3.
For the test runs requiring analysis for other metals, aliquots were taken from the
front-half and HNO3/H2O2 fractions. These aliquots were prepared as indicated in
Figure 6-3 and analyzed for other metals by ICAP by EPA Method 6010. Because of the
detection limitations of ICAP for As, Cd, Pb, and Se, additional analyses were conducted
by GFAA. Based on the levels of the metals present at both sampling locations, GFAA
was used for analysis of As (Method 7060, Cd (Method 7131), and Pb (Method 7421) on
the stack front-half fraction, and on the economizer outlet and stack HNO3/H2O2
fractions. For Se (Method 7740), the front-half and HNO3/H2O2 fractions from both the
economizer outlet and stack were analyzed by GFAA. To improve the detection limit
for metals run by GFAA, all of the sample remaining after removal of aliquots for Hg
and ICAP analyses was reduced to near dryness prior to sample preparation.
6.2 CDD/CDF
The sampling and analytical method used for determining flue gas emissions of
CDD/CDF was EPA Method 23.11 Sample recovery techniques incorporated the latest
EPA Office of Research and Development guidance on replacing the methylene chloride
rinses with toluene rinses. Samples were simultaneously collected at the economizer
outlet and in the stack. Samples times during each run were two hours.
6-8
-------
ContBJnerS
Add Probe Flnse
(Labeled APR)
Contalnar2
Acetone Probe Rinse
(Labeled PR)
ConMnw 1
Flftsc
(Labeled F)
Container 4
HNCyHftlmplngers
(Labeled HN)
(Include CondensatB
tmplnger.lfUsed)
I
Containers
Permanganate Impinge
(Labeled KM)
Container 6
HCI Rinse
(Labeled IR)
Reduce to Dryneea
In a Tared Beaker
Desiccate to
Constant Weight
Residue
Weight In Beaker
Determine Filter
PartJcutatB Weight
Sokiblllze Residua
wfth Cone. HNO 3
Acidify to pH 2
wtthConc.HNO3
Romovs
50-100 ml AHquot
for Hfl Analysis
CVAAS
Fraction 28
Acidify
Remaining
Sample to pH
of 2 with
Cone. HNO 3
Fraction 2A
Digest wfth Acid
and Permanganate
at 95% for 2 hr
and Analyze
for Hg try CVAAS
Fractions
Digest with Add
and Permanganate
at WC for 2 hr
and Analyze
for Hg by CVAAS
Reduce Volume
to near
Dry ness and
Digest with
HNO3andKp2
Divide Into O.S g
Sections and Digest
Each Section with
Cone. HF and HNO 3
Reduce Volume to
Near Dry ness and
Digest with HF and
Cone. HNO 3
Analyze by
ICAP for
Target Metals
Analyze for
Hg by CVAAS
Rtter and Dilute
to Known Volume
Fraction 1
_L
Remove 50 to 100 mL
Aliquot for Hg
Analysis by CVAAS
Fraction 18
Digest with Add and
Permanganate at 95"C In
a Water Bath for 2 hr
Analyze by (CAP for
Target Metals
Fraction 1A
Analyze for
Metals by QFASS
Fraction 1A
Analyze AHquot for
Hg Using CVAAS
Figure 6-3. Metals Sample Preparation and Analysis Scheme
-------
Gooseneck
Nozzle
Stack
Temperature /
Sensor >
A
Temperature Sensor
X
FiHer Holder
Condenser
Temperature Sensor
Flue Gas
S-Type Pilot Tube
Temperature Sensor
Heat Traced
Quartz Probe
Uner
Water Knockout 100 ml HPLC Water Empty
Implnger
Manometer
Figure 6-4. CDD/CDF Sampling Train Configuration
Vacuum
Line
K
I
8
-------
All of the CDD/CDF analyses, as well as preparation of the XAD-H collection
modules, were performed by Twin Cities Testing in St. Paul, Minnesota. Preparation of
all other sampling train equipment was conducted by Radian.
6.2.1 Sampling Equipment Preparation
The CDD/CDF sampling method used the sampling train shown in Figure 6-4.
The sampling system was similar to a Method 5 train with the exception of the following:
All components (glass probe liner/nozzle, all other glassware, filters) were
pre-cleaned using solvent rinses and extraction techniques; and
A condensing coil and XAD-IIฎ resin absorption module for collection of
CDD/CDF were located between the filter and impinger train.
In addition to the standard EPA Method 5 requirements, the CDD/CDF
sampling method includes several preparation steps for ensuring that the sampling train
components are not contaminated with organics that may interfere with analysis. The
glassware, glass fiber filters, and absorbing resin were cleaned, and the filters and resin
were checked for residue before they were packed. The remaining preparation included
calibration and leak checking of all sampling train equipment, including meter boxes,
thermocouples, nozzles, pitot tubes, and umbilicals.
6.2.2 Sampling Equipment Operation and Recovery
The CDD/CDF preliminary measurement procedures and sampling procedures
were identical to those described in Section 6.1.2 for the multiple metals sampling. To
facilitate transfer from the sampling location to the recovery trailer, the sampling train
was disassembled into the following sections: probe liner, filter holder,
6-9
-------
N)
Back Half
Front Half of Filter Connecting
Probe Monte Probe Uner Filter Housing Filter Support Housing Line Condenser Filter Resin Trap
1 J 1 1 1 1
1 1
1
RlnsewHh Attach Brush and Rlnsewlth Rlnsewlth Rlnsewlth Rlnsewlth Carefully SecureXAD
Acetone 250 mL Flask RlnsewHh Acetone Acetone Acetone Acetone Remove filter Trao
until all toBallJoInt Acetone (3x)
Paniculate i (3x)
to Removed f
Otnmm "III-
rflnM Wun
Ao6tona
Empty Flask
Into 950 mL
Bottte
Brush Uner
and Rinse
with 3
AHquotsof
Acetone
1
Check Uner
to See It
Partculate
to Removed,
If Not Repeat
J
Rinse Rinse Rinse
with wtth with
Toluene Toluene Toluene
(3
X) (3X) (3
1 1
ป)
(3x) (3x) (3x) from Support Openings
lii with Tweezers with Glass
T f T Balls and
Rinse ntma Rinse Brush Loose Clamps
with T~ with Partculate
Toluene *"" Toluene onto Filter Place In
(3>4 Toluene (3x)
(t toast once (3x) (at toast once
tot the rinse
stand 5 minutes
In unit)
,
Cooler for
Storage
tot the rinse Seal In
stand 5 minutes Petit Dtoh
In unit)
F
SM
1st Implnger
(knockout)
Weigh
Implnger
2nd Implnger
I
Weigh
Implnger
3rd Implnger
t
Weigh
Implnger
4th Implnger
5th Implnger
(Silica Gel)
1 I
Weigh
Implnger
I I I J
Record
Weight
and
Calculate
Gain
Record
Weight
and
Calculate
Gain
Record
Weight
and
Calculate
Gain
Record
Weight
and
Calculate
Gain
Weigh
Implnger
Record
and
Calculate
Gain
I I J I I
Discard
Discard
Discard
Discard
Save
for
Regeneration
Recover Into
prewelghed
bottle
Figure 6-5. CDD/CDF Field Recovery Scheme
-------
filter-to-condenser glassware, condenser/sorbent module, and impingers. Each of these
sections were capped with methylene chloride-rinsed aluminum foil or ground glass caps
before removal to the recovery trailer. Once in the trailer, sample recovery followed the
scheme shown in Figure 6-5. The samples were recovered and stored in cleaned amber
glass bottles to prevent light degradation.
All CDD/CDF recovery rinses were completed using toluene instead of
methylene chloride. All solvents used for train recovery were pesticide grade. To
prevent the introduction of chemical impurities which interfere with the quantitative
analytical determination, the highest grade reagents were used for train recovery.
Field recovery resulted in the sample components listed in Table 6-1. The
sorbent module was stored on ice at all times. The samples were shipped to the
analytical laboratory accompanied by written information designating target analyses.
6.2.3 Analytical Procedures
High resolution gas chromatography (HRGC) and high resolution mass
spectrometry (HRMS) were used to determine CDD/CDF concentrations. The target
CDD/CDF congeners are listed in Table 6-2.
Each of the field sample fractions was combined into a single sample and
analyzed according to the scheme in Figure 6-6. For the CDD/CDF analysis,
isotopically-labeled surrogate compounds and internal standards were added to the
samples before the extraction process was initiated. The internal standards and
surrogates that were used are described in detail in EPA Method 23.
6-11
-------
TABLE 6-2 TARGET CDD/CDF CONGENERS
CAMDEN COUNTY MWC (1992)
DIOXINS:
2,3,7,8 tetrachlorodibenzo-p-dioxin (2,3,7,8 TCDD)
Total tetrachlorinated dibenzo-p-dioxins (TCDD)
1,2,3,7,8 pentachlorodibenzo-p-dioxin (1,2,3,7,8 PeCDD)
Total pentachlorinated dibenzo-p-dioxins (PeCDD)
1,2,3,4,7,8 hexachlorodibenzo-p-dioxin (1,2,3,4,7,8 HxCDD)
1,2,3,6,7,8 hexachlorodibenzo-p-dioxin (1,2,3,6,7,8 HxCDD)
1,2,3,7,8,9 hexachlorodibenzo-p-dioxin (1,2,3,7,8,9 HxCDD)
Total hexachlorinated dibenzo-p-dioxins (HxCDD)
1,2,3,4,6,7,8 heptachlorodibenzo-p-dioxin (1,2,3,4,6,7,8 HpCDD)
Total heptachlorinated dibenzo-p-dioxins (HpCDD)
Total octachlorinated dibenzo-p-dioxins (OCDD)
FURANS:
2,3,7,8 tetrachlorodibenzofiirans (2,3,7,8 TCDF)
Total tetrachlorinated dibenzofurans (TCDF)
1,2,3,7,8 pentachlorodibenzofuraii (1,2,3,7,8 PeCDF)
2,3,4,7,8 pentachlorodibenzofuran (2,3,4,7,8 PeCDF)
Total pentachlorinated dibenzofurans (PeCDF)
1,2,3,4,7,8 hexachlorodibenzofuran (1,2,3,4,7,8 HxCDF)
1,2,3,6,7,8 hexachlorodibenzofuran (1,2,3,6,7,8 HxCDF)
2,3,4,6,7,8 hexachlorodibenzofuran (2,3,4,6,7,8 HxCDF)
1,2,3,7,8,9 hexachlorodibenzofuran (1,2,3,7,8,9 HxCDF)
Total hexachlorinated dibenzofurans (HxCDF)
1,2,3,4,6,7,8 heptachlorodibenzofuran (1,2,3,4,6,7,8 HpCDF)
1,2,3,4,7,8,9 heptachlorodibenzofuran (1,2,3,4,7,8,9 HpCDF)
Total heptachlorinated dibenzofurans (HpCDF)
Total octachlorinated dibenzofurans (OCDF)
6-14
-------
TABLE 6-1 CDD/CDF SAMPLE FRACTIONS SHIPPED
TO ANALYTICAL LABORATORY
CAMDEN COUNTY MWC (1992)
Container/
Component
1
2
3
Code
F
PRa
SM
Fraction
Filter(s)
Acetone and toluene rinses of nozzle, probe,
front-half/back-half filter holder, filter
support, connecting glassware, and condenser
XAD-II resin trap (sorbent module)
"Rinses include acetone and toluene which were recovered into the same sample bottle.
6-13
-------
Data for the mass spectrometer were recorded and stored on a computer file, as
well as printed on paper. Results such as amount detected, detection limit, retention
time, and internal standard and surrogate standard recoveries were calculated by
computer. The chromatograms were retained by Twin City Testing and were also
included in the analytical report delivered to Radian.
6.3 Volatile Organic Compounds
Sampling for VOC was conducted according to SW-846 Method 0030.12 The
VOST is designed to collect VOC with boiling points between 86ฐF and 212ฐF. Sampling
for VOC was limited to two test conditions and included simultaneous sampling at the
economizer exit and stack. During each VOST run, four pairs of collection traps were
used, with each pair being used for 20 minutes at a sampling rate of 1 L/min.
The list of target analytical species is given in SW-846 Method 824012 and
presented in Table 6-3. Flue gas detection limits for most of the compounds are about
1.0 ^g/m3, except for polar molecule water-soluble compounds which have higher
detection limits. Preparation of the resin traps used for sample collection and analysis of
collected samples was conducted by Air Toxics, Ltd in Rancho Cordova, California.
Preparation of the sampling trains and associated equipment was performed by Radian.
6.3.1 Sampling Equipment and Preparation
A schematic of the VOST is shown in Figure 6-7. The flue gas was sampled from
the stack through a glass probe containing a glass wool plug. The probe temperature
was maintained above 300ฐF. The gas sample was cooled to 68ฐF by a water-cooled
condenser and was passed through a pair of resin traps in series, a silica gel drying tube,
a rotameter, a sampling pump, and a dry gas meter. The first resin trap contained
approximately 1.6 g of Tenax and the second trap contained approximately 1 g of Tenax
followed by 1 g of petroleum-based charcoal. The rotameter indicated the volumetric
6-16
-------
Acetone^Toluene
Rinses
Concentrate
at Temperature
<37ฐC(980F)
Silica Gel Column
Chromatography Cleanup;
Concentrate Elua
luate to 1 ml
withN2
Basic Aluminum Column
Chromatography Cleanup;
Concentrate Eluate to
0.5 mL with N2
FK-21 Carbon/Celite 545
Column Chromatogaphy
Cleanup; Concentrate
Eluate t.OmLin
Rotary Evaporator
Concentrate Eluate to
200 mL with N2
Store in Freezer
Analyze with DB-5
Capillary column; if
TCDF is Found, Continue
Analyze with
DB-225
Column
Quantify Results
According to
Section 5.3.2.6
of Reference Method 23
ง
3
Figure 6-6. Extraction and Analysis Schematic for CDD/CDF Samples
6-15
-------
Sampling Module
Metering Valves
Heated Probe
oo
Quartz Wool
Tenax
Tenax
Charcoal
Silica Gel
Charcoal
To Ambient
To Umbilical
Flask
Flask
Control Module
(7) Temp. Adj.
Probe Heat
Pump
Power
Calibrated
Rotameter
0,
Metering
Valve
Valve ^__
Pu
| Temp. Readout
T1 Probe
T2 Probe Exit
T3 Tenax #1
I T4 Tenax #2
T5 Meter In
T6 Meter Out
^ Pressure
Q Gauge -"H2O
Vacuum
iauge - "Hg
\r^=\ O H^
5=J / X
'n / \
J / Dry Gas \
] Meter 1 LPR ]
mp y Calibrated /
-^ v y
Figure 6-7. Schematic of YOST Sampling Train
-------
TABLE 6-3. VOLATILE COMPOUNDS QUANTIFIED BY
SW-846 METHOD 8240
CAMDEN COUNTY MWC (1992)
Compound
Compound
Acetone
Acrolein
Acrylonitrile
Benzene
Bromochloromethane (I.S.)
Bromodichloromethane
p-Bromofluorobenzene (sun.)
Bromoform
Bromomethane
2-Butanone
Carbon disulfide
Carbon tetrachloride
Chlorobenzene
Chlorobenzene-d5 (I.S.)
Chlorodibromomethane
Chloroethane
2-Chloroethyl vinyl ether
Chloroform
Chloromethane
Dibromomethane
l,4-Dichloro-2-butane
Dichlorodifluoromethane
1,1-Dichloroethane
1,2-Dichloroethane
l,2-Dichloroethane-d4 (SUIT.)
1,1-Dichloroethene
trans- 1,2-Dichloroethene
1,2-Dichloropropane
cis-l,3-Dichloropropene
trans- 1,3-Dichloropropene
1,4-Difluorobenzene (I.S.)
Ethanol
Ethylbenzene
Ethyl methacrylate
2-Hexanone
lodomethane
Methylene chloride
4-Methyl-2-pentanone
Styrene
1,1,2,2-Tetrachloroethane
Tetrachloroethene
Toluene
Toluene-d8 (SUIT.)
1,1,1-Trichloroethane
1,1,2-Trichloroethane
Trichloroethene
Trichlorofluoromethane
1,2,3-Trichloropropane
Vinyl acetate
Vinyl chloride
Xylene
6-17
-------
6.3.3 Analytical Procedures
The sorbent cartridges were analyzed according to SW-846 Method 5041/8240.
Method 5041 defines thermal desorption techniques for processing the resin traps.
Analysis was then completed by purge and trap GC/MS as shown in Figure 6-8. This
procedure utilizes HRGC and low resolution mass spectroscopy (LRMS). One sample
was screened by GC/FID to determine the relative concentration of the species prior to
mass spectroscopy.
6.4 Fly Ash Carbon Content
A daily fly ash sample was collected for analysis of unburned carbon. The daily
sample was withdrawn from a single point in the economizer outlet using an EPA
Method 5 sampling train. The nozzle was sized to allow isokinetic sampling at
approximately 0.5 dscfm. A cyclone was used in front of the filter to facilitate collection
of a large volume of ash without clogging the filter. The train was run for the duration
of the test period each day to ensure that a representative sample was collected. Each
sample was analyzed for carbon content by Commercial Testing and Engineering in
South Holland, Illinois, using ASTM Method D3178-84, Carbon and Hydrogen in the
Analysis Sample of Coal and Coke.13
6.5 Particle Size Distribution
Flue gas samples were collected from the stack during each of the Unit A tests to
determine the size distribution of emitted particles. Sampling was conducted using an
Andersen Mark HI pre-impactor and a Mark II 8-stage cascade impactor. During each
of these tests, two trains were operated to collect duplicate composite samples over the
duration of the test day. Particles were separated based on their inertial properties as
they flowed through succeeding stages with smaller acceleration jets (higher velocities).
Larger particles impacted on the initial collection stages and smaller particles were
6-20
-------
gas sampling rate, and the dry gas meter recorded the total gas volume that passed
through the meter during the sampling period.
Prior to field use, the glass tubes and condensers used with the VOST were
cleaned with a non-ionic detergent in an ultrasound bath, rinsed three times with
organic-free water, and dried at 212ฐF. The traps were filled and conditioned according
to the above referenced protocol and analyzed by GC/FID to verify that the traps were
free from background contamination. Preparation of the sampling equipment included
calibration of dry gas meters and temperature measuring devices.
6.3.2 Sampling Equipment Operation and Recovery
The VOST probe was inserted to a single point of average gas velocity in the
centroid area of the duct or stack. Since the target species were gases, isokinetic
sampling was not required. The trains were leak checked before and after sampling with
each pair of traps.
The handling procedures for the VOST traps emphasized the need to minimize
potential sample contamination. The VOST traps were stored in a clean cooler,
separate from all other types of samples collected at the site. Used traps were stored on
cold packs or ice. The time that the traps were exposed to ambient air during train
assembly and disassembly was minimized.
One pair of field blank traps was collected per test run by removing the end caps
from a pair of traps for the length of time required to exchange two pairs of traps during
sampling (approximately 5 minutes). Also, one pair of trip blanks was collected per test
site. The trip blank consisted of a pair of traps that were taken to the site and were
stored with the other VOST samples, but remained capped throughout the test.
6-19
-------
collected in the downstream stages. The impactor was placed inside the stack so that the
sample stream maintained all of the flue gas's physical characteristics, such as
temperature and viscosity. The sample was then isokinetically extracted through the
nozzle so that a representative distribution of particles was collected. All desiccation
and weighing of filtrates and rinses to determine the mass of collected particulate was
conducted in Radian's Perimeter Park Laboratory.
6.5.1 Sampling Equipment Preparation
The PSD train is shown in Figure 6-9. The sampling train is similar to the EPA
Method 5 train except that a pre-separator and a cascade impactor were used in the
stack instead of a glass-lined probe and external filter. Prior to sampling, the impactor
housing, nozzle, and filter holders were cleaned. All filters were desiccated, placed
inside a folded sheet of cleaned aluminum foil, and tared on a five-place balance prior to
use. The filter used in the impactor were Reeve Angel 934AH substrates. The foil and
filters were handled with tweezers to avoid weight gain due to fingerprints. Replicate
weighings at lest six hours apart had to agree within 0.05 mg in order to accept the
weight.
After preparing the impingers, the probe was attached to the impingers, and the
system was checked for a leak rate of less than 0.0005 m3/min (0.02 cfm). After a
successful leak check, the impactor and nozzle were attached to the probe. The nozzle
size was selected to allow isokinetic sampling at the flow rate required for proper
particle separation. The impactor was preheated to approximately stack temperature by
placing it inside the stack with the nozzle sealed and out of the flue gas flow.
6.5.2 Sampling Equipment Operation and Recovery
Sampling was conducted at a single point of average flue gas velocity at a fixed
sampling rate. The sampling rate was adjusted to obtain the required flow rate through
the impactor based on expected gas conditions and was not adjusted during the run.
6-22
-------
VOST ANALYSIS PROTOCOL
Condensate
Tenax and/or
Tenax-Charcoal Tube
Add 5.0 mL of
Condensate to
the Purging Device
Spike the Tube(s) with Benzene
250 ng bromochloromethane
250 ng 1,4-dffluorobenzene
250 ng ds- chlorobenzene
While at Room Temperature
Spike the Condensate
Prior to Purging with
Internal Standards (250 ng
each of bromochloromethane
1,4-dffluorobenzene
and ds-chlorobenzene
Additional Spikes
d4-1,2-dich!oroethane 250 ng
p-bromofluorobenzene 250 ng
da-toluene 250 ng
Place Tube(s) in Desorption
Unit and Desorb for
10 Minutes at 180*0 onto
the Analytical Trap
Spike the Condensate
Prior to Purging with
Surrogate Standards
(250 ng each of
d 4 1,2 dichloroethane,
dg-toluene, and
p-bromofluorobenzene)
Use the Purge and Trap
Apparatus as Described in
Method 5041. Rapidly Heat the
Analytical Trap to 180ฐC
4-5 minutes
Analyze the Desorbed
Compounds by GC/MS per
Method 5041
Analyze the Condensate
by GC/MS per
Method 8240
Figure 6-8. VOST Analysis Protocol
6-21
-------
After sampling was completed, the impactor was cooled in a vertical position prior
to recovery. During the recovery operation, each filter was examined for particle
bounce, overloading, and reentrainment. Any particles lost to surfaces upstream of a
stage substrate were recovered by dry brushing and added to that substrate. The filter
substrate and collected paniculate from each stage were placed inside the same piece of
cleaned aluminum foil with which the filter was tared and sealed by crimping the edges
of the foil. This approach minimized the loss of particulate from the sample during
shipment. Particles from the nozzle and pre-separator were collected in a separate
fraction using an acetone rinse. In the final calculations, the weight of collected
particulate in this fraction was added to that of the first stage.
The substrates were desiccated and allowed to come to a constant weight
(ฑ0.05 mg). The final values were reported to the nearest 0.01 mg.
6.6 Volumetric Flow Rate and Moisture Content
The volumetric flow and moisture content of the flue gas during each run were
based on the data collected from the multiple metals train.
6.6.1 Determination of Duct Gas Velocity by EPA Method 2
The volumetric flow rate (duct gas velocity) was measured according to EPA
Method 2 (40 CFR Part 60, Appendix A). A Type K thermocouple and S-type pitot tube
were used to measure flue gas temperature and velocity, respectively. The pitot tubes
were inspected before being used and were leak checked before and after each run,
following the protocols in the method.
6-24
-------
Temperature
Sensor
Thermometer
Andersen Mark
Eight-Stage Impactor
fc
Straight Nozzle ^ / ^^
Manometer
Figure 6-9. Sampling Train for Particle Size Distribution Tests
-------
The stack CEMS was also an extractive system. The effluent gas sample was
drawn from the stack via a filter and heated sampling line to a Bodenseewerk Mekos
multicomponent analyzer, which analyzes the gas sample for O2, CO2, H2O, CO, THC,
CH4, SO2, HC1, and NOX. The gas sample exiting the Mekos analyzer passed through an
electrical gas cooler to remove moisture and was then delivered to Westinghouse/
Maihak Oxygor O2 and JUM THC. analyzers. A Monitrol T5C-1000 opacity monitor
installed on each stack continuously monitored opacity levels.
The analog gas concentration data from the analyzers was transmitted to the
respective Odessa CEM data acquisition system. The Odessa DAS calculated the
pollutant emission levels in parts per million (ppm) corrected to 7% O2. These values
were then archived on the DAS once every minute. At the end of each test day, the
one-minute readings for all flue gas parameters were transferred to a floppy disk by
plant personnel and given to Radian.
6.7.2 Calibration
The CEMS had been certified using EPA QA/QC protocols for CEMS (40 CFR
Part 60, Appendix F). During the testing, the CEMS were calibrated daily using a
two-point calibration. A low-level calibration gas (typically a zero concentration gas) and
a high-level calibration gas were used for this procedure. All calibrations were
completed by passing the calibration gas through the entire sampling system. The results
of these calibrations were printed in a daily report.
6.8 Process Data Collection
Combustor conditions and SD/ESP operating parameters were monitored using
CCRRFs existing data acquisition systems. Combustor operating parameters that were
recorded included one-minute average steam production rate, furnace temperature, and
6-26
-------
The parameters measured at each traverse point included:
Pressure drop across the pitot;
Stack temperature;
Stack static; and
Ambient pressure.
A Method 5 computer program was used to calculate the average velocity during the
sampling period.
6.6.2 Determination of Flue Gas Moisture Content by EPA Method 4
The flue gas moisture content was determined according to EPA Method 4
(40 CFR Part 60, Appendix A). Before sampling, the initial weight of the impingers was
recorded. When sampling was completed, the final weights of the impingers were
recorded, and the weight gain was calculated. The weight gain and the volume of gas
sampled were used to calculate the average moisture content (percent) of the flue gas.
6.7 Continuous Emission Monitors
6.7.1 Equipment Description
The flue gas composition at the economizer exit and stack was monitored during
each test using the permanently installed CEMS operated by the CCRRF. The CEMS
on both units were identical.
The CEMS at the economizer exit included extractive SO2 and O2 monitors. Gas
samples were extracted through a sintered probe and heated sampling line and delivered
to a gas conditioner for moisture removal. The gas was then supplied to Western
Research SO2 and Rosemount Analytical O2 analyzers.
6-25
-------
economizer exit temperature. Operating parameters of the SD/ESP that were recorded
included:
Lime slurry and dilution water flow rates;
SD inlet and outlet temperatures;
ESP secondary voltage, secondary amperage, and spark rate per field; and
Stack SO2 and opacity levels.
Each of the SD/ESP parameters were recorded as instantaneous values read once each
minute, rather than as one-minute averages. All of these data were continuously logged
onto the plant DAS systems -- a Bailey NET-90 for the combustor parameters and a
Belco Merlin system for the SD/ESP parameters. At the end of each test day, the data
spanning the testing period were downloaded by CCRRF and Belco personnel onto a
floppy disk and given to Radian.
6-27
-------
7.1 Overview of Data Quality
The QAPP established specific QA objectives for precision and accuracy for
measurement of each flue gas emission parameter, including Hg, other metals,
CDD/CDF, VOST, O2, and particulates. The primary QC results used to evaluate
precision and accuracy for each analytical parameter are summarized in Table 7-1.
Results of matrix spike/matrix spike duplicates were used as QC indicators for Hg and
the other metals. Results of surrogate spikes were used as QC indicators for analyses
using GC/MS methods. Measured QC values that are not within the specified data
quality objectives are discussed in detail in Sections 7.4 and 7.5. Other data quality
indicators for each type of analysis are presented throughout the remainder of Section 7.
There are no cases where data quality issues prevent sound conclusions from
being made regarding the effectiveness of carbon injection in reducing emissions of Hg,
Cd, Pb, CDD/CDF, and volatile organics. With the exception of a limited number of
samples, the quality of measurement data generated for the test parameters fully meet
the data quality objectives outlined in the QAPP. Generally, there is no impact on the
acceptability of the data quality, except for Se. Data quality issues related to Se are
summarized in Section 7.4.
7.2 Sampling Quality Control
Sampling activities conducted during the Camden County MWC testing include
the following for stack gases:
EPA multi-metals method for determination of Hg, other metals, and
paniculate matter;
EPA Method 23 for determination of CDD/CDF;
EPA SW-846 Method 0030 for determination of volatile organics; and
In-stack Anderson cascade impactor particle size distribution.
7-2
-------
7.0 QUALITY ASSURANCE/QUALITY CONTROL
As a part of the testing at the Camden County MWC, Radian designed and
implemented a quality assurance/quality control (QA/QC) effort tailored to meet the
specific needs of this project. The testing was conducted in accordance with QA/QC
procedures described in the Quality Assurance Project Plan (QAPP). The results of the
QA/QC effort demonstrate that the data are reliable; defensible, and meet project
objectives for completeness, representativeness, and comparability. The data meet the
QA objectives for precision and accuracy and there are no data quality issues that effect
conclusions regarding the effectiveness of carbon injection.
The primary objectives of the QA/QC effort were to control, assess, and
document data quality. In order to accomplish these objectives, the QA/QC approach
consisted of the following key elements:
Definition of data quality objectives that reflect the overall technical
objectives of the project;
Design of a sampling, analytical, QA/QC, and data analysis system to meet
these objectives;
Evaluation of the measurement system performance; and
Initiation of corrective action when measurement system performance did
not meet the specifications.
These elements include the use of validated or standard sampling and analytical
procedures, along with specified calibration requirements, QC checks, data reduction and
validation procedures, and sample tracking.
A summary of analysis results for QA/QC samples, which includes measures of
precision and accuracy, and limitations in the use of this data is presented in this section.
7-1
-------
Quality control activities associated with sampling are described in the QAPP. These
activities include adherence to accepted reference method protocols, use of standardized
data recording sheets, equipment calibration, and collection of field blanks. Records
documenting these sampling activities are presented in the Appendices of this report.
7.2.1 Multi-Metals Method Flue Gas Sampling Quality Control
Stack sampling QC data, including isokinetic sampling rates, sample volume
collected, maximum recorded leak rate, and maximum allowable leak rate, are
summarized in Table 7-2 for each multi-metals method run. All of the data quality
indicators are within acceptable limits, with the exception of low isokinetic sampling
rates for three runs and high leak rates for four runs.
The isokinetic sampling rates for Phase I Outlet Run 1 (81%), Phase II Outlet
Run 26 (89%), and Phase n Inlet Run 35 (88%) were below the QC objective of 90 to
110% isokinetic. The low isokinetic sampling rate for these runs do not significantly
effect the metals results because the isokinetic sampling rate was only slightly outside the
QC objective. Emission rates for these test runs may have a slight high bias due to the
low isokinetic sampling rate.
The acceptance criteria for sample train leak checks is a leak rate of less than 4%
of the average sampling rate or 0.02 dscf, whichever is less. This criteria was met by all
of the outlet sampling trains and by 49 of 53 inlet sampling trains. Two of the four high
leak rates met the 0.02 dscf criteria, but were 5% of the sample rate (Phase II Inlet
Runs 18 and 22). The other two high leak rates were 10% (Phase II Inlet Run 19) and
20% (Phase n Inlet Run 30). The final sample volume for these four test runs were not
corrected for the high leak rates. If corrections had been made to account for leaks, flue
gas flow rates would be 1 to 7% lower than shown.
7-4
-------
TABLE 7-1. COMPARISON TO QUALITY CONTROL OBJECTIVES
CAMDEN COUNTY MWC (1992)
Parameter
Mercury
Cadmium
Lead
Other Metals0
CDD/CDF
Volatile
Organics
Oxygen
QC Analysis
Matrix Spike
Matrix Spike
Matrix Spike
Matrix Spike
Surrogate Spike
Method Spike
Surrogate Spike
Method Spike
Daily CEM Cal
Annual RATA
Precision
Measured
0% - 15.4%
0% - 17.4%
0% - 10.2%
0% - 20.6%b
6% - 42.2%b
2.8% - 55%b
1.5% - 60.3%b
4.9% - 17.0%
0% - 14.4%b
Objective
<20 RPD
<20RPD
<20 RPD
<20 RPD
<40% RSD
<40% RSD
<40% RSD
<40% RSD
< 10% CV
Ratio'
22/22
12/12
12/12
154/155
9/10
16/17
35/36
6/6
47/48
Accuracy
Measured
70% - 138%b
82% - 116%
66% - 146%b
0% - 146%b
33%.- 128%b
68% - 260%b
80% - 308%b
87% - 127%
92% - 98%
Objective
70% - 130%
70% - 130%
70% - 130%
70% - 130%
50% - 150%
50% - 150%
50% - 150%
50% - 150%
80% - 120%
Ratio
40/44
24/24
21/24
288/298
95/100
49/51
178/180
36/36
4/4
"Number of samples meeting QC objective compared to total number of samples analyzed.
bMeasurements outside of the specified objectives are discussed in Section 7.4 for each analytical parameter and matrix.
'Summary statistics do not include selenium. Selenium met 41.7% of the data quality accuracy objectives.
-------
TABLE 7-2, CONTINUED
Ron Number
Isokmetic
(%)
Standard
Meter
Volume
(dsd)
Average
Sample Rate
(dscfin)
Manmnm Leak
Check
(dscf@inHg)
4% Sample
Rate
(dscfin)
Acceptable
Leak Rate?*
Phase I - Metals - Outlet
10
11
12
13
14
15
101
103
101
102
101
102
Phase II - Metals - Inlet
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15R
16
17
18
19
20
99.7
102
101
101
102
110
104
109
105
109
108
108
108
107
108
106
107
108
106
108
43.60
38.81
41.87
36.21
38.88
34.74
0.727
0.647
0.698
0.603
0.648
0.579
29.90
27 .59
27.10
28.19
22.44
24.89
24.99
22.23
25.20
20.99
22.27
22.36
22.39
23.46
22.42
22.59
24.14
19.66
23.16
20.56
0.498
0.460
0.452
0.470
0.374
0.415
0.416
0.371
0.420
0350
0371
0373
0373
0.391
0.374
0377
0.402
0328
0.386
0.343
0.007 @ 10
0.007 @ 10
0.005 @ 10
0.007 @ 10
0.010 @ 10
0.008 @ 11
0.029
0.026
0.028
0.024
0.026
0.023
Y
Y
Y
Y
Y
Y
0.012 @ 8
0.017 @ 6
0.014 @ 7
0.012 @ 10
0.006 @ 6
0.009 @ 7
0.010 @ 10
0.015 @ 14
0.008 @ 12
0.008 @ 14
0.009 @ 11
0.005 @ 12
0.008 @ 12
0.009 @ 10
0.009ฎ 4
0.008 @ 14
0.011 @ 12
0.015 @ 6
0.040ฎ 4
0.008 @ 14
0.020
0.018
0.018
0.019
0.015
0.017
0.017
0.015
0.017
0.014
0.015
0.015
0.015
0.016
0.015
0.015
0.016
0.013
0.015
0.014
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
Y
-------
TABLE 7-2. METALS STACK SAMPLING QUALITY CONTROL DATA
CAMDEN COUNTY MWC - PHASE I & II (1992)
Ron Number
Isokmetic
(%)
Standard
Meter
Volume
(dsd)
Average
Sample Rate
(dscfin)
Maximum Leak
Check
(dsd @ in Hg)
4% Sample
Rate
(dscfin)
Acceptable
Leak Rate?"
Phase I - Metals - Inlet
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
96.8
102
102
101
102
97.3
100
98.0
100
99.1
101
99.0
99.6
98.6
99.8
38.52
31.32
33.89
36.41
36.20
35.81
33.87
33.52
36.52
34.65
32.73
35.15
32.67
31.29
29.16
0.642
0.522
0.565
0.607
0.603
0.597
0.565
0.559
0.609
0.577
0.546
0.586
0.544
0.522
0.486
0.005 @ 4
0.012 @ 10
0.009 @ 9
0.009 @ 10
0.008 @ 14
0.008 @ 14
0.005 @ 14
0.015 @ 15
0.015 @ 5
0.008 @ 10
0.009 @ 10
0.008 @ 15
0.011 @ 15
0,009 @ 8
0.009 @ 15
0.026
0.021
0.023
0.024
0.024
0.024
0.023
0.022
0.024
0.023
0.022
0.023
0.022
0.021
0.019
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Phase I - Metals - Outlet
1
2
3
4
5
6
7
8
9
81.0
107
104
108
102
99.9
102
100
103
38.78
41.30
37.46
41.68
41.96
35.81
41.39
40.57
39.24
0.646
0.688
0.624
0.695
0.699
0.597
0.678
0.665
0.643
0.006ฎ 7
0.010 @ 8
0.016 @ 8
0.008 @ 10
0.012 @ 10
0.008 @ 10
0.014 @ 10
0.006 @ 11
0.018 @ 11
0.026
0.028
0.025
0.028
0.028
0.024
0.027
0.027
0.026
Y
Y
Y
Y
Y
Y
Y
Y
Y
-------
TABLE 7-2, CONTINUED
Ron Number
kokffietic
(%)
Standard
Meter
Volume
(dsd)
Average
Sample Rate
(dscfin)
Phase II - Metals - Outlet
. 12
13
14
15
15R
16
17
18
19
20
21
22
23
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
99.1
98.4
103
103
102
104
108
107
107
107
109
97.3
102
107
89.0
105
97.4
103
106
106
103
107
102
102
106
105
104
101
37.24
35.14
38.41
35.33
36.82
38.40
39.56
33.13
35.43
37.21
32.27
34.85
37.33
74.47
59.21
75.72
69.18
70.99
75.38
37.93
39.99
40.79
72.61
71.13
76.32
35.31
32.09
37.95
0.621
0.586
0.640
0.589
0.614
0.640
0.659
0.552
0.590
0.620
0.556
0.581
0.622
0.621
0.493
0.631
0.577
0.592
0.628
0.632
0.678
0.680
0.605
0393
0.636
0389
0335
0.633
Maximum Leak
Check
(dscf@inHg)
4% Sample
Rate
(dscfin)
Acceptable
Leak Rate?*
0.018 @ 10
0.010 @ 7
0.014 @ 10
0.018 @ 11
0.018 @ 11
0.017 @ 10
0.006 @ 8
0.012 @ 9
0.010 @ 8
0.013 @ 7
0.016 @ 10
0.016 @ 8
0.005 @ 8
0.018 @ 8
0.018 @ 8
0.010 @ 8
0.010 @ 9
0.005 @ 8
0.010 @ 9
0.011 @ 10
0.012 @ 9
0.012 @ 9
0.019 @ 7
0.018 @ 7
0.018 @ 9
0.015 @ 6
0.006ฎ 5
0.014 @ 10
0.025
0.023
0.026
0.024
0.025
0.026
0.026
0.022
0.024
0.025
0.022
0.023
0.025
0.025
0.020
0.025
0.023
0.024
0.025
0.025
0.027
0.027
0.024
0.024
0.025
0.024
0.021
0.025
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
oo
"The values shown in the table for 4% of the sample rate were compared to a value of 0.02 dscfm. The maximum allowable leak rate was established
as the lesser of two values.
-------
TABLE 7-2, CONTINUED
Run Number
Isokinetic
(%)
Standard
Meter
Volume
(dscf)
Average
Sample Rate
(dscfim)
Maximum Leak
Check
(dscf@inHg)
4% Sample
Rate
(dscfm)
Acceptable
Leak Rate?*
Phase II - Metals - Inlet |
21
22
23
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
107
101
104
104
99.6
105
106
105
108
102
104
104
102
87.7
102
105
107
106
23.04
20.34
21.05
42.01
38.08
45.15
38.81
45.41
37.76
24.41
20.00
25.10
47.05
39.70
45.38
2232
20.84
20.60
0.384
0339
0.351
0.350
0.317
0376
0323
0378
0.315
0.407
0333
0.418
0392
0.331
0378
0.372
0.347
0.343
0.006 @ 14
0.015 @ 4
0.008 @ 14
0.003 @ 5
0.011 @ 12
0.006 @ 14
0.006ฎ 7
0.009ฎ 5
0.065 @ 4
0.009 @ 14
0.006ฎ 4
0.011 @ 13
0.011 @ 7
0.010 @ 6
0.007 @ 14
0.009 @ 15
0.012 @ 8
0.014 @ 4
0.015
0.013
0.014
0.014
0.013
0.015
0.013
0.015
0.013
0.016
0.013
0.017
0.016
0.013
0.015
0.015
0.014
0.014
V
N
Y
Y
Y
Y
Y
Y
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Phase II - Metals - Outlet
1
2
3
4
5
6
7
8
9
10
11
101
101
102
102
98.6
104
98.7
105
103
102
101
36.12
34.47
35.50
34.80
36.65
39.85
35.63
33.41
38.36
35.09
37.24
0.602
0.575
0.592
0.580
0.611
0.664
0.594
0.557
0.639
0.585
0.621
0.009ฎ 7
0.010 @ 8
0.007 @ 10
0.012 @ 15
0.005 @ 12
0.010 @ 14
0.012 @ 8
0.012 @ 10
0.015 @ 10
0.016 @ 6
0.012 @ 7
0.024
0.023
0.024
0.023
0.024
0.027 -
0.024
0.022
0.026
0.023
0.025
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y '
-------
TABLE 7-3. CDD/CDF STACK SAMPLING QUALITY CONTROL DATA
CAMDEN COUNTY MWC - PHASE II (1992)
Rim Number
Isokbetic
(%)
Standard
Meter
Volume
(dsd)
Phase II - CDD/CDF - Inlet
25
26
27
28
29
30
34
35
36
103
99.5
101
99.2
100
104
101
103
96.3
42.01
37.75
43.66
35.88
44.01
33.89
42.79
35.10
42.12
Average
Sample Rate
(dscfin)
Maximum Leak
Check
(dscf@mHg)
4% Sample
Rate
(dscfin)
Acceptable
Leak Rate?*
0.350
0.315
0.364
0.299
0.367
0.282
0.357
0.293
0.351
0.018 @ 9
0.017 @ 15
0.014 @ 7.5
0.012 @ 8
0.014 @ 9
0.030 @ 6.5
0.012 @ 10
0.006ฎ 7
0.014 @ 10
0.014
0.013
0.015
0.012
0.015
0.011
0.014
0.012
0.014
N
N
Y
Y
Y
N
Y
Y
Y
Phase II - CDD/CDF - Outlet
25
26
27
28
29
30
34
35
36
105
103
108
102
103
107
103
103
98.4
72.58
78.78
83.68
81.09
78.28
78.67
78.94
76.63
72.85
0.605
0.657
0.697
0.676
0.652
0.656
0.658
0.639
0.607
0.014 @ 10
0.018 @ 8
0.018 @ 10
0.012 @ 13
0.010 @ 10
0.014 @ 10
0.012 @ 9
0.008 @ 10
0.016 @ 8
0.024
0.026
0.028
0.027
0.026
0.026
0.026
0.026
0.024
Y
Y
Y
Y
Y
Y
Y
Y
Y
"The values shown in the table for 4% of the sample rate were compared to a value of 0.02 dscfm. The maximum allowable leak rate was established
as the lesser of two values.
-------
As mentioned in Section 4.1.1, the moisture content for the multi-metals method
trains during Phase II Outlet Runs 1 and 3 and Phase I Inlet Run 6 appeared erroneous.
Revised values were estimated by subtracting 2.0% from the outlet or adding 2.0% to the
inlet flue gas moisture content for these runs. The 2.0% adjustment was selected based
on the average difference between the inlet and outlet moisture contents during the
other runs. Also, during Phase II Inlet Run 25, the silica gel impinger broke following
successful final leak check of the train, and the silica absorbed water from the impinger
bucket. To estimate the actual moisture level, a silica gel weight gain of 8.3 g was used
based on the average weight gain during other inlet runs.
7.2.2 Method 23 Flue Gas Sampling Quality Control
Stack sampling QC data, including isokinetic sampling rates, sample volume
collected, maximum recorded leak rate, and maximum allowable leak rate, are
summarized in Table 7-3 for each Method 23 run. All of the data quality indicators are
within acceptable limits, with the exception of high leak rates for three runs.
As with the EPA multi-metals method, the acceptance criteria for sampling train
leak checks is a leak rate of less than 4% of the average sampling rate or 0.02 dscf,
whichever is less. All of the outlet trains and six of nine inlet trains met this criteria.
Two of the trains (Inlet Runs 25 and 26) met the 0.02 dscf criteria, but had leaks of 5%
of the sampling rate. The third train (Inlet Run 30) had a 10% leak rate. The final
sample volume for these test runs were not corrected for the high leak rates. If
corrections had been made, flue gas flow rates would be 1 to 3% lower than shown.
7.2.3 Volatile Organic Flue Gas Sampling Quality Control
Stack sampling QC data, including average standard meter volume and maximum
recorded leak rate, are summarized in Table 7-4 for each VOST run. All of the data
quality indicators are within acceptable limits for all runs.
7-9
-------
7.2.4 Particle Size Distribution (PSD)
There were no problems observed during the PSD sampling, except for the two
PSD trains operated during Condition A4. One train had a loose impinger connection
which was discovered at the end of the run, and the other train had sampling pump
problems. Post-test review of the collected data from the first train indicated that the
flue gas moisture content was lower than for other trains and that the isokinetic flow rate
was high. As a result, the samples collected by the first train were invalidated. Post-test
review of data from the second train indicated that all QA/QC criteria were met.
Therefore, the data from this train are acceptable.
7.3 Sample Storage and Holding Time
Sample hold times specified in the QAPP were met for all samples, with the
exception of the CDD/CDF samples and 33 front-half fractions (acetone and nitric
probe rinses, and filter) for the analysis of Hg. The QAPP called for a maximum
CDD/CDF sample hold time of 21 days. All of these samples were analyzed 28 to
30 days after completion of the tests. Although the CDD/CDF hold times exceeded the
QAPP objectives, they were within the 30-day hold time limit in EPA Method 23.
Therefore, the data are acceptable.
The hold time for Hg, as specified by SW-846 Method 7470, is 38 days. The hold
times were missed by 1 to 10 days for Phase II Inlet Runs 13 through 15, 22, 23, and the
Field Blank and Outlet Runs 13 through 39. The hold times were missed because the
filters and acetone probe rinses were weighed for paniculate matter at Radian's
Morrisville, North Carolina laboratory, before the samples were sent to Radian's
Austin, Texas laboratory for sample digestion and analysis. Any potential loss of Hg in
the front-half fractions due to extended hold time would lower the reported Hg
concentrations. For these runs, the Hg concentration was not noticeably lower than for
other runs at similar operating conditions that met the hold time limits. As a result, the
values detected are considered acceptable for calculating removal efficiencies.
7-12
-------
TABLE 7-4. VOST STACK SAMPLING QUALITY CONTROL DATA
CAMDEN COUNTY MWC - PHASE II (1992)
Run Number
Average
Standard Meter
Volume
(dscf)
Maximum Leak
Check
(dscf @ in Hg)
Acceptable
Leak Rate?
Phase II - VOST - Inlet
25
26
27
28
29
30
0.718
0.670
0.665
0.681
0.671
0.664
0@20
0@ 17
0@ 19
0@ 18
0@ 18
0@ 18
Y
Y
Y
Y
Y
Y
Phase II - VOST - Outlet
25
26
27
28
29
30
0.670
0.669
0.670
0.676
0.673
0.673
0 @ 22.5
0@21
0@22
0@21
0@21
0@21
Y
Y
Y
Y
Y
Y
7-11
-------
Analytical QC criteria for the metals train analyses were:
70 to 130% recovery for laboratory control samples and matrix spike
samples; and
<20% relative percent difference (RPD) for duplicates or <20% relative
standard deviation (RSD) for replicates.
Recoveries and RPDs for matrix spike/matrix spike duplicates for Hg analyzed
for Phase I and II are presented in Table 7-5. Recoveries and RPDs for laboratory
control samples and analytical spikes for the other metals analyzed for Phase I are
presented in Table 7-6. Phase II recoveries and RPDs are presented in Table 7-7.
Verification of system accuracy was provided by the performance evaluation audit
of blind metals samples provided by RTI. Measured and audit values for the two blind
samples are provided in Table 7-8. All of the sample recoveries were between 90 and
104%, well within the QC criteria of 70 to 130%.
Mercury Analytical Quality Control
Of these analyses, 40 out of 44 (91%) met the accuracy QC criteria. All of the
samples met the precision QC criteria. The matrix spike recoveries for the Outlet
Run 27 sample were 131 and 135%, and recoveries for the Outlet Run 33 sample were
135 and 138%. The high recovery for these samples indicates either a potential spiking
problem or a matrix interference. Because the laboratory control samples, the
calibration quality control samples, and other front-half fraction recoveries met the QC
criteria, a matrix interference is not likely. As shown in Table 7-5, analysis of 3 other
front-half fractions showed 90 to 114% Hg recoveries. Since a spiking problem is
indicated for the four matrix spikes and other QC analyses met the data quality
objectives, the Hg analytical results are judged to be acceptable.
7-14
-------
7.4 Analytical Quality Control
Analytical methods used during the carbon injection testing included the
following:
Metals by SW-846 Method 7470 for Hg, Method 7060 for As, Method 7131
for Cd, Method 7421 for Pb, Method 7740 for Se, and Method 6010 for
other metals by ICAP;
Chlorinated CDD/CDF by EPA Method 23;
Volatile organics by SW-846 Method 8240; and
Gravimetric analysis for Method 5 and particle size distribution.
Results for matrix spikes, method spikes, control samples, field blanks, and audit samples
are summarized in this section. These samples served the dual purpose of controlling
and assessing measurement data quality, and provided the basis for precision and
accuracy estimates.
No significant blank contamination problems were identified during the analysis of
field and laboratory blanks and no blank corrections were performed for reported
emissions data.
7.4.1 Multiple Metals Analytical Quality Control
Quality control associated with the determination of metals in stack gas samples
included the analysis of laboratory control samples, matrix spike/matrix spike duplicates,
analytical spikes, and audit samples.
7-13
-------
TABLE 7-5, CONTINUED
Mercury Recovery (%)
Matrix
Spike
Matrix
Spike
Duplicate
Relative
Difference
(%)
Permanganate Fraction - Matrix Spike/Matrix Spike Duplicate Results
Phase I - Permanganate Fraction (Outlet Run 7)
Phase I - Permanganate Fraction (Outlet Run 15)
Phase II - Permanganate Fraction (Inlet Run 7)
Phase II - Permanganate Fraction (Outlet Run 9)
Phase II - Permanganate Fraction (Outlet Run 14)
Phase II - Permanganate Fraction (Outlet Run 21)
Phase II - Permanganate Fraction (Inlet Run 26)
Phase II - Permanganate Fraction (Inlet Run 32)
Phase II - Permanganate Fraction (Outlet Run 32)
98
93
84
98
104
109
126
114
126
87
96
84
102
104
110
122
112
124
11.9
3.2
0
4.0
0
0.9
3.2
1.8
1.6
-------
TABLE 7-5. MATRIX SPIKE RESULTS FOR MERCURY IN FLUE GAS
CAMDEN COUNTY MWC - PHASE I AND II (1992)
Mercury Recovery (%)
Matrix
Spike
Matrix
Spike
Duplicate
Relative
Difference
(%)
Front Fraction - Matrix Spike/Matrix Spike Duplicate Results
Phase I - Front Fraction (Outlet Run 13)
Phase II - Front Fraction (Outlet Run 27)
Phase II - Front Fraction (Inlet Run 29)
Phase II - Front Fraction (Outlet Run 33)
Phase II - Front Fraction (Inlet Run 39)
114
135
90
138
101
110
131
105
135
98
3.6
3.0
15.4
2.2
3.0
Nitric/Peroxide Fraction - Matrix Spike/Matrix Spike Duplicate Results
Phase I - Nitric Fraction (Outlet Run 10)
Phase I - Nitric Fraction (Outlet Run 15)
Phase II - Nitric Fraction (Outlet Run 10)
Phase II - Nitric Fraction (Inlet Run 12)
Phase II - Nitric Fraction (Outlet Run 15)
Phase II - Nitric Fraction (Inlet Run 19)
Phase II - Nitric Fraction (Outlet Run 23)
Phase II - Nitric Fraction (Outlet Run 38)
78
128
85
105
106
98
103
105
75
125
91
107
106
107
110
104
3.9
2.4
6.8 1
0.9 1
0 1
8.8
6.6
1.0
-------
TABLE 7-6, CONTINUED
Metal Recovery (%)
Lead
Mancume
fcf..LJ. .!.._
MiNyuucouin
Nickel
Selenium
Silver
Thaffium
VUUKDQIII
Laboratory Control Sample Results
Laboratory Control Sample 1
Laboratory Control Duplicate 1
Relative Difference (%)
101
102
1.0
96
96
0
98
99
1.0
Analytical Spike/Analytical Spike Duplicate Results
Front Half, Outlet Run 6
Front Half Duplicate, Outlet Run 6
Relative Difference (%)
Back Half, Outlet 6
Back Half Duplicate, Outlet Run 6
Relative Difference (%)
74
82
10.2
79
72
9.3
NA
NA
102
102
0
NA
NA
~
88
88
0
97
98
1.0
NA
NA
-
85
86
1.2
99
98
1.0
95
95
0
101
99
2.0
97
98
1.0
113
109
3.6
128
138
7.5
NA
NA
--
89
87
2.3
NA
NA
-
0
0
0
NA
NA
-
87
87
0
-J
e*
oo
*NA = Not analyzed, this analytical spike was only analyzed by GFAAS, not ICAP.
-------
TABLE 7-6. MATRIX SPIKE RESULTS FOR METALS IN FLUE GAS, PHASE I
CAMDEN COUNTY MWC (1992)
Metal Recovery (%)
Antimony
Anenk
Barium
BcryUhnn
Boron
Cadmium
Chromhmi
Cobah
Copper
Laboratory Control Sample Results
Laboratory Control Sample 1
Laboratory Control Duplicate 1
Relative Difference (%)
99
100
1.0
101
%
5.1
97
97
0
98
99
1.0
115
111
35
94
95
1.0
97
98
1.0
97
%
1.0
%
%
0
Analytical Spike/Analytical Spike Duplicate Results
Front Half, Outlet Run 6
Front Half Duplicate, Outlet Run 6
Relative Difference (%)
Back Half, Outlet 6
Back Half Duplicate, Outlet Run 6
Relative Difference (%)
NA'
NA
13
16
20.6
108
107
0.9
109
113
3.6
NA
NA
--
91
90
1.1
NA
NA
-
98
97
1.0
NA
NA
-
112
114
1.8
94
97
3.1
82
82
0
NA
NA
--
86
86
0
NA
NA
--
84
85
1.2
NA
NA
--
87
86
1.2
-------
TABLE 7-7, CONTINUED
Metal Recovery (%)
Antimony
Anenk
Barium
BCffyllllUB
Boron
Cadmium
Chromium
Cobalt
Copper
Analytical Spike/Analytical Spike Duplicate Results
Front Half, Outlet 33
Front Half Duplicate, Outlet Run 33
Relative Difference (%)
Front Half, Inlet 39
Front Half Duplicate, Inlet Run 39
Relative Difference (%)
Front Half, Outlet 39
||
H Front Half Duplicate, Outlet Run 39
II
Relative Difference (%)
Back Half, Outlet Run 9
Back Half Duplicate, Outlet Run 9
Relative Difference (%)
Back Half, Outlet 23
Back Half Duplicate, Outlet Run 23
Relative Difference (%)
Back Half, Inlet 31
Back Half Duplicate, Inlet Run 31
Relative Difference (%)
Back Half, Outlet 38
Back Half Duplicate, Outlet Run 38
1 Relative Difference (%)
77
83
7.5
95
91
4.3
68
67
1.5
80
84
4.9
89
94
5.5
84
90
6.9
88
89
1.1
126
129
2.4
84
86
2.4
91
99
8.4
84
83
1.2
92
90
2.2
98
98
0
88
91
3.4
99
100
1.0
98
93
5.2
98
97
1.0
91
92
1-1.
99
100
1.0
95
98
3.1
98
97
1.0
94
94
0
91
89
2.2
92
92
0
87
88
1.1
96
97
1.0
89
91
2.2
90
90
0
NA
NA
-
NA
NA
-
NA
NA
--
112
111
0.9
118
119
0.8
88
92
4.4
114
113
0.9
100
84
17.4
96
92
4.3
87
87
0
83
84
1.2
102
101
1.0
108
105
2.8
96
100
4.1
96
97
1.0
96
94
2.1
96
96
0.0
92
90
2.2
100
101
1.0
93
94
1.1
96
95
1.0
93
93
0
96
93
3.2
94
93
1.1
87
88
1.1
101
102
1.0
93
93
0
94
93
1.1
97
97
0
97
91
6.4
95
94
1.1
88
89
1.1
98
98
0
92
94
2.2
92
92
0
-------
TABLE 7-7. MATRIX SPIKE RESULTS FOR METALS IN FLUE GAS, PHASE II
CAMDEN COUNTY MWC (1992)
MeUl Recovery (%)
Antimony
Arsenic
Barium
Bayffiam
Boron
Cadmium
Chromium
Cobak
Copper
Laboratory Control Sample Results
Laboratory Control Sample 1
Laboratory Control Duplicate 1
Laboratory Control Sample 2
Laboratory Control Duplicate 2
Relative Standard Deviation (%)
99
100
98
96
L7
100
101
99
101
1.0
97
97
99
99
1.2
98
99
98
,98
0.5
115
111
103
103
5.6
116
116
109
103
5.7
97
98
100
100
1.5
97
%
100
100
2.1
%
%
98
98
1.2
Analytical Spike/Analytical Spike Duplicate Results
Front Half, Outlet Run 14
Front Half Duplicate, Outlet Run 14
Relative Difference (%)
Front Half, Outlet 27
Front Half Duplicate, Outlet Run 27
Relative Difference (%)
Front Half, Inlet 29
Front Half Duplicate, Inlet Run 29
Relative Difference (%)
78
70
10.8
97
92
5.2
106
110
3.7
108
104
3.8
141
118
17.8
80
76
5.1
101
98
3.0
96
95
1.0
97
99
2.0
95
93
2.1
90
90
0.0
89
89
0.0
NAB
NA
--
NA
NA
--
NA
NA
--
99
%
3.1
104
90
14.4
97
98
1.0
101
97
4.0
93
92
1.1
94
96
2.1
96
95
1.0
89
90
1.1
92
93
1.1
97
94
3.1
93
92
1.1
100
103
3.0
aNot analyzed. Boric acid is added to the front fraction during sample preparation, therefore, invalidating the analysis for boron.
-------
TABLE 7-7, CONTINUED
Metal Recovery (%)
Lead
Motybdenam
1 Nickel
Selenium
Silver
Thaffium
VanMfium
Analytical Spike/Analytical Spike Duplicate Results
Front Half, Outlet 33
Front Half Duplicate, Outlet Run 33
Relative Difference (%)
Front Half, Inlet 39
Front Half Duplicate, Inlet Run 39
Relative Difference (%)
Front Half, Outlet 39
Front Half Duplicate, Outlet Run 39
Relative Difference (%)
Back Half, Outlet Run 9
Back Half Duplicate, Outlet Run 9
Relative Difference (%)
Back Half, Outlet 23
Back Half Duplicate, Outlet Run 23
Relative Difference (%)
Back Half, Inlet 31
Back Half Duplicate, Inlet Run 31
Relative Difference (%)
Back Half, Outlet 38
Back Half Duplicate, Outlet Run 38
Relative Difference (%)
97
93
42
111
106
4.6
100
100
0
87
88
1.1
101
104
2.9
98
94
4.2
99
97
2.0
94
94
0
97
94
3.1
94
93
1.1
88
88
0
104
108
3.8
91
92
1.1
92
91
1.1
95
94
1.1
96
93
3.2
96
94
2.1
87
87
0
98
98
0
92
93
1.1
93
92
1.1
93
94
1.1
94
92
2.2
90
91
1.1
89
89
0
98
96
2.1
92
94
2.2
95
95
0
94
93
1.1
5.6
0
200
73
84
14.0
99
87
12.9
83
78
6.2
0
0
--
62
66
6.3
83
85
2.4
87
84
3.5
84
83
1.2
87
87
0
85
95
11.1
91
93
2.2
91
90
1.1
102
93
9.2
84
86
2.4
93
95
2.1
88
88
0
100
98
2.0
92
90
2.2
92
92
0
98
97
1.0
95
93
2.1
96
95
1.0
91
91
0
100
100
0
92
94
2.2
93
92
1.1
-------
TABLE 7-7, CONTINUED
MeUl Recovery (%)
Lead
Manganese
Molybdenom
Nickel
Selenium
Silver
HuOram
Vanadium
Laboratory Control Sample Results
Laboratory Control Sample 1
Laboratory Control Duplicate 1
Laboratory Control Sample 2
Laboratory Control Duplicate 2
Relative Standard Deviation (%)
94
96
100
102
3.7
96
96
99
99
1.8
98
99
99
98
0.6
97
98
97
99
1.0
91
91
88
89
1.7
95
95
100
101
3.3
101
99
97
98
1.7
97
98
100
100
15
Analytical Spike/Analytical Spike Duplicate Results
Front Half, Outlet Run 14
Front Half Duplicate, Outlet Run 14
Relative Difference (%)
Front Half, Outlet 27
Front Half Duplicate, Outlet 27
Relative Difference (%)
Front Half, Outlet 29
Front Half Duplicate, Outlet Run 29
Relative Difference (%)
66
66
0
90
94
4.3
132
146
10.1
97
94
3.1
91
90
1.1
98
102
4.0
99
96
3.1
90
91
1.1
92
93
1.1
97
93
4.2
90
90
0
89
90
1.1
69
66
4.4
64
66
3.1
111
132
173
87
86
1.2
77
77
0
91
92
1.1
108
99
8.7
98
90
8.5
82
88
7.1
99
97
2.0
94
94
0
95
96
1.0
-------
Cadmium Analytical Quality Control
All analyses met the accuracy and precision QC criteria.
Lead Analytical Quality Control
All analyses for Phase I met the accuracy and precision QC criteria. For Phase II,
analytical spikes for Pb had two front-half fraction spikes that were outside control limits,
with Outlet Run 14 at 66% and Outlet Run 29 at 132% and 146% recovery. Four other
front fraction analytical spikes had 90 to 106% recovery. The laboratory control samples
analyzed were 96 to 100% recovery, which is well within the QC objective of 80 to 120%
recovery for these analyses. The poor Pb recoveries for the three analytical spikes
suggest a potential spiking problem or a matrix interference. Since the laboratory
control samples, the calibration quality control samples, and other front fraction
recoveries are in control, the Pb analytical results are acceptable.
Other Metals Analytical Quality Control
For the other metal samples analyzed, 288 out of 298 (96%) met the accuracy QC
criteria for Phase I and n. All analyses, except for some analytical spikes for Sb, As, and
Tl, met the accuracy and precision QC criteria. The spike recoveries for Sb in the
back-half fraction, Phase I Outlet Run 6, were 13 and 16% recovery. As indicated by the
spike recovery results, a low bias is likely for the Sb back fraction results. The
laboratory control samples analyzed show acceptable recoveries for all of the elements.
The exceedances for As and Tl are expected to have limited impact on data quality.
Selenium recoveries met the QC criteria for only 10 of the 24 (41%) samples
analyzed. Because most of the samples did not meet the data quality objective, the
values determined for Se were viewed as questionable and are not reported.
7-24
-------
TABLE 7-8. AUDIT SAMPLE RESULTS FOR METAL ANALYSIS
CAMDEN COUNTY MWC PHASE H (1992)
Sample ID
3967-56H-03
3967-56H-04
Analyte
Cadmium
Lead
Mercury
Cadmium
Lead
Mercury
SW-846
Method
7131
7421
7470
7131
7421
7470
Measured
Concentration
Gxg/D
34
380
1.47
62
400
4.16
Audit
Concentration
G*g/D
36
420
1.6
60
420
4.0
Recovery
(%)
94.4
90.4
91.9
103.3
95.2
104.0
7-23
-------
TABLE 7-9. SURROGATE RECOVERY RESULTS FOR CDD/CDF
CAMDEN COUNTY MWC - PHASE II (1992)
Run
% Recovery
TCDD
Inlet
Run 25
Run 26
Run 27
Run 28
Run 29
Run 30
Run 34
Run 35
Run 36
Field Blank
Relative Standard Deviation
74
83
91
81
77
82
86
84
76
78
6.3
PeCDF
93
101
107
89
120
105
91
94
88
96
10.1
HxCDF 478
HxCDD 478
HpCDF 789
90
99
92
89
94
95
97
102
83
87
6.2
80
87
81
91
85
88
98
96
.72
83
9.0
Outlet
Run 25
Run 26
Run 27
Run 28
Run 29
Run 30
Run 34
Run 35
Run 36
Field Blank
Relative Standard Deviation
83
83
94
84
86
87
87
64
82
43
18.7
128
108
110
76
86
95
98
86
108
41
42.2
88
93
99
95
90
94
88
74
86
33
22.8
103
84
89
100
101
87
98
80
80
29
25.3
100
105
98
108
108
98
114
108
96
97
6.0
104
99
104
102
112
108
107
84
105
30
25.3
7-26
-------
7.4.2 CDD/CDF Analytical Quality Control
Quality control associated with the determination of CDD/CDFs in stack gas
samples included method spikes and audit samples. Additionally, all samples were
spiked with isotopically labeled surrogates. The CDD/CDF stack gas analytical data are
of acceptable quality. Analytical QC criteria for the CDD/CDF train analyses were:
50 to 150% recovery for surrogates and method spikes; and
<40% RSD for replicates.
Surrogate recoveries for the stack gas CDD/CDF analyses are summarized in
Table 7-9. The accuracy QC objective was met for all samples except for the Outlet
field blank. This sample was low for all compounds, ranging from 29 to 43%. Since this
sample was a field blank, the low recoveries have no impact on the quality of the
CDD/CDF emissions data. However, results reported for the field blank may have a
slightly low bias.
Method spike recoveries for the stack gas CDD/CDF analyses are summarized in
Table 7-10. The data quality objective for accuracy and precision was met for all
compounds except for Spike 709A for 2,3,7,8-TCDD and 2,3,7,8-TCDF. The recovery for
2,3,7,8-TCDD was 155% and 2,3,7,8-TCDF was 260%. Because this is a method spike,
there is not a possibility of the matrix interference causing high recoveries. The recovery
for 2,3,7,8-TCDD and 2,3,7,8-TCDF in Spike 707 and 708B met the data quality
objectives and, therefore, it appears that the high values are caused by a spiking error.
7-25
-------
Results from the analysis of method blanks and field blanks are summarized in
Table 7-11. No blank corrections were made in calculating stack emission rates. Any
background contamination in the samples or analytical system would, therefore, tend to
cause a high bias in the reported emission rates.
Verification of system accuracy was provided by the analysis of blind CDD/CDF
audit samples provided by RTI. Measured values for- the audit samples are provided in
Table 7-12. All isomers were within acceptable limits except for hexa-CDF, for which
measured values in both samples were slightly high.
7.4.3 Stack Gas Volatile Organic Compound Quality Control
Quality control associated with the determination of VOC in stack gas samples
included analysis of method spikes and audit samples. In addition, all samples were
spiked with isotopically labeled surrogates. Analytical QC criteria for the VOST analyses
were:
50 to 150% recovery for surrogates and method spikes;
<40% RSD for replicates.
The analysis of one VOST tube for Outlet Run 26 was lost due to an instrument
failure. However, because four sets of tubes were collected during each run, three valid
sample sets were still obtained, and the 100% completeness objective was met for this
run.
The results of the field blank results for each test are provided in Tables 7-13 and
7-14. No blank corrections were made in calculating stack emission rates. Any
background contamination in the samples would, therefore, tend to cause a high bias in
the reported emission rates.
7-28
-------
TABLE 7-10. CDD/CDF METHOD SPIKE RESULTS
CAMDEN COUNTY MWC PHASE H (1992)
Isomers
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OctaCDD
2,3,7,8- TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Octa CDF
Spike Recovery (%)
Spike
707
120
110
94
88
77
98
105
130
110
120
100
99
73
88
100
120
100
Spike
709A
155
100
98
100
86
100
100
260
iio
140
110
97
94
92
100
110
105
Spike
708B
110
88
93
86
74
90
95
115
95
77
96
91
68
83
90
110
100
RSD*
(%)
18
11
2.8
8.3
7.9
5.5
5.0
55
8.2
29
7.1
4.4
18
5.1
6.0
5.1
2.8
"RSD = (standard deviation/mean) x 100
7-27
-------
TABLE 7-12. CDD/CDF AUDIT RESULTS
CAMDEN COUNTY MWC (1992)
Isomer
2378 TCDD
Total TCDD
12378 PeCDD
Total PeCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
Total HxCDD
1234678 HpCDD
Total HpCDD
Total OCDD
2378 TCDF
Total TCDF
12378 PeCDF
23478 PeCDF
Sample 3967-56H-01
Measured
Cone.
(mg/L)
0.390
0.390
0.460
0.460
0.450
0.0062
0.0018
0.460
0.450
0.450
0.300
0.580
0.580
0.560
0.0035
Audit
Cone.
(mg/L)
0.375
0.375
0.375
0.375
0.375
0.0
0.0
0.375
0.375
0.375
0.375
0.500
0.500
0.500
0.0
Recovery
(%)
104
104
123
123
120
123
120
120
80
116
116
112
...
Sample 3967-56H-02
Measured
Cone.
(mg/L)
0.510
1.100
0.0033
0.760
0.970
ND
ND
0.970
0.0077
0.0077
0.0052
0.400
0.400
0.390
0.0020
Audit
Cone.
(mg/L)
0.500
1.250
0.0
0.750
0.750
0.0
0.0
0.750
0.0
0.0
0.0
0.375
0.375
0.375
0.0
Recovery
(%)
102
88
101
129
129
107
107
104
-------
TABLE 7-11. CDD/CDF FLUE GAS BLANK RESULTS
CAMDEN COUNTY MWC PHASE H (1992)
Isomer
2,3,7,8-TCDF
Total TCDF
23,7,8-TCDD
Total TCDD
1,23,7,8-PeCDF
23,4/7,8-PeCDF
Total PeCDF
1,23,7,8-PeCDD
Total PeCDD
1,23,4,7,8-HxCDF
1,23,6,7,8-HxCDF
1,23,7,8,9-HxCDF
23,4,6,7,8-HxCDF
Total HxCDF
1,23,4,7,8-HxCDD
1,23,6,7,8-HxCDD
1,23,7,8,9-HxCDD
Total HxCDD
1,23,4,6,7,8-HpCDF
U3,4,7,8,9-HpCDF
Total HpCDF
1,23,4,6,7,8-HpCDD
Total HpCDD
OCDF
OCDD
Method Blank Results
(ซg)
Blank 707
[0.0210]
NDa
[0.0270]
ND
[0.0065]
[0.0056]
ND
[0.0051]
ND
[0.0030]
[0.0018]
[0.0020]
[0.0018]
ND
[0.0055]
[0.0052]
[0.0033]
ND
[0.0039]
[0.0027]
ND
0.0054
0.0054
[0.0100]
0.0880
Blank 709
[0.0180]
ND
[0.0260]
ND
[0.0045]
[0.0078]
ND
[0.0047]
ND
0.0058
[0.0023]
[0.0030]
[0.0033]
ND
[0.0042]
[0.0057]
[0.0074]
ND
[0.0099]
[0.0100]
ND
0.0110
0.0220
[0.0100]
0.1500
Field Blank Results
(ng)
Inlet
0.040
0.500
[0.028]
ND
0.046
0.050
0.430
0.014
0.086
0.050
0.050
0.046
0.018
0390
0.014
0.018
0.012
0.150
0.180
0.047
0.290
0.160
0290
0.180
0360
Outlet
[0.0180]
0.0110
[0.0160]
ND
0.0067
[0.0130]
0.0067
[0.0067]
ND
0.0072
0.0059
0.0051
[0.0026]
0.0350
0.0041
[0.0075]
0.0028
0.0069
0.0280
0.0092
0.0450
0.0230
0.0230
0.0510
0.1700
*ND = Not Detected
[ ] = less than 5 times the method detection limit.
7-29
-------
TABLE 7-13. VOST FIELD BLANK RESULTS - CONDITION BIO
CAMDEN COUNTY MWC - PHASE II (1992)
Compound
Trichlorofluoromethane
Carbon Disulfide
Methylene Chloride
Benzene
Toluene
Chlorobenzene
ir.,p-Xylene
Compound
Trichlorofluoromethane
Carbon Disulfide
Methylene Chloride
Benzene
Toluene
Chlorobenzene
m,p-Xylene
INLET (total ng)
Run 25
Tube
A
120
35
ND
240
10
26
12
Tube
B
140
130
12
290
15
41
22
Tube
C
ND"
40
18
91
ND
14
17
Tube
D
ND
33
10
59
ND
ND
31
Blank
Tube
ND
ND
ND
48
ND
ND
44
Run 26
Tube
A
34
88
27
35
ND
ND
23
Tube
B
ND
78
ND
24
ND
ND
17
Tube
C
18
46
ND
24
ND
ND
15
Tube
D
34
150
14
360
ND
25
58
Blank
Tube
ND
ND
ND
12
ND
ND
ND
Run 27
Tube
A
ND
220
19
120
ND
15
40
Tube
B
ND
62
10
81
ND
17
18
Tube
C
ND
32
72
73
ND
ND
60
Tube
D
ND
29
12
59
ND
14
ND
Blank
Tube
ND
ND
ND
ND
ND
ND
26
OUTLET (total ng)
Run 25
Tube
A
280
48
26
120
20
ND
29
Tube
B
210
28
38
100
19
ND
18
Tube
C
ND
16
28
65
26
ND
28
Tube
D
ND
11
39
48
13
ND
32
Blank
Tube
ND
ND
11
14
ND
ND
ND
Run 26
Tube
A
ND
ND
ND
ND
ND
ND
ND
Tube
B
ND
ND
13
32
15
ND
15
Tube
C
65
21
24
35
20
ND
29
Tube
D
37
29
40
50
25
ND
27
Blank
Tube
ND
ND
ND
ND
ND
ND
ND
Run 27
Tube
A
ND
12
19
31
27
ND
21
Tube
B
ND
29
27
41
28
ND
35
Tube
C
18
ND
24
36
26
ND
28
Tube
D
11
20
31
36
27
ND
28
Blank
Tube
ND
ND
ND
ND
ND
ND
ND
'ND = Not Detected.
Note: Final values were not field blank corrected.
-------
TABLE 7-12. CONTINUED
Isomer
Total PeCDF
123478 HxCDF
123678 HxCDF
123789 HxCDF
234678 HxCDF
Total HxCDF
1234678 HpCDF
1234789 HpCDF
Total HpCDF
Total OCDF
" Outside acceptable
Sample 3967-56H-01
Measured
Cone.
(mg/L)
0.560
0.710
ND
ND
ND
0.710
0.420
0.0009
0.420
0.560
Audit
Cone.
(mg/L)
0.500
0.500
0.0
0.0
0.0
0.500
0.500
0.0
0.500
0.500
Recovery
(%)
112
142a
142a
84
84
112
Sample 3967-56H-02
Measured
Cone.
(mg/L)
0.390
0.490
ND
ND
ND
0.490
0.310
ND
0.310
0.380
Audit
Cone.
(mg/L)
0.375
0.375
0.0
0.0
0.0
0.375
0.375
0.0
0.375
0.375
Recovery
(%)
104
131"
131"
83
83
101
units of recovery for HxCDF of 40-130%.
-------
Surrogate recoveries for the stack gas volatile organic analyses are summarized in
Tables 7-15 and 7-16. The data quality objectives for the surrogate recoveries were met
for all analyses, with the exception of high surrogate recoveries for l,2-Dichloroethane-d4
and 4-Bromofluorobenzene in the Inlet Run 27C tube. The high recoveries suggest a
potential high bias in results reported for this one pair of tubes. This potential high bias
for this one pair of tubes has no impact on the overall data quality for the VOST results.
Analytical method spike results for the stack gas volatile organic analyses are
summarized in Table 7-17. The data quality objectives were met for all of the analyses.
Verification of the accuracy of the VOST system was provided through sampling
and analysis of two EPA cylinder audits. Triplicate pairs of VOST tube samples were
collected for each cylinder. The results of the audit samples are summarized in
Table 7-18. Recoveries for all measured compounds were within the project data quality
objectives of 50 to 150%, except for vinyl chloride. The coefficient of variance (CV) for
individual runs of each audit gas by each train ranged from 0 to 24.9%, well within the
acceptable CV of 40%.
Although the recovery and CV values for all of the compounds except vinyl
chloride are within the data quality objectives, the reported concentrations for the inlet
sampling train are consistently higher than for the outlet sampling train. This suggests
the possibility of a systematic bias in operation of the two trains. Review of sampling
train and analytical QC data for the audit samples suggests that the low recoveries for
the outlet train samples may be attributable to differences in chromatograph
performance when the samples were run. Specifically, the inlet audit samples were run
as part of Data Package A, while the outlet audit samples were run as part of Data
Package F. As indicated on Table 7-17, method spike recoveries for Data Package F
were lower than for the other packages. Review of the flue gas samples included in each
data package indicates that only the outlet samples for Run 30 may have been influenced
7-34
-------
TABLE 7-14. VOST FIELD BLANK RESULTS - CONDITION Bll
CAMDEN COUNTY MWC - PHASE II (1992)
Compound
Trichlorofluoromethanc
Carbon Bisulfide
Methylene Chloride
Benzene
Toluene
Clilo: obenzene
r.i.p-Xylene
Compound
Trichlorofluoromethane
Carbon Disulfide
Methylene Chloride
Benzene
Toluene
Chlorobenzene
m.p-Xylene
INLET (toUl ng)
Rua 28
Tube
A
28
280
24
65
ND
14
21
Tube
B
11
76
11
60
ND
11
31
Tube
C
ND'
16
ND
24
ND
ND
ND
Tube
D
ND
32
12
36
ND
ND
14
Blank
Tube
ND
ND
ND
ND
ND
ND
44
Run 29
Tube
A
12
140
15
89
ND
ND
51
Tube
B
ND
110
13
86
ND
15
49
Tube
C
ND
98
15
200
ND
26
56
Tube
D
ND
64
ND
250
ND
36
60
Blank
Tube
ND
ND
ND
ND
ND
ND
ND
Run 30
Tube
A
ND
67
ND
150
ND
23
ND
Tube
B
ND
42
ND
140
ND
27
ND
Tube
C
ND
37
ND
65
ND
20
ND
Tube
D
ND
290
10
74
ND
20
ND
Blank
Tube
ND
ND
ND
ND
ND
ND
ND
OUTLET (total ng)
Run 28
Tube
A
54
28
23
61
37
ND
56
Tube
B
38
24
20
45
28
ND
73
Tube
C
17
25
24
41
26
ND
36
Tube
D
11
13
15
32
19
ND
55
Blank
Tube
ND
ND
ND
ND
ND
ND
15
Run 29
Tube
A
41
55
23
50
45
ND
29
Tube
B
17
27
20
21
15
14
25
Tube
C
ND
21
18
21
11
ND
17
Tube
D
13
14
17
23
ND
ND
43
Blank
Tube
ND
ND
ND
ND
ND
ND
ND
Run 30
Tube
A
37
23
31
37
30
ND
23
Tube
B
ND
12
14
20
15
ND
35
Tube
C
ND
24
13
30
15
ND
ND
Tube
D
ND
16
13
29
38
ND
14
Blank
Tube
ND
ND
ND
ND
ND
ND
ND
'ND = Not Detected
Note: Final values were not field blank corrected.
-------
TABLE 7-15, CONTINUED
Tube
i
Surrogate Recovery (%)
l,2-Dichloroethane-d4
Toluene-d8
4-Bromofluorobenzene
Sample Hold
Time
(days)
Inlet, Run 27
27A
27B
27C
27D
Field Blank '
Relative Standard Deviation
125
100
308
102
108
60.3
116
99
107
99
95
8.1
102
107
160
96
100
23.5
8
8
8
8
8
Inlet, Run 28
28A
28B
28C
28D
Field Blank
Relative Standard Deviation
105
108
105
108
108
1.5
103
96
100
99
88
5.9
83
92
93
92
94
4.9
9
9
9
9
9
-------
TABLE 7-15. SURROGATE RECOVERY RESULTS AND HOLD TIMES FOR
INLET VOST
CAMDEN COUNTY MWC - PHASE II (1992)
Tube
Surrogate Recovery (%)
l,2-Dichloroethane-d4
Toluene-d8
4-Bromofluorobenzene
Sample Hold
Time
(days)
Inlet, Run 25
25A
25B
25C
25D
Field Blank
Relative Standard Deviation
105
95
108
103
114
6.6
100
120
91
103
99
10.4
106
113
104
101
104
4.3
8
8
8
8
8
Inlet, Run 26
26A
26B
26C
26D
Field Blank
Relative Standard Deviation
97
100
96
99
93
2.8
101
94
96
101
99
3.2
102
102
132
106
96
13.1
8
8
8
8
8
-------
TABLE 7-16. SURROGATE RECOVERY RESULTS AND HOLD TIMES FOR
OUTLET VOST
CAM DEN COUNTY MWC - PHASE II (1992)
Tube
Surrogate Recovery (%)
l,2-Dichloroethane-d4
Toluene-d8
4-Bromofluorobenzene
Sample Hold
Time
(days)
Outlet, Run 25
25A
25B
25C
25D
Field Blank
Relative Standard Deviation
104
102
111
108
100
4.2
101
117
97
104
99
7.6
113
100
109
104
114
5.5
8
8
8
8
9
Outlet, Run 26
26A
26B
26C
26D
Field Blank
Relative Standard Deviation
NAa
85
99
102
102
8.4
NA
133
110
108
99
12.9
NA
106
87
100
80
12.7
NA
9
9
9
9
a
oo
-------
TABLE 7-15, CONTINUED
Tube
Surrogate Recovery (%)
l,2-Dichloroethane-d4
Toluene-d8
4-Bromofluorobenzene
Sample Hold
Time
(days)
Inlet, Run 29
29A
29B
29C
29D
Field Blank
Relative Standard Deviation
105
113
109
111
110
2.7
93
85
90
89
93
3.7
100
104
106
108
102
3.0
9
9
9
9
9
Inlet, Run 30
30A
30B
30C
30D
Field Blank
Relative Standard Deviation
110
114
90
91
94
11.3
88
90
103
110
108
10.2
100
111
101
95
100
5.8
9
9
11
11
11
-------
TABLE 7-16, CONTINUED
Tube
Surrogate Recovery (%)
l,2-Dichloroethane-d4
Toluene-d8
4-Bromofluorobenzene
Sample Hold
Time
(days)
Outlet, Run 29
29A
29B
29C
29D
Field Blank
Relative Standard Deviation
105
98
100
105
92
5.4
91
102
97
93
107
6.7
102
93
92
106
100
6.1
8
9
9
9
11
Outlet, Run 30
30A
30B
3.0C
30D
Field Blank
Relative Standard Deviation
98
96
97
88
96
4.2
105
112
114
141
108
12.4
103
92
82
84
82
10.2
11
11
11
11
12
"NA = Not analyzed due to instrument malfunction.
-------
TABLE 7-16, CONTINUED
Tube
Surrogate Recovery (%)
l,2-Dichloroethane-d4
Toluene-d8
4-Bromofluorobenzene
Sample Hold
Time
(days)
Outlet, Run 27
27A
27B
27C
27D
Field Blank
Relative Standard Deviation
98
113
112
115
118
6.9
102
94
101
91
95
5.1
84
102
92
102
90
8.4
9
9
9
9
9
Outlet, Run 28
28A
28B
28C
28D
Field Blank
Relative Standard Deviation
114
113
119
109
113
3.1
93
94
92
119
92
12.0
103
104
112
80
100
11.9
8
8
8
8
8
-------
TABLE 7-18. VOLATILE ORGANIC SAMPLING TRAIN AUDIT RESULTS
CAMDEN COUNTY MWC - PHASE II (1992)
Compound
Audit
Concentration
(ppb)
Inlet Train (ppb)
Run 1
Run 2
Run 3
Recovery
(%)'
Outlet Train (ppb)
Run 1
Run 2
Run 3
Recovery
(%)
Cylinder 5 14A
Vinyl Chloride
Acetone
Methylene Chloride
Chloroform
Carbon Tetrachloride
Benzene
Toluene
Tetrachloroethene
14.6
16.8
11.6
1S.6
15.2
8.3
NDC
ND
16
10
IS
ND
11
7.9
ND
0.41
18
10
17
ND
14
7.5
ND
ND
21
15
22
0.29
18
54
~
109
101
115
ซ
94
3.7
14
ND
11
7.9
11
0.84
8.7
5.1
11
ND
12
9.3
12
0.43
9.3
5.9
ND
ND
10
8.8
11
0.27
8.9
34"
~
66
75
73
56
Cylinders MB
Vinyl Chloride
Carbon Disulfide
Methylene Chloride
Chloroform
Carbon Tetrachloride
Benzene
Toluene
1 ,2-Dichloroethane
Tetrachloroethene
19.3
-
34.0
9.7
30.2
10.1
12
0.38
1.2
43
11
44
ND
0.47
13
12
0.35
ND
39
11
40
ND
0.49
12
12
ND
ND
39
11
44
ND
ND
13
62
-
119
113
141
125
7.9
ND
ND
23
7.0
22
0.51
ND
6.4
7.5
ND
ND
21
6.9
22
0.46
ND
6.0
7.5
ND
ND
21
6.8
21
0.56
ND
5.4
40"
-
64
71
72
59
Rermปปrv hnsed on difference between audit concentration and average of measured concentrations.
kRecovery exceeded acceptance limits of 50 to 150%.
CND = Not Detected.
-------
TABLE 747. VOLATILE ORGANIC SAMPLING TRAIN
METHOD SPIKE RECOVERY RESULTS
PRECISION AND ACCURACY VOST ANALYSES
CAMDEN COUNTY MWC - PHASE II (1992)
Compound
Vinyl Chloride
1,1-Dichloroethene
Chloroform
1 ,2-Dichloropropane
Toluene
Ethyl Benzene
% Recovery*
Method
Spike A
111
112
122
112
114
113
Method
Spike B
117
119
118
109
115
112
Method
Spike C
127
127
117
102
103
98
Method
Spike D
127
117
121
111
116
116
Method
Spike E
127
117
121
111
116
116
Method
Spike F
77
88
87
93
107
111
Relative
Standard
Deviation1*
(%)
17.0
11.8
11.8
7.0
4.9
6.0
'Method Spike A - F correspond to Air Toxics, Ltd. data packages 9206080 A-F.
b RSD = Standard Deviation x 100
Mean
-------
TABLE 7-19. CEM DAILY CALIBRATION CHECKS - UNIT A
CAMDEN COUNTY MWC (1992)
Condition
Al
A2
A3
__a
A4
b
__b
b
A5
Date
05/29/92
05/30/92
06/01/92
06/06/92
06/06/92
06/07/92
06/08/92
06/09/92
06/10/92
RSD (%)c
Economizer O2
Zero = 0%
Actual (%)
0.1
0.2
0.5
0.0
0.0
0.0
0.0
0.0
NA
Error
(%)
0.3
0.7
2.1
0.0
0.0
0.0
0.0
0.0
NA
0.8
Span = 20.9%
Actual
(%)
19.8
19.2
17.9
13.7
21.1
21.0
20.9
20.7
NA
Error
(%)
-5.2
-8.1
-14.4
-34.5
1.0
0.3
0.0
-0.8
NA
5.5
Slack 02
Zero = 0%
Actual
(%)
-0.2
-0.2
NA
NA
-0.1
-0.1
-0.1
NA
-0.2
Error
(%)
-0.7
-0.7
NA
NA
-0.4
-0.4
-0.4
NA
-0.7
0.1
Span = 20.9%
Actual
(%)
22.0
22.1
NA
NA
21.6
22.0
21.9
21.6
NA
Error
(%)
5.2
5.4
NA
NA
3.3
5.2
4.8
3.3
NA
1.0
"Results of economizer CEM calibration conducted at 04:30. Recalibration conducted at 14:30. Emissions testing began at 15:57.
bCalibration results for these days included in table to show CEM stability. Results not included in RSD calculation except as noted in
footnote c.
CRSD (Relative Standard Deviation)=Standard Deviation/Span. Excludes CEM data from morning of 6/06/92; see footnote a. Includes
data from 6/09/92 when data from 6/10/92 not available.
-------
by the low method spike recoveries for Data Package F. Review of the flue gas data
does not indicate any clear difference between the Run 30 samples and those from other
runs. As a result, the VOST results reported in Sections 3 and 4 are considered valid.
7.5 Continuous Emission Monitors
The plant's CEMs were used to monitor flue gas composition at the economizer
exit and stack. The CEMs on Units A and B were of identical design and are described
in Section 6.7. The QC criteria established in the QAPP for the CEM system were
limited to the measurement of O2 and included an accuracy criterion of 80 to 120% of
the reference method value and a precision criterion of less than 10% deviation between
the measured and calibration gas values during each daily calibration check.
To confirm the accuracy of each of these systems, each of the monitors was
certified in accordance with the QA/QC protocols for CEMS in 40 CFR Part 60,
Appendix F. These tests were conducted on Units A and B in February 1992. The
measured relative accuracy of the Unit A economizer and stack O2 monitors was 91.1%
and 98.2% of the reference method, respectively. For Unit B, the relative accuracy of
the economizer and stack monitors was 95.4% and 96.0% of the reference method.
To confirm measurement precision during the testing period, 2-point (zero and
span) calibration checks were conducted each day. The results of the daily calibration
check on Units A and B are presented in Tables 7-19 and 7-20, respectively. As noted in
Table 7-19, the economizer O2 monitor on Unit A exhibited significant span drift
between May 29 and June 6. This drift was corrected by recalibrating the monitor prior
to the start of testing on June 6. With the exception of the Unit A economizer CEM
calibration check on June 1, each of the daily checks met the QC criteria of less than the
10% deviation.
7-43
-------
During Conditions A4 and A5, the economizer O2 monitor on Unit A indicated
O2 levels that were higher than during other runs and that were similar to the levels
measured by the stack monitor. This suggests higher combustor O2 levels during these
two conditions and less air infiltration across the SD/ESP system. Based on review of
plant process data and discussion with plant personnel, it was concluded that the
economizer CEM was reporting erroneously high O2 levels. As a result, the economizer
O2 measurements for these two test conditions were calculated by subtracting 2.4% frpm
the stack O2 reading for the same run. This adjustment factor was based on the average
difference in O2 between the inlet and outlet sampling locations during the other test
conditions on both Units A and B.
7-46
-------
TABLE 7-20. CEM DAILY CALIBRATION CHECKS - UNIT B
CAMDEN COUNTY MWC (1992)
Condition
Bl
B2
B3
B4
B5
B6
B7
B8
B9
BIO
Bll
B12
B13
Date
05/11/92
05/12/92
05/13/92
05/14/92
05/15/92
06/02/92
06/03/92
06/04/92
06/05/92
06/08/92
06/09/92
06/11/92
06/12/92
RSD (%)'
Economizer O,
Zero = 0%
Actual (%)
NA
NA
NA
0.0
NA
0.0
0.0
0.0
0.0
0.0
NA
0.0
0.0
Error
NA
NA
NA
0.0
NA
0.0
0.0
0.0
0.0
0.0
NA
0.0
0.0
0.0
Span = 20.9%
Actual
NA
NA
NA
20.3
NA
20.1
20.2
20.2
20.0
20.1
NA
20.1
20.2
Error
NA
NA
NA
-2.9
NA
-3.8
-3.3
-33
-4.3
-3.8
NA
-3.8
-3.3
0.4
Stack O2
Zero = 0%
Actual
0.4
0.4
0.4
0.4
0.4
-0.3
-0.3
-0.3
-0.3
-0.3
-0.3
-0.3
-0.3
Error
1.9
1.9
1.9
1.9
1.9
-1.1
-1.1
-1.1
-1.1
-1.1
-1.1
-1.1
-1.1
0.0
Span = 20.9%
Actual
21.6
21.2
21.5
20.6
21.3
20.9
20.9
21.1
20.5
20.9
20.7
20.4
20.9
Error
3.3
1.4
2.9
-1.4
1.9
0.0
0.0
1.0
-1.9
0.0
-1.0
-2.4
0.0
1.7
b.
*RSD (Relative Standard Deviation)=Standard Deviation/Span
-------
8.0 REFERENCES
1. Clean Air Act Amendments of 1990, P.L. 101-549, U.S. Congress,
Washington D.C., November 15, 1990.
2. Nebel, K.L. and D.M. White. "A Summary of Mercury Emissions and Applicable
Control Technologies for Municipal Waste Combustors," Report prepared by
Radian Corporation for the Office of Air Quality Planning and Standards, U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina,
September 1991, Docket No. A-89-09, U.S. Environmental Protection Agency,
Washington D.C.
3. Brown, B. and K.S. Felsvang. "Control of Mercury and Dioxin Emissions from
United States and European Municipal Solid Waste Incinerators by Spray Dryer
Absorption Systems," In Proceedings, 1991 International Conference on Municipal
Waste Combustion, Volume 3, EPA-600/R-92-209c (NTIS PB93-124196),
pp 287-317.
4. Nebel, K.L., et al. "Emission Test Report OMSS Field Test on Carbon Injection
for Mercury Control," EPA-600/R-92-192 (NTIS PB93-105518), Air and Energy
Engineering Research Laboratory, Research Triangle Park, North Carolina,
September 1992.
5. Guest, T.L. and O. Knizek. "Mercury Control at Burnaby's Municipal Waste
Incinerator," 84th Annual Meeting & Exhibition, Air & Waste management
Association, Vancouver, B.C., June 1991.
6. White, D.M., et al. "Parametric Evaluation of Activated Carbon Injection for
Control of Mercury Emissions from a Municipal Waste Combustor," Paper
No. 92-40.06, 1992 Annual Meeting, Air & Waste Management Association,
Kansas City, Missouri, June 1992.
7. Brna, T.G., J.D. Kilgroe, and C.A. Miller. "Reducing Mercury Emission from
Municipal Waste Combustion with Carbon Injection into Flue Gas," ECO
World '92 Conference, Washington, D.C., June 1992.
8. Brna, T.G. "Toxic Metal Emissions from MWCs and their Control," In
Proceedings: 1991 International Conference on Municipal Waste Combustion,
Volume 3, EPA-600/R-92-209c (NTIS PB 93-124196), pp 23-39.
9. Code of Federal Regulations (CFR), Environmental Protection (40), Part 266,
Appendix IX Methods Manual for Compliance With the BIF Regulations,
Section 3.1 Methodology for the Determination of Metals Emissions in Exhaust
Gases from Hazardous Waste Incineration and Similar Combustion Sources, U.S.
Government Printing Office Washington, D.C. 20402, July 1991.
8-1
-------
10. 40 CFR, Part 60 Standards of Performance for New Stationary Sources,
Appendix A Test Methods, Method 5-Determination of Paniculate Emissions
from Stationary Sources, U.S. Government Printing Office Washington, D.C.
20402, July 1991.
11. 40 CFR, Part 60 Standards of Performance for New Stationary Sources,
Appendix A Test Methods, Method 23-Determination of Polychlorinated
Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans from Stationary Sources,
U.S. Government Printing Office Washington, D.C. 20402, July 1991.
12. Test Methods for Evaluating Solid Waste, Physical/Chemical Methods, SW-846,
Manual, Third Addition, Doc. 955-00 1-0000001, Available from Superintendent of
Documents, U.S. Government Printing Office Washington, D.C. 20402,
November 1986.
13. 1984 Annual Book of ASTM Standards, Part 26 Gaseous Fuels; Coal and Coke;
Atmospheric Analysis, D-3178-84 Standard Test Methods for Carbon and
Hydrogen in the Analysis of Sample of Coal and Coke, ASTM,
Philadelphia, Pennsylvania, 1984.
14. 40 CFR, Part 60 Standards of Performance for New Stationary Sources,
Appendix F-Quality Assurance Procedures, U.S. Government Printing Office
Washington, D.C. 20402, July 1991.
8-2
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/R-93-181
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Emission Test Report, Field Test of Carbon Injection
for Mercury Control, Camden County Municipal
Waste Combustor
5. REPORT DATE
September 1993
6. PERFORMING ORGANIZATION CODE
7.AUTHOR(s)D>M. White, W. E. Kelly, M. J. Stucky,
J. L. Swift, andM.A. Palazzolo
8. PERFORMING ORGANIZATION REPORT NO.
DCN: 93-239-022-42-01
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
Radian Corporation
P. O. Box 13000
Research Triangle Park, North Carolina 27709
V CONTRACT/GRANT NO.
68-D9-0054 Task 71. and
68-W9-0069 Task 25
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
13. TYPE Of REPORT AND PERIOD COVERED
Task Final; 1/92 - 4/93
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES AEERL project officer is James D.
541-2854.
Kilgroe, Mail Drop 65, 919 /
16. ABSTRACT,
The report gives results of parametric tests to evaluate the injection of pow-
dered activated carbon to control volatile pollutants in municipal waste combustor
(MWC) flue gas. The tests were conducted at a spray dryer absorber/electrostatic
precipitator (SD/ESP)-equipped MWC in Camden County, New Jersey. Primary test
objectives were to evaluate the effect of carbon type, feed rate, feed method, and
ESP operating temperature on emissions of mercury (Hg) and chlorinated dioxins and
furans (CDD/CDF), and to assess the impact of carbon injection on the p.articulate
matter control performance of the ESP. Secondary objectives were to examine the
impact of carbon injection on emissions of other metals and volatile organic com-
pounds (VOCs). The tests included operating three different carbon injection systems
and examining 16 different SD/ESP and carbon injection system operating conditions.
Test results indicate that carbon injection upstream of a SD/ESP can achieve high
levels (> 90%) of Hg and CDD/CDF reduction. Key system operating parameters are
carbon feed rate, carbon feed method, and ESP temperature. No detrimental im-
pacts on ESP performance were identified. The study also found that carbon injection
does not have a significant impact on emissions of the other metals examined or of
VOCs.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. cos AT I Field/Group
Pollution
Activated Carbon
Mercury (Metal)
Wastes
Combustion
Flue Gases
Halohydrocarbons
Furans
Electrostatic Precipi-
tators
Particles
Organic Compounds
Volatility
Pollution Control
Stationary Sources
Municipal Waste Com-
bustion
Dioxins
Particulate
Volatile Organics
13B
11G
07B
14G
21B
07C
131
07C
20M
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
198
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
8-3
------- |